CN120256976A - A data processing method and system based on liquid leakage detection - Google Patents

Abstract

The present application relates to the field of data processing technology, and discloses a data processing method and system based on liquid leakage detection. The method includes: collecting sensor signals and processing to obtain basic leakage data; analyzing signal changes to determine the type of leakage and the scope of influence; matching risk thresholds to determine risk levels; correlating historical records to analyze leakage patterns; calculating failure probabilities to generate maintenance strategies; and constructing operation status graphics to form hierarchical management reports. The present application transforms leakage management from the traditional passive response mode to a data-driven active prevention mode, while achieving optimal allocation of maintenance resources, reducing maintenance costs while improving leakage prevention and control efficiency.

Description

Data processing method and system based on liquid leakage detection

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a data processing method and system based on liquid leakage detection.

Background

In modern building management, the problem of liquid leakage has been an important challenge for maintenance of the facility. Traditional liquid leakage detection methods mainly rely on periodic inspection and post-hoc response, i.e. inspection of critical parts such as pipes, valves, joints, etc. by manual periodic inspection, or treatment after leakage has occurred and causes significant losses. There are various liquid leak detection devices on the market including touch sensors, photoelectric sensors, acoustic detection systems, etc., which can monitor leakage in real time and issue alarms. Building Management Systems (BMS) also often integrate basic leak alarm functions, triggering a corresponding emergency response procedure when a leak is detected. Some advanced management systems also attempt to utilize statistical methods to simply analyze historical leakage data, identify high risk areas, and assist in maintenance decisions.

However, existing methods of liquid leak detection and management suffer from significant drawbacks. First, most systems only focus on detection and alarm of leakage, lack deep mining and analysis capability of detection data, and cannot extract valuable leakage rules from mass monitoring data. Secondly, the existing risk assessment method is often too simple, and multi-dimensional factors such as leakage type, influence range, region importance and the like cannot be fully considered, so that risk level judgment is inaccurate. Third, current leakage management mostly adopts a passive response mode, lacks predictive maintenance mechanism based on data analysis, and cannot effectively prevent leakage events. Fourth, maintenance resource allocation is typically based on experience or a fixed plan, rather than being optimized according to actual risk levels and cost-effectiveness, resulting in wasted resources. Finally, existing systems generally lack data presentation capabilities oriented to different management levels, and are difficult to support the overall management requirements from the operational layer to the decision layer.

Disclosure of Invention

The application provides a data processing method and a system based on liquid leakage detection, which are used for converting leakage management from a traditional passive response mode to a data-driven active prevention mode, realizing the optimal configuration of maintenance resources, reducing the maintenance cost and improving the leakage prevention and control efficiency.

The application provides a data processing method based on liquid leakage detection, which comprises the steps of collecting monitoring signals of a photoelectric liquid leakage sensor through an RS485 bus, conducting signal denoising and outlier filtering on the monitoring signals to obtain leakage basic data comprising sensor ID, position information, detection time stamp and signal strength, calculating signal strength change rate and duration according to the leakage basic data, combining a building water supply and drainage system topological structure to obtain leakage characteristic data for identifying a leakage type and an influence range, matching a risk threshold value based on the identification leakage type, comparing the leakage characteristic data with the risk threshold value to determine leakage risk level, conducting time, position and type correlation analysis on the leakage risk level and historical leakage records, extracting periodic distribution of leakage occurrence, equipment aging correlation and use strength correlation to form leakage rule data, calculating failure probability of each pipeline section and equipment according to the leakage rule data, combining maintenance cost and resource limiting parameters to generate maintenance strategy data comprising frequency, maintenance period and equipment update, and executing report form maintenance strategy construction of the maintenance strategy based on the operation state control system, integrating the maintenance strategy and executing the maintenance strategy construction of the liquid monitoring system.

In a second aspect, the present application provides a liquid leak detection-based data processing system comprising:

the filtering module is used for collecting monitoring signals of the photoelectric liquid leakage sensor through the RS485 bus, carrying out signal denoising and outlier filtering on the monitoring signals to obtain leakage basic data comprising sensor ID, position information, detection time stamp and signal strength;

The calculation module is used for calculating the signal intensity change rate and the duration according to the leakage basic data and combining the topological structure of the building water supply and drainage system to obtain leakage characteristic data for marking the leakage type and the influence range;

the comparison module is used for determining a leakage risk level by comparing the leakage characteristic data with a risk threshold value based on the identification leakage type matching risk threshold value;

The extraction module is used for carrying out time, position and type association analysis on the leakage risk level and the historical leakage record, extracting periodic distribution, equipment aging association and use intensity association of leakage occurrence, and forming leakage rule data;

The generation module is used for calculating the failure probability of each pipeline section and equipment according to the leakage rule data, and generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion by combining maintenance cost and resource limiting parameters;

And the integration module is used for constructing a liquid system running state diagram based on the maintenance strategy data, integrating the real-time monitoring data and the maintenance execution condition and generating a management-hierarchy leakage prevention and control data report.

A third aspect of the present invention provides a computer device comprising a memory and at least one processor, the memory having instructions stored therein, the at least one processor invoking the instructions in the memory to cause the computer device to perform the above-described liquid leak detection based data processing method.

A fourth aspect of the present invention provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the above-described liquid leak detection-based data processing method.

In the technical scheme provided by the application, the monitoring signal of the photoelectric liquid leakage sensor is acquired through the RS485 bus, and the signal denoising and outlier filtering are carried out to obtain high-quality leakage basic data, so that the problems of unstable signal and easy interference of the traditional leakage detection system are solved, and the accuracy of leakage detection is greatly improved. And secondly, calculating the change rate and duration of the signal intensity according to the leakage basic data, combining the topological structure of the building water supply and drainage system, realizing the accurate identification of the leakage type and the influence range, and overcoming the limitation that the traditional method can not distinguish the sudden leakage and the slow leakage. By matching the risk threshold value based on the identification leakage type, the leakage characteristic data is compared with the risk threshold value, so that the scientific and reasonable leakage risk level is determined, the situation that the traditional leakage risk assessment is too subjective and rough is changed, and the quantification and standardization of the risk assessment are realized. And carrying out time, position and type association analysis on the leakage risk level and the historical leakage record, extracting periodic distribution, equipment aging association and use strength association of leakage occurrence to form leakage rule data, fully utilizing an artificial intelligent algorithm to carry out deep mining on the historical data, so that the system has predictive capability, and radically changing the passive response mode of the traditional leakage management. And calculating the failure probability of each pipeline section and equipment according to the leakage rule data, combining the maintenance cost and the resource limiting parameters, generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion, realizing the optimal configuration of maintenance resources by applying an artificial intelligence algorithm, and maximally reducing the leakage risk under the condition of limited resources. And finally, constructing a liquid system running state diagram based on maintenance strategy data, integrating real-time monitoring data and maintenance execution conditions, generating a management-hierarchy leakage prevention and control data report, meeting the information requirements of different management levels, and supporting management activities of all levels from operation to decision. According to the application, the artificial intelligent algorithm is deeply fused with the domain knowledge, and the accuracy of leakage detection is remarkably improved through acquisition, processing, analysis and visualization of leakage data.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of an embodiment of a data processing method based on liquid leak detection in an embodiment of the present application;

FIG. 2 is a schematic diagram of one embodiment of a data processing system based on liquid leak detection in an embodiment of the application;

Fig. 3 is a block diagram schematically illustrating the structure of a computer device in an embodiment of the present invention.

Detailed Description

The embodiment of the application provides a data processing method and system based on liquid leakage detection. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

For ease of understanding, a specific flow of an embodiment of the present application is described below with reference to fig. 1, where an embodiment of a data processing method based on liquid leak detection in an embodiment of the present application includes:

step S101, collecting monitoring signals of a photoelectric liquid leakage sensor through an RS485 bus, and carrying out signal denoising and outlier filtering on the monitoring signals to obtain leakage basic data comprising sensor ID, position information, detection time stamp and signal strength;

Step S102, calculating the change rate and duration of the signal intensity according to the leakage basic data, and combining the topological structure of the building water supply and drainage system to obtain leakage characteristic data for marking the leakage type and the influence range;

Step S103, matching a risk threshold value based on the identification leakage type, and determining a leakage risk level through comparing leakage characteristic data with the risk threshold value;

step S104, carrying out correlation analysis on time, position and type of leakage risk level and historical leakage record, extracting periodic distribution, equipment aging correlation and use intensity correlation of leakage occurrence, and forming leakage rule data;

step S105, calculating failure probability of each pipeline section and equipment according to the leakage rule data, and generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion by combining maintenance cost and resource limiting parameters;

And step S106, constructing a liquid system running state diagram based on the maintenance strategy data, integrating the real-time monitoring data and the maintenance execution condition, and generating a management-hierarchy leakage prevention and control data report.

It will be appreciated that the subject of the present application may be a data processing system based on liquid leak detection, or may be a terminal or server, and is not limited in this regard. The embodiment of the application is described by taking a server as an execution main body as an example.

Specifically, monitoring signals are collected from photoelectric liquid leakage sensors distributed at key positions of a building through an RS485 bus. These sensors are installed in key positions such as toilets, tea rooms, equipment rooms, roof pump rooms of buildings, etc. where liquid leakage is likely to occur, and when liquid leakage occurs, the optical path inside the sensor is changed, thereby generating a monitoring signal. These signals are transmitted to the central processing unit through the RS485 bus, and as the sensing signals are susceptible to electromagnetic interference and environmental noise, signal denoising and outlier filtering processing are required. The system adopts a Kalman filtering method to denoise an original signal, and the method effectively removes environmental interference noise through a prediction-correction iterative process to obtain a primary filtering signal. And then comparing the signals with the historical state data of the sensor, and eliminating abnormal jump values caused by sudden electromagnetic interference to obtain effective signal data. The effective signal data is converted into standardized liquid leakage signal intensity by combining with a sensor calibration curve, and combined with a sensor ID, position information and a detection time stamp, leakage basis data is constructed. The system calculates the rate of change and duration of signal strength from the leakage basis data. The signal intensity change rate is obtained by performing time-series differential calculation on the signal intensity in the leakage basis data, and represents the amount of change in signal intensity per unit time. The duration is obtained by analyzing the continuity of the rate of change of the signal intensity, and represents the length of the time period in which the signal intensity continuously changes. The system extracts the topological structure of the building water supply and drainage system from the building information database and marks the relative position relation of each sensor in the water supply and drainage system. The system identifies the leak type as a sudden leak when the rate of change of signal strength detected by the sensor is above a preset threshold (e.g., 0.5V/s), and as a slow leak when the rate of change is below the threshold but for a longer duration. And combining the leakage duration time with the topological structure of the building water supply and drainage system, and calculating the possible flowing direction and area of the liquid by using a fluid diffusion propagation algorithm to form an influence range. This information is integrated into leakage characteristic data.

Based on the identified leakage type matching the corresponding risk threshold, a risk threshold parameter set corresponding to the leakage type is extracted from the leakage risk database. Based on the range of influence in the leakage characteristic data, the system calculates a leakage amount estimation value while taking into account the importance coefficient of the leakage area. For example, the importance coefficient (e.g., 0.9) between devices is higher than that of a general toilet (e.g., 0.6). The system ranks the leakage by comparing the leakage estimate to a critical value in the risk threshold parameter set. Meanwhile, the system evaluates the value of facilities around the leakage area, calculates an area sensitivity index, comprehensively considers the factors, and determines the final leakage risk level as light, moderate or heavy risk. And carrying out association analysis on the leakage risk level and the historical leakage record. The system retrieves historical leakage records from the leakage event database, performs multidimensional matching with the current leakage risk level, and constructs a leakage event time sequence table containing space-time marks. The time sequence table is analyzed through Fourier time-frequency transformation, the leakage distribution characteristics of a long period, a medium period and a short period are separated by the system, and the periodic distribution is extracted. And combining the difference value of the equipment installation time and the leakage occurrence time, constructing a nonlinear regression equation by the system, and acquiring the equipment aging relevance. And obtaining the correlation of the use intensity of each area by calculating the correlation coefficient of the use intensity and the leakage event. The analysis results are fused through a self-adaptive weight network to form leakage risk quantization indexes, and the leakage risk quantization indexes are combined with the hydraulic characteristics of the geographic positions and the failure modes of the equipment types to construct leakage rule data.

The failure probability of each pipe section and equipment is calculated. The system converts periodic distribution data in the leakage regular data into a time sequence predicted value, and counts the number of leakage events in each time period, multiplied by the weight of the severity, divided by the length of the time period to obtain a leakage risk value in unit time. According to the equipment aging relevance data, the system extracts the initial failure rate of the equipment, multiplies the initial failure rate by the square root of the service time, and adds the environmental factor influence value to calculate the failure probability. In combination with maintenance costs (including equipment prices, installation man-hours, material fees) and resource limitation parameters (including available man-hours, number of stock equipment, budget limits), the system calculates a cost-benefit ratio, sets inspection frequency and maintenance period, generates equipment update advice, and forms maintenance policy data. A liquid system operational state diagram is constructed based on the maintenance policy data. The system superimposes the topological frame of the pipeline distribution and the equipment position with failure probability data to divide a risk area. And fusing the real-time monitoring data with the running state diagram to form a monitoring situation diagram. And extracting maintenance execution records from the work order system, and calculating the task completion rate and the solution efficiency. According to different management levels, the system generates reports of different levels and integrates the reports into a unified leakage prevention and control data report.

Taking a commercial building as an example, 235 photoelectric liquid leakage sensors are deployed on 58 floors, and the system realizes comprehensive monitoring and prevention of liquid leakage of the building through the data processing method. When the sensor between the 35 th layer equipment detects that the signal intensity is rapidly increased from 0.2V to 3.5V, the system judges that the sudden leakage is serious in risk grade, and automatically closes the related electromagnetic valve and informs maintenance personnel. Through historical data analysis, the system finds that similar problems often occur in the 18 th month after installation of the pipeline connection of the area, a targeted maintenance strategy is generated accordingly, the inspection frequency is increased from quarter to month, and the occurrence frequency of leakage events is effectively reduced.

In the embodiment of the application, the monitoring signal of the photoelectric liquid leakage sensor is acquired through the RS485 bus, and the signal denoising and outlier filtering are carried out to obtain high-quality leakage basic data, so that the problems of unstable signal and easy interference of the traditional leakage detection system are solved, and the accuracy of leakage detection is greatly improved. And secondly, calculating the change rate and duration of the signal intensity according to the leakage basic data, combining the topological structure of the building water supply and drainage system, realizing the accurate identification of the leakage type and the influence range, and overcoming the limitation that the traditional method can not distinguish the sudden leakage and the slow leakage. By matching the risk threshold value based on the identification leakage type, the leakage characteristic data is compared with the risk threshold value, so that the scientific and reasonable leakage risk level is determined, the situation that the traditional leakage risk assessment is too subjective and rough is changed, and the quantification and standardization of the risk assessment are realized. And carrying out time, position and type association analysis on the leakage risk level and the historical leakage record, extracting periodic distribution, equipment aging association and use strength association of leakage occurrence to form leakage rule data, fully utilizing an artificial intelligent algorithm to carry out deep mining on the historical data, so that the system has predictive capability, and radically changing the passive response mode of the traditional leakage management. And calculating the failure probability of each pipeline section and equipment according to the leakage rule data, combining the maintenance cost and the resource limiting parameters, generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion, realizing the optimal configuration of maintenance resources by applying an artificial intelligence algorithm, and maximally reducing the leakage risk under the condition of limited resources. And finally, constructing a liquid system running state diagram based on maintenance strategy data, integrating real-time monitoring data and maintenance execution conditions, generating a management-hierarchy leakage prevention and control data report, meeting the information requirements of different management levels, and supporting management activities of all levels from operation to decision. According to the application, the artificial intelligent algorithm is deeply fused with the domain knowledge, and the accuracy of leakage detection is remarkably improved through acquisition, processing, analysis and visualization of leakage data.

In a specific embodiment, the process of executing step S101 may specifically include the following steps:

(1) Dividing the photoelectric liquid leakage sensor into a plurality of monitoring areas according to a building floor structure, and forming a monitoring network in each monitoring area through an RS485 bus;

(2) Removing environmental interference noise from the monitoring signal by adopting a Kalman filtering method to obtain a primary filtering signal;

(3) Comparing the primary filtering signal with the historical state data of the sensor, and eliminating abnormal jump values caused by electromagnetic interference to obtain effective signal data;

(4) Converting the effective signal data into standardized liquid leakage signal intensity by combining the effective signal data with a sensor calibration curve;

(5) The normalized liquid leak signal strength is combined with the sensor ID, location information, and detection time stamp to construct leak basis data.

The photoelectric liquid leakage sensor is divided into a plurality of monitoring areas according to the floor structure of a building, and the photoelectric liquid leakage sensor is deployed at key positions of different floors according to the spatial structural characteristics of the building and the liquid leakage risk distribution condition. The division of the monitoring areas follows the "nearest law" and the "risk concentration law", and the similar functional areas of the same floor or adjacent floors are divided into one monitoring unit, for example, a 58-layer building is divided into three main monitoring areas of a low area (1-20 layers), a middle area (21-40 layers) and a high area (41-58 layers), and the inside of each area is subdivided into a toilet area, an equipment area and a public area according to functions. In each monitoring area, the sensors are connected through an RS485 bus to form a monitoring network, the RS485 bus adopts a master-slave communication architecture, the multipoint data transmission is supported, the anti-interference capability is strong, the transmission distance can reach 1200 meters, and the vertical space of a large building can be covered. Each sensor is allocated with a unique address code, the central controller sends a data request instruction to each sensor in a polling mode, and the sensor returns a current monitoring signal value after receiving the instruction.

And (3) removing the environmental interference noise by adopting a Kalman filtering method for the acquired monitoring signals. Kalman filtering is a recursive filtering algorithm that processes signals through two stages, prediction and correction. The prediction stage predicts the state of the current moment by using the state estimation and the error covariance of the previous moment, and the correction stage corrects the predicted value through the actual measured value. For signal processing of the liquid leakage sensor, a signal state equation is firstly established, the sensor signal is expressed as a form of real leakage state plus random noise, then the statistical characteristics of the noise are determined according to the historical data of the sensor, and the optimal estimated value is calculated. In this way, even in strong electromagnetic interference environments such as elevator machine rooms, water pump rooms, etc., a smooth and stable primary filtered signal can be obtained. And comparing and analyzing the primary filtering signal and the historical state data of the sensor, and eliminating abnormal jump values caused by electromagnetic interference. The specific method is to construct a historical state database of the sensors and record signal characteristic distribution of each sensor under normal conditions, wherein the signal characteristic distribution comprises signal mean value, standard deviation and variation trend. And calculating the deviation of the current received primary filtering signal from the historical mean value, and comparing the deviation with a preset threshold value. When the deviation exceeds a threshold value and the front and back signal changes do not accord with the physical change rule, the abnormal jump value is judged and removed, and effective signal data accord with the normal change rule is reserved. The method is particularly suitable for processing signal anomalies caused by short-time strong interference, such as electromagnetic interference generated by elevator starting and high-power equipment switching moment.

After the effective signal data is acquired, it is converted to normalized liquid leakage signal intensity in combination with the sensor calibration curve. The sensor calibration curve is a corresponding relation curve between the liquid leakage amount and the signal output by exposing the sensor to environments with different liquid amounts under laboratory conditions and measuring the output signal value. There may be sensitivity differences between different batches and models of sensors, so each sensor has its unique calibration curve parameters. The conversion process firstly reads the calibration curve parameters of the corresponding sensor from the sensor configuration database, then brings the effective signal data into the calibration function to calculate, and obtains the standardized liquid leakage signal strength, wherein the strength value is uniformly expressed by the liquid infiltration height (millimeter), so that cross-equipment comparison and threshold setting are facilitated. The normalized liquid leakage signal strength is combined with the sensor ID, the location information and the detection timestamp to construct structured leakage basis data. The sensor ID is a unique identification code assigned to each sensor by the system, typically using a 16-bit hexadecimal number, containing a vendor code, a type code, and a serial number. The position information comprises the installation floor, the area code and specific coordinates of the sensor, and is expressed by a unified space reference system. The detection time stamp adopts a UNIX time format with millisecond precision to record the accurate time of data acquisition. This information, along with the normalized liquid leakage signal strength, is organized into leakage basis data according to predefined data structure templates, stored in a system database, and used as input for subsequent leakage signature analysis.

Taking a commercial complex building as an example, the building has 58 layers, 235 photoelectric liquid leakage sensors are installed in areas such as a toilet, an equipment room and the like, and 3000 m induction wires are used. The system divides the building into 6 monitoring areas, and each area is connected with a sensor in the area through an independent RS485 bus to form a monitoring network. When a sensor of the 28-floor toilet detects leakage of the water pipe interface, the original signal is in a fluctuation state by electromagnetic interference generated by operation of a nearby elevator, and the original value is jumped between 0.2V and 1.8V. After the kalman filter is applied, a smoothed primary filtered signal is obtained of about 0.85V. The system searches the historical state data of the sensor, finds that the signal value is between 0.1V and 0.3V under normal conditions, the current value is obviously higher, but the change trend accords with the characteristic of liquid leakage, so that the signal is reserved as effective signal data. According to the calibration curve of the model sensor, the liquid infiltration height corresponding to the signal intensity of 0.85V is 5.2 millimeters, and a complete leakage basis data is constructed by combining the sensor ID (SAF 2374E6C 9), position information (28 layers of northwest angles of men and women), and detection time stamps (2023-08-1514:28:37.245), so that an accurate data basis is provided for subsequent leakage type judgment and risk assessment.

In a specific embodiment, the process of executing step S102 may specifically include the following steps:

(1) Carrying out time sequence difference on the signal intensity in the leakage basic data, and calculating the signal intensity variation in unit time to obtain the signal intensity variation rate;

(2) Continuously analyzing the signal strength change rate, counting the time period length of continuous change of the signal strength, and obtaining leakage duration;

(3) Extracting a topological structure of a building water supply and drainage system from a building information database, and marking the relative position relation of each sensor in the water supply and drainage system;

(4) Comparing the signal strength change rate with a preset change rate threshold, and identifying the sudden leakage type when the change rate is higher than the threshold and the slow leakage type when the change rate is lower than the threshold;

(5) Calculating the possible flowing direction and area of the liquid by adopting a fluid diffusion propagation algorithm based on the leakage duration and the topological structure of the building water supply and drainage system to form an influence range;

(6) And integrating the leakage type, the influence range and the original leakage basic data to construct leakage characteristic data for identifying the leakage type and the influence range.

Specifically, in the data processing method based on liquid leakage detection, after acquiring leakage basis data, it is necessary to further extract leakage characteristic information. Firstly, carrying out time series difference on signal intensity in leakage basic data, and calculating signal intensity variation in unit time to obtain a signal intensity variation rate. Time series difference is a common data processing method, and the change trend of data is reflected by calculating the data difference between adjacent time points. For liquid leak detection, the signal strength change rate calculation formula is:

Wherein DeltaI _t represents the signal intensity change rate at the time t, the unit is V/s, I _t represents the standardized liquid leakage signal intensity at the time t, I _t-τ represents the standardized liquid leakage signal intensity at the time t-tau, and tau represents the sampling time interval, and the unit is s. By this formula, discrete signal strength data is converted to a continuous variable reflecting the leak rate.

Next, the signal intensity change rate is continuously analyzed, the time period length of the continuous change of the signal intensity is counted, and the leakage duration is obtained. The continuity analysis first sets a baseline threshold value Δi _base for the rate of change of the signal, and considers the signal to be in a continuously changing state when the absolute value of the rate of change of the signal at a plurality of consecutive time points is greater than the threshold value. Duration T _leak is calculated as the time difference from the point in time when the rate of change of the signal first exceeded the baseline threshold to the point in time when the baseline threshold was last exceeded. This method effectively distinguishes between temporary fluctuations in the signal and real leakage events.

Meanwhile, the topological structure of the building water supply and drainage system is extracted from the building information database, and the relative position relation of each sensor in the water supply and drainage system is marked. The topology structure of the building water supply and drainage system refers to the spatial layout and connection relation of water supply pipelines, drainage pipelines, valves, water tanks and other devices in the building. The database stores the information of pipeline trend, diameter, material, connection point type and the like, and the installation position of each sensor in the network. By extracting the information, a spatial mapping relation between the sensor network and the water supply and drainage system is constructed, and a spatial basis is provided for subsequent leakage influence range analysis.

And comparing the signal strength change rate with a preset change rate threshold value to determine the leakage type. The change rate threshold ΔI _threshold is a critical value determined based on empirical data and experimental analysis, and is typically set at 0.5V/s. When the signal strength change rate ΔI _t>ΔI_threshold, it indicates a fast liquid leakage rate, which is identified as a sudden leakage type, and when 0< ΔI _t≤ΔI_threshold, it indicates a slow liquid accumulation or leakage, which is identified as a slow leakage type.

Based on the leakage duration and the topological structure of the building water supply and drainage system, the direction and the area in which the liquid possibly flows are calculated by adopting a fluid diffusion propagation algorithm, so that an influence range is formed. The fluid diffusion propagation algorithm simulates the flow path of liquid in a building by taking into consideration the factors of gravity, water absorption of ground materials, ground gradient and the like. The algorithm takes the leakage point as a starting point, and calculates the range and the direction of possible liquid diffusion according to the spatial characteristics and the physical laws of the building. This diffusion analysis based on a physical model reflects the actual leakage situation more accurately than a simple geometrical diffusion. And integrating the leakage type, the influence range and the original leakage basic data to construct leakage characteristic data for identifying the leakage type and the influence range. The integration process adopts a structured data format, and comprises sensor basic information, leakage characteristic information and influence range information. Such structured leakage feature data facilitates subsequent risk level assessment and emergency handling.

In a specific embodiment, the process of executing step S103 may specifically include the following steps:

(1) Extracting a risk threshold parameter set corresponding to the identification leakage type from a leakage risk database, wherein the risk threshold parameter set comprises liquid leakage quantity critical values of different region types;

(2) Calculating a leakage quantity estimated value according to the influence range in the leakage characteristic data, and simultaneously acquiring an importance coefficient of a leakage area from a building information system;

(3) Comparing the leakage quantity estimated value with a liquid leakage quantity critical value in the risk threshold parameter set, and dividing the leakage quantity estimated value into a primary leakage value, a middle-level leakage value and a high-level leakage value;

(4) Performing value evaluation on surrounding facilities of the leakage area, and calculating an area sensitivity index based on the importance degree and the possible loss degree of the facilities;

(5) The regional sensitivity index is weighted and combined with the primary leakage value, the intermediate leakage value and the high leakage value to form a risk score;

(6) And determining the leakage risk level as a mild risk, a moderate risk or a severe risk according to the score interval in which the risk score is located.

Specifically, a risk threshold parameter set corresponding to the identified leak type is extracted from a leak risk database. The leakage risk database is a data storage system for storing different types of leakage risk assessment criteria, and comprises risk parameters of two leakage types, namely sudden leakage and slow leakage. The risk threshold parameter set contains liquid leakage threshold values of different area types, and the threshold values are preset according to the characteristics and importance of different functional areas in the building. For example, the leak volume threshold for a data center room area is lower than a normal office area because the same volume of liquid leak may have more serious consequences in a data center.

A leakage amount estimation value is calculated from an influence range in the leakage characteristic data. The leakage amount estimation is calculated using the product of the area of the impact and the average liquid depth. The specific formula is as follows:

V_leak＝A_impact×D_avg×F_corr

Wherein V _leak represents an estimated leakage amount in cubic meters, A _impact represents an area of an affected area in square meters, D _avg represents an average liquid depth in meters, and F _corr represents a correction factor, dimensionless. The correction factor takes into account factors that affect the actual leakage amount, such as the ground material water absorption, evaporation rate, etc. Meanwhile, the importance coefficient of the leakage area is obtained from the building information system, and the coefficient is a quantization index for reflecting the importance degree of the area to the building function, and is generally between 0 and 1, and the larger the value is, the more important the area is. And comparing the leakage quantity estimated value with a liquid leakage quantity critical value in the risk threshold parameter set, and dividing the leakage quantity estimated value into a primary leakage value, a middle-level leakage value and a high-level leakage value. The comparison process selects the corresponding critical value standard according to the type of the leakage area. The leakage amount estimation value is divided into a primary leakage value when the leakage amount estimation value is smaller than or equal to a first critical value, a middle leakage value when the leakage amount estimation value is larger than the first critical value but smaller than or equal to a second critical value, and a high leakage value when the leakage amount estimation value is larger than the second critical value. This hierarchical approach provides the underlying data for subsequent risk assessment.

The surrounding facilities of the leak area are evaluated for value, and an area sensitivity index is calculated based on the importance of the facilities and the extent of possible loss. The regional sensitivity index comprehensively considers the direct value, functional importance and substitution difficulty of the facility. The calculation method is to carry out weighted summation on the evaluation values of all facilities in the area, and the weight is determined according to the relative importance of the facilities. This way of assessment reflects the overall level of economic loss and functional impact that a leak event may cause.

The regional sensitivity index is weighted with the primary leakage value, the intermediate leakage value, and the high leakage value to form a risk score. The weighted combination adopts a linear weighting method, and the leakage value grade and the regional sensitivity are respectively assigned with weights, so that the relative importance of the leakage value grade and the regional sensitivity in risk assessment is reflected. The risk score calculation formula is:

R_score＝w₁×L_value+w₂×S_index

Wherein, R _score represents a risk score, a dimensionless value, L _value represents a score corresponding to a leakage value grade, S _index represents a regional sensitivity index, w ₁ and w ₂ represent weight coefficients of the leakage value grade and the regional sensitivity, respectively, and w ₁+w₂ =1. In this way, both the severity of the leak itself and the actual impact that the leak may have are taken into account.

And determining the leakage risk level as a mild risk, a moderate risk or a severe risk according to the score interval in which the risk score is located. The specific division basis is a preset score interval range, namely, the light risk is judged when the risk score is between 0 and 40, the medium risk is judged when the risk score is between 41 and 70, and the heavy risk is judged when the risk score is greater than 70. Different risk classes correspond to different response policies and processing priorities.

Taking a certain comprehensive office building as an example, a leakage event caused by loosening of a water pipe interface occurs between devices of a 32 th layer. First, a risk threshold parameter set corresponding to the "slow leak" leak type is extracted from the leak risk database, and the liquid leak amount critical values are 0.05 cubic meters (first critical value) and 0.15 cubic meters (second critical value) for this region type between devices, respectively. The leakage characteristic data shows an impact range of 10 square meters, and the estimated leakage amount is calculated to be 0.08 cubic meters, and the importance coefficient between the devices is obtained from the building information system to be 0.85 (because the devices contain important network communication devices). And comparing the estimated leakage value of 0.08 cubic meters with a critical value to determine a medium-grade leakage value. The facilities in the area are evaluated, key equipment such as a network switch, a distribution box and the like is included in the equipment room, and the area sensitivity index is calculated to be 78. The mid-level leakage value (corresponding score 50) was weighted with the regional sensitivity index 78, weights 0.6 and 0.4, respectively, to yield a risk score of 61.2. According to the score interval division standard, 61.2 falls into the interval 41-70, so that the risk level of the leakage event is determined to be a moderate risk, a corresponding alarm mechanism and a corresponding processing flow are triggered, and the method comprises the steps of dispatching maintenance personnel to carry out inspection and maintenance and informing a network administrator to closely monitor the running state of equipment.

In a specific embodiment, the process of executing step S104 may specifically include the following steps:

(1) Retrieving a historical leakage record from a leakage event database, performing multidimensional matching with the current leakage risk level, and constructing a leakage event time sequence table containing space-time marks;

(2) Applying Fourier time-frequency transformation to the leakage event time sequence table, separating leakage distribution characteristics of a long period, a medium period and a short period, and extracting periodic distribution through a peak detection algorithm;

(3) According to the time difference between the equipment installation time and the leakage occurrence time in the historical leakage record, constructing a nonlinear regression equation by combining equipment material parameters and running environment factors, and acquiring equipment aging relevance;

(4) Collecting people flow density data and equipment operation frequency data of different areas of a building, and determining the coupling coefficient of the use intensity of each area and a leakage event through a cross entropy calculation method to obtain the correlation of the use intensity;

(5) Carrying out multi-level fusion on the periodic distribution, the equipment aging relevance and the use intensity relevance through a self-adaptive weight network to form a leakage risk quantization index with space-time prediction capability;

(6) And combining the leakage risk quantization index with the hydraulic characteristics of the geographic position and the failure modes of the equipment type to construct leakage rule data with a prediction function.

Specifically, the historical leakage record is retrieved from the leakage event database and multidimensional matching is performed with the current leakage risk level. The leakage event database is a structured data warehouse for storing detailed information of all historical leakage events in a building, and comprises multi-dimensional data such as leakage time, position, type, reason, processing mode and the like. The multidimensional matching refers to that according to the characteristics (such as region type, equipment type and leakage type) of the current leakage event, historical events with similar characteristics are screened out from a database, and a leakage event time sequence table containing space-time marks is constructed. The space-time markers refer to the occurrence time and spatial position information of each leakage event, and facilitate subsequent time sequence analysis and spatial correlation analysis.

Fourier time-frequency transformation is applied to the leakage event schedule to isolate leakage distribution characteristics of different periods. Fourier time-frequency transformation is a mathematical tool that converts a time-domain signal into the frequency domain, revealing implicit periodic components in the signal. In the liquid leakage analysis, a leakage event time sequence table is regarded as a time sequence signal, the frequency spectrum characteristics of the leakage event time sequence table are calculated through Fourier transformation, and periodic components with different frequencies are identified. The long period generally refers to seasonal or annual leakage laws, such as peak leakage due to frost cracking in winter, the medium period refers to monthly or periodic laws, such as leakage laws caused by difference in use intensity between weekdays and weekends, and the short period refers to intra-day or hourly laws, such as leakage caused by pressure fluctuation during peak periods of water. And identifying a significant peak value on the spectrogram through a peak detection algorithm, and extracting periodic distribution data corresponding to leakage distribution characteristics of different periods.

And constructing a nonlinear regression equation according to the time difference between the equipment installation time and the leakage occurrence time in the historical leakage record and combining the equipment material parameters and the running environment factors to obtain the equipment aging relevance. The mathematical expression of the nonlinear regression equation is:

P_fail(t)＝α·e^β·t+γ·t²+λ·M_factor·E_coef+ε

Wherein P _fail (t) represents the failure probability of the equipment after the use time t, alpha, beta, gamma and lambda are regression coefficients and are obtained through historical data fitting, t is the equipment use time in months, M _factor is equipment material parameters, pipelines and joints made of different materials have different durability, E _coef is an operation environment factor comprising factors such as temperature fluctuation, humidity and water quality, and epsilon is a random error term. Through the equation, the relation between the service life of the equipment and the leakage risk is quantized, and a basis is provided for updating decisions of the equipment.

And collecting people flow density data and equipment operation frequency data of different areas of the building, and determining the coupling coefficient of the use intensity of each area and the leakage event by a cross entropy calculation method to obtain the correlation of the use intensity. The mathematical expression of the cross entropy calculation method is:

H(p,q)=-∑_xp(x)logq(x)

Wherein H (p, q) represents cross entropy, p _usage represents area use intensity distribution, q _leak represents leakage event distribution, H _max represents normalization factor, C _coupling represents coupling factor, U _factor represents use type factor, p (x) represents actual distribution of area use intensity, probability distribution of actual use frequency or intensity of each area of a building, q (x) represents distribution of leakage event, probability distribution of occurrence of leakage event of each area is represented, and x represents different areas or position points in the building. Reflecting the wear impact of different modes of use on the device. The smaller the cross entropy, the more similar the intensity distribution is used to the leakage event distribution, the more strongly correlated the two are. In this way, the extent of correlation of the area usage pattern with the risk of leakage is quantified. And carrying out multi-level fusion on the periodic distribution, the equipment aging relevance and the use intensity relevance through a self-adaptive weight network to form a leakage risk quantization index with space-time prediction capability. The self-adaptive weight network is a calculation model for automatically adjusting the weight of each index according to the data characteristics, and can adapt to complex nonlinear relations more than the linear combination of fixed weights. The fusion process comprises four steps of data standardization, correlation analysis, weight calculation and weighted summation, and a final leakage risk quantization index is generated. And combining the leakage risk quantization index with the hydraulic characteristics of the geographic position and the failure modes of the equipment type to construct leakage rule data with a prediction function. The hydraulic characteristics comprise physical parameters such as pipe network pressure distribution, flow rate change, water hammer effect and the like, and the equipment failure modes comprise typical failure types such as corrosion perforation, interface loosening, sealing element aging and the like. Through the comprehensive analysis, the generated leakage rule data not only comprises quantitative prediction of leakage risks, but also comprises possible leakage positions, leakage types, optimal overhaul time and other information.

Taking a large commercial building as an example, the liquid leak detection system records leak event data over three years. When a moderate risk leakage event occurs between the 42 th floor devices, a total of 87 pieces of data are retrieved from the leakage event database. 32 records similar to the current event characteristics are screened out through multidimensional matching of time, position and leakage type, and a leakage event time sequence table containing detailed space-time marks is constructed. Fourier time-frequency transform analysis was applied to the schedule, and significant 90-day periods (corresponding to seasonal changes) and 7-day periods (corresponding to stress changes in restarting after a weekend device shutdown) were found. The periodic distribution characteristics are extracted through a peak detection algorithm, so that a basis is provided for predicting the next possible leakage time.

Meanwhile, the average installation time of the regional equipment is 36 months, and leakage is more likely to occur between 24 and 30 months after installation. The correlation function of equipment aging and leakage is obtained by fitting a nonlinear regression equation in combination with the material parameters (good durability and easy influence of water quality) and environmental factors (large temperature fluctuation and hard water quality) of the copper pipeline used in the area. The function shows that the probability of failure starts to increase significantly after 28 months of use. By analyzing the people stream data and the equipment operation log, the use intensity of the area on the working day is found to be 2.7 times of that of the weekend, and the coupling coefficient of the use intensity and the leakage event is obtained by cross entropy calculation to be 0.78, which shows that the use intensity is highly related to the leakage risk. The analysis results are fused through a self-adaptive weight network to form leakage risk quantification indexes, and complete leakage rule data is constructed by combining the characteristics of the regional pipe network such as large pressure fluctuation, frequent valve operation and the like and the characteristics of loose interfaces as main failure modes. Based on the data, the next potential leakage event is predicted to happen after 35 days, and preventive inspection and maintenance should be carried out within 30 days, so that equipment damage and service interruption are effectively avoided.

In a specific embodiment, the process of executing step S105 may specifically include the following steps:

(1) Sequencing periodic distribution data in the leakage regular data according to time nodes, counting the number of leakage events in each time period, multiplying the number by the weight of the leakage severity, dividing the weight by the length of the time period to obtain leakage risk values in unit time, and calculating leakage risk coefficients of all pipeline sections;

(2) Extracting the initial failure rate of the equipment according to the equipment aging association, multiplying the initial failure rate by the square root of the used time of the equipment, and adding the linear combination of the environmental corrosion factor and the used pressure coefficient to calculate the failure probability of each pipeline section and the equipment;

(3) Extracting the original price, the installation time and the replacement material cost of each device from the facility management system, multiplying the original price, the installation time and the replacement material cost by the expansion coefficient of the general cargo, and adding the estimated value of economic loss during the shutdown period to construct a maintenance cost matrix;

(4) Extracting the total number of monthly available time, the number of inventory devices and the budget limit based on an enterprise resource management system, and respectively setting a human resource upper limit, a material resource upper limit and a cost resource upper limit to form a resource limiting parameter set;

(5) Multiplying the failure probability by the failure loss evaluation value, dividing the failure probability by the single checking maintenance cost to obtain a cost benefit ratio, and setting checking frequency and maintenance period according to the three steps of high, medium and low when the ratio is larger than the resource allocation threshold value;

(6) Multiplying the failure probability of the equipment by the cost benefit ratio, subtracting the equipment updating difficulty coefficient to obtain updating priority score, and generating equipment updating suggestions according to the score order to finally form maintenance strategy data.

Specifically, the periodic distribution data in the leakage regular data is ordered according to time nodes, in such a way that historical leakage events are arranged from early to late according to occurrence time and are divided into time periods with fixed lengths, such as monthly, quarterly or annually. And counting the leakage events occurring in each time period to obtain the number of the events, and multiplying the number by the weight of the leakage severity. The leakage severity weight is a value determined according to the leakage risk level, the light risk value is 1, the moderate risk value is 2, and the heavy risk value is 5, and the difference of the hazard degrees of leakage events of different risk levels is reflected. Dividing the weighted event number by the time period length, calculating leakage risk values in unit time, and distributing the risk values to each pipeline section according to the pipeline connection relation and the spatial distribution characteristics to obtain leakage risk coefficients of each pipeline section. The leakage risk coefficient directly reflects the probability of leakage of the pipeline, and provides basic data for subsequent failure probability calculation.

The initial failure rate of the equipment is extracted according to the aging association data of the equipment, and the initial failure rate of the equipment refers to the basic failure rate of the new installation state of the equipment under normal use conditions, and is usually provided by a manufacturer or obtained through experimental tests. The method is characterized in that the initial failure rate of the equipment is multiplied by the square root of the used time of the equipment, the calculation mode reflects the rule that the failure rate of the equipment is gradually increased along with the increase of the used time, and the square root function enables the failure rate to be slow in the initial stage and accelerated in the later stage, so that the aging characteristics of most of the equipment are met. Then a linear combination of the environmental corrosion factor and the pressure coefficient is used. The environmental corrosion factor considers the influences of humidity, temperature fluctuation, water quality and the like on the service life of the equipment, and the pressure coefficient reflects the influence of the ratio of the working pressure of the equipment to the rated pressure on the fault rate. The failure probability of each pipeline section and equipment is obtained through the comprehensive calculation, and a quantitative basis is provided for maintenance decision.

Basic data such as the original price, the installation man-hour, and the replacement material cost of each device are extracted from the facility management system. The original price is the purchase cost of the equipment, the installation time is the manual time required for completing the installation of the equipment, and the replacement material cost comprises auxiliary materials and consumable cost required in the replacement process. These base cost data are multiplied by the generic expansion coefficients to make adjustments reflecting the current actual cost level. And then adding an estimated economic loss value during the shutdown, wherein the economic loss of the shutdown comprises indirect loss such as income loss caused by service interruption during maintenance, customer satisfaction reduction and the like. By integrating these cost factors, a maintenance cost matrix is constructed, which is a two-dimensional table, with rows representing different equipment types, columns representing different maintenance types (e.g., daily inspections, periodic maintenance, and equipment replacement), and matrix element values being the corresponding maintenance costs. The maintenance cost matrix provides a data basis for subsequent cost-benefit analysis.

The maintenance resource limit data is extracted based on the enterprise resource management system, including a total number of monthly available hours, a number of inventory devices, and a budget limit. The total number of monthly available hours refers to the total working time that maintenance personnel can use for equipment maintenance in one month, the number of stock equipment refers to the number of spare parts available for replacement in a warehouse, and the budget limit refers to the upper limit of funds allocated to maintenance work. The data are respectively set as a human resource upper limit, a material resource upper limit and a cost resource upper limit to form a resource limiting parameter set, and the resource limiting parameter set is used for restricting the establishment process of the maintenance strategy so as to ensure that the generated maintenance plan is within the resource available range.

The failure probability is multiplied by the failure loss evaluation value to calculate the total loss that may be caused when each equipment or pipe section fails. The fault loss assessment value comprises factors such as direct maintenance cost, indirect service loss, reputation influence and the like. The total loss is divided by the single check maintenance cost to yield a cost benefit ratio. The cost-effectiveness ratio measures the proportional relationship between preventive maintenance investment and possible avoided losses, with higher ratios indicating greater returns for maintenance investment. When the ratio is greater than the resource allocation threshold, preventive maintenance is indicated to be economically justified. The resource allocation threshold is a critical value set according to enterprise maintenance policies and risk tolerance. The inspection frequency and maintenance period of the equipment are set to three high, medium and low gears according to the magnitude of the cost-effectiveness ratio. The high-grade is corresponding to frequent inspection and maintenance and is generally applied to high-risk or critical equipment, the medium-grade is applied to general importance equipment, and the low-grade is applied to low-risk or non-critical equipment.

Multiplying the probability of failure of the device by the cost-effectiveness ratio, this calculation takes into account both the probability of failure of the device and the economic justification of preventive maintenance. And then subtracting the equipment updating difficulty coefficient, wherein the equipment updating difficulty coefficient reflects the technical complexity of equipment replacement, the influence degree on the system operation and the resource consumption level. Through this calculation, an updated priority score for each device is obtained. The higher the update priority score, the more the device needs to be prioritized for replacement. And generating equipment updating suggestions according to the score order, wherein the equipment updating suggestions comprise information such as suggested replacement time, replacement mode, expected return on investment and the like. And finally, integrating the inspection frequency, the maintenance period and the equipment updating suggestion into complete maintenance strategy data, and providing scientific basis for facility management.

Taking a commercial building as an example, a liquid leak detection system in a 58-story building detects a minute leak at the water supply pipe interface between the 25 th story facilities. By analyzing the leakage rule data, the frequency of leakage of the type of interface in the interval of 24-36 months after installation is found to be obviously improved, and most of the interfaces are in medium risk level. The data are time-segmented in quarters, each leakage event is multiplied by a corresponding severity weight (in this case, the weight is 2), and the leakage risk value of this type of interface per unit time in 3 rd year is calculated to be 0.18 times/month. The initial failure rate of the interface is 0.01 times/year, the service time is 30 months, the environmental corrosion factor and the service pressure coefficient are respectively 0.03 and 0.02, and the failure probability is calculated to be 0.12. The data is extracted from the facility management system, the cost for replacing the interface comprises 300 yuan of original price, 2 hours of installation time (100 yuan/hour) and 50 yuan of material cost, the expansion coefficient of the ventilation is 1.2, the shutdown loss is estimated to be 1000 yuan/time, and a maintenance cost matrix is constructed. The enterprise resource management system shows that available maintenance man-hour in the month is 200 hours, spare parts are sufficient in inventory, and maintenance budget is abundant. The failure loss evaluation value of this interface is calculated to be 5000 yuan, the maintenance cost of a single check is 150 yuan, the cost benefit ratio is 4, which is higher than the resource allocation threshold value 3, so that the check frequency is set to be high-grade (check once a month), and the maintenance period is set to be medium-grade (maintenance once a quarter). Considering that the update difficulty coefficient is 1.5, the update priority score is calculated to be 0.33, and the update priority score is ranked among all devices needing to be updated, and the replacement is recommended to be planned within one year. Through the data-driven decision process, scientific and reasonable maintenance strategy data are formed, and leakage risks and maintenance cost are effectively reduced.

In a specific embodiment, the process of executing step S106 may specifically include the following steps:

(1) According to the inspection frequency, the maintenance period and the equipment updating suggestion in the maintenance strategy data, combining the building space structure information to construct a liquid system topological frame reflecting the pipeline distribution and the equipment position;

(2) Overlapping a topological frame of the liquid system with failure probability, dividing the topological frame into a high-risk area, a medium-risk area and a low-risk area according to the risk degree, and generating a liquid system running state diagram;

(3) Acquiring latest leakage basic data from a central monitoring system, and fusing the latest leakage basic data with a liquid system running state diagram to form a real-time monitoring situation diagram;

(4) Extracting maintenance execution records from a work order management system, wherein the maintenance execution records comprise completion time, maintenance content and processing results, and calculating maintenance task completion rate and problem solving efficiency indexes;

(5) Carrying out data layering extraction on the real-time monitoring situation map and the maintenance execution record according to different levels of management staff to generate a facility operation layer report, a management decision layer report and a strategic planning layer report;

(6) Integrating the facility operation layer report, the management decision layer report and the strategic planning layer report into a unified leakage prevention and control data report, and archiving and storing according to a time sequence.

Specifically, according to the inspection frequency, the maintenance period and the equipment updating proposal in the maintenance strategy data, a liquid system topological frame reflecting the pipeline distribution and the equipment position is constructed by combining the building space structure information. The topological frame of the liquid system is a spatial relation diagram of a water supply and drainage pipeline network, valve equipment and sensors in a building, and a two-dimensional plane diagram and a vertical elevation diagram are integrated into a complete three-dimensional network structure by extracting spatial coordinate information and connection relation in a Building Information Model (BIM) database. The topological frame adopts a node-edge representation method, wherein nodes represent entities such as pipeline connection points, valves, sensors and the like, and edges represent pipeline connection relations. In the construction process, the maintenance strategy data are associated with the topology nodes, and the equipment positions with different inspection frequencies, maintenance periods and update priorities are marked to form a topology network diagram with maintenance attributes.

And then, superposing the topological frame of the liquid system with failure probability data, wherein the specific method is to endow each node and each edge in the topological frame with a corresponding failure probability value. Then dividing the risk degree into a high-risk region, a medium-risk region and a low-risk region, wherein the dividing standard is based on a failure probability threshold value, wherein the region with the failure probability larger than 0.3 is marked as the high-risk region and is represented by red, the region with the failure probability between 0.1 and 0.3 is marked as the medium-risk region and is represented by yellow, and the region with the failure probability smaller than 0.1 is marked as the low-risk region and is represented by green. By the color coding mode, leakage risk distribution of different areas in a building is intuitively displayed, and a liquid system running state diagram is generated. The running state diagram adopts a layered display mode, so that the whole distribution can be checked, and the local detail can be amplified and observed.

The latest leakage basic data is obtained from the central monitoring system, and comprises real-time monitoring information such as sensor ID, position information, detection time stamp, signal strength and the like. The real-time data and the running state diagram of the liquid system are subjected to space position matching and data fusion, corresponding nodes of the topological diagram are found according to the sensor ID, and the real-time monitoring data is used as node attributes to be added to the topological diagram, so that a real-time monitoring situation diagram with static risk distribution and dynamic monitoring states is formed. The situation map displays the currently active leakage points by using flashing marks, and the leakage severity is represented by different colors and flashing frequencies, and meanwhile, the leakage duration and the accumulated leakage quantity estimated value are displayed.

And extracting maintenance execution records from a work order management system, wherein the work order management system is an information system for recording equipment maintenance task allocation, execution and result feedback. The extracted data comprises the fields of work order number, maintenance task type, allocation time, completion time, maintenance content, processing personnel, processing results and the like. And carrying out statistical analysis on the records, and calculating the completion rate of the maintenance task and the problem solving efficiency index. The maintenance task completion rate is equal to the number of completed work orders divided by the total number of allocated work orders, and the problem solving efficiency index is equal to the number of work orders for solving the problem at one time divided by the total number of work orders. These indicators reflect the efficiency of the maintenance team and the performance of the maintenance strategy.

And carrying out data layering extraction on the real-time monitoring situation map and the maintenance execution record according to different levels of management personnel, and generating reports aiming at different management levels. The core of hierarchical extraction is to screen out relevant information from the original data and present the relevant information with proper granularity according to the responsibility range and focus of the management level. The facility operation layer report forms are oriented to maintenance technicians and comprise detailed equipment monitoring data, fault positions, specific maintenance tasks and operation instructions, information granularity is the finest, the management decision layer report forms are oriented to middle layer managers and comprise risk condition summaries, maintenance resource allocation suggestions and efficiency index analysis of all areas, management efficiency and resource optimization are concerned, the strategic planning layer report forms are oriented to high-level decision makers and comprise overall risk trend, major event summaries and resource input-output analysis, and information is the most summarized and focused. The report forms of all the layers adopt different data aggregation modes, the operation layer takes the original data as the main part, the management layer performs moderate summarization, and the strategic layer is highly aggregated into key performance indexes.

And finally, integrating the facility operation layer report, the management decision layer report and the strategic planning layer report into a unified leakage prevention and control data report, wherein abnormal indexes in the upper layer report can drill down layer by layer to check detailed data by adopting linkage design. The integrated report comprises four main blocks of risk overview, event tracking, maintenance efficiency and resource utilization, and each block has a corresponding chart, index and analysis text. And after the report is generated, archiving and storing are carried out according to the time sequence, and a historical report library is established, so that the risk change trend and maintenance effect evaluation can be conveniently tracked. The filing storage adopts a grading strategy, the recent report forms keep complete data, the long-term report forms keep key indexes and event records, and the storage requirements and the query efficiency are balanced.

Taking a commercial complex as an example, the building has 58 layers in total, and 235 liquid leakage sensors are deployed. Based on the maintenance strategy data generated in advance, a complete topological framework of the liquid system is constructed, and the framework comprises spatial position relations of all water supply and drainage pipelines, valves, water pumps and sensors. After the maintenance team completes the replacement of the pipeline interface between the 37 th layer devices according to the maintenance strategy, the work order system records the maintenance completion time, maintenance content and processing results. Meanwhile, the running state diagram is automatically updated, the area is adjusted from a high-risk area to a low-risk area, and the risk color is changed from red to green. When a new leakage signal appears in the 42 th-layer bathroom, the central monitoring system immediately transmits leakage basic data to a situation map, leakage points are marked on the situation map through spatial position matching, and the leakage points are judged to be medium risk according to the signal intensity and displayed by a yellow flashing mark. The maintenance manager checks detailed leakage positions and peripheral equipment conditions through the facility operation layer report, dispatches maintenance personnel for processing, the department manager knows that the recent leakage events are mainly concentrated in high-area toilets through the management decision layer report and decides to adjust maintenance resource allocation, and the general manager decides to continuously support preventive maintenance strategies according to the fact that the frequency of the leakage events is reduced and the maintenance cost is reduced since new maintenance strategies are implemented through the strategic planning layer report. The hierarchical data reporting system realizes seamless conversion from data to decision, and provides comprehensive support for the liquid leakage risk management of the building.

The above describes a data processing method based on liquid leak detection in the embodiment of the present application, and the following describes a data processing system based on liquid leak detection in the embodiment of the present application, referring to fig. 2, an embodiment of the data processing system based on liquid leak detection in the embodiment of the present application includes:

the filtering module is used for collecting monitoring signals of the photoelectric liquid leakage sensor through the RS485 bus, carrying out signal denoising and outlier filtering on the monitoring signals to obtain leakage basic data comprising sensor ID, position information, detection time stamp and signal strength;

The calculation module is used for calculating the signal intensity change rate and the duration according to the leakage basic data and combining the topological structure of the building water supply and drainage system to obtain leakage characteristic data for marking the leakage type and the influence range;

the comparison module is used for determining a leakage risk level by comparing the leakage characteristic data with a risk threshold value based on the identification leakage type matching risk threshold value;

The extraction module is used for carrying out time, position and type association analysis on the leakage risk level and the historical leakage record, extracting periodic distribution, equipment aging association and use intensity association of leakage occurrence, and forming leakage rule data;

The generation module is used for calculating the failure probability of each pipeline section and equipment according to the leakage rule data, and generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion by combining maintenance cost and resource limiting parameters;

And the integration module is used for constructing a liquid system running state diagram based on the maintenance strategy data, integrating the real-time monitoring data and the maintenance execution condition and generating a management-hierarchy leakage prevention and control data report.

Through the cooperation of the components, the monitoring signal of the photoelectric liquid leakage sensor is acquired through the RS485 bus, and the signal denoising and outlier filtering are carried out, so that high-quality leakage basic data are obtained, the problems of unstable signal and easy interference of the traditional leakage detection system are solved, and the accuracy of leakage detection is greatly improved. And secondly, calculating the change rate and duration of the signal intensity according to the leakage basic data, combining the topological structure of the building water supply and drainage system, realizing the accurate identification of the leakage type and the influence range, and overcoming the limitation that the traditional method can not distinguish the sudden leakage and the slow leakage. By matching the risk threshold value based on the identification leakage type, the leakage characteristic data is compared with the risk threshold value, so that the scientific and reasonable leakage risk level is determined, the situation that the traditional leakage risk assessment is too subjective and rough is changed, and the quantification and standardization of the risk assessment are realized. And carrying out time, position and type association analysis on the leakage risk level and the historical leakage record, extracting periodic distribution, equipment aging association and use strength association of leakage occurrence to form leakage rule data, fully utilizing an artificial intelligent algorithm to carry out deep mining on the historical data, so that the system has predictive capability, and radically changing the passive response mode of the traditional leakage management. And calculating the failure probability of each pipeline section and equipment according to the leakage rule data, combining the maintenance cost and the resource limiting parameters, generating maintenance strategy data comprising inspection frequency, maintenance period and equipment updating suggestion, realizing the optimal configuration of maintenance resources by applying an artificial intelligence algorithm, and maximally reducing the leakage risk under the condition of limited resources. And finally, constructing a liquid system running state diagram based on maintenance strategy data, integrating real-time monitoring data and maintenance execution conditions, generating a management-hierarchy leakage prevention and control data report, meeting the information requirements of different management levels, and supporting management activities of all levels from operation to decision. According to the invention, the artificial intelligent algorithm is deeply fused with the domain knowledge, and the accuracy of leakage detection is remarkably improved through acquisition, processing, analysis and visualization of leakage data.

Referring to fig. 3, in an embodiment of the present invention, there is further provided a computer device, which may be a server, and an internal structure thereof may be as shown in fig. 3. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected by a system bus. Wherein the computer is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used for communicating with an external terminal through a network connection. Which computer program, when being executed by a processor, carries out the above-mentioned method.

It will be appreciated by those skilled in the art that the architecture shown in fig. 3 is merely a block diagram of a portion of the architecture in connection with the present inventive arrangements and is not intended to limit the computer devices to which the present inventive arrangements are applicable.

An embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above method. It is understood that the computer readable storage medium in this embodiment may be a volatile readable storage medium or a nonvolatile readable storage medium.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided by the present invention and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk or an optical disk, etc. which can store the program code.

While the application has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that the foregoing embodiments may be modified or equivalents may be substituted for some of the features thereof, and that the modifications or substitutions do not depart from the spirit and scope of the embodiments of the application.

Claims (10)

1. A data processing method based on liquid leakage detection, characterized in that the data processing method based on liquid leakage detection comprises: The monitoring signal of the photoelectric liquid leakage sensor is collected through the RS485 bus, and the signal is denoised and outliers are filtered to obtain the basic leakage data including sensor ID, location information, detection timestamp and signal strength; Calculate the signal strength change rate and duration based on the basic leakage data, and obtain leakage characteristic data that identifies the leakage type and impact range in combination with the building water supply and drainage system topology; Based on the identified leakage type matching risk threshold, the leakage risk level is determined by comparing the leakage feature data with the risk threshold; Perform correlation analysis on the leakage risk level and historical leakage records in terms of time, location and type, extract the periodic distribution of leakage occurrence, equipment aging correlation and usage intensity correlation, and form leakage pattern data; Calculate the failure probability of each pipeline section and equipment based on the leakage law data, and generate maintenance strategy data including inspection frequency, maintenance cycle and equipment update suggestions in combination with maintenance cost and resource limitation parameters; A liquid system operation status diagram is constructed based on the maintenance strategy data, real-time monitoring data and maintenance execution status are integrated, and a management hierarchical leakage prevention and control data report is generated. 2. The data processing method based on liquid leakage detection according to claim 1 is characterized in that the monitoring signal of the photoelectric liquid leakage sensor is collected through the RS485 bus, and the monitoring signal is subjected to signal denoising and outlier filtering to obtain leakage basic data including sensor ID, location information, detection timestamp and signal strength, including: The photoelectric liquid leakage sensor is divided into a plurality of monitoring areas according to the floor structure of the building, and a monitoring network is formed in each monitoring area through an RS485 bus; Using the Kalman filtering method to remove environmental interference noise from the monitoring signal to obtain a primary filtering signal; Comparing the primary filter signal with the historical state data of the sensor, eliminating abnormal jump values caused by electromagnetic interference, and obtaining valid signal data; The effective signal data is combined with a sensor calibration curve to convert into a standardized liquid leakage signal intensity; The standardized liquid leakage signal strength is combined with the sensor ID, location information and detection timestamp to construct the leakage basic data. 3. The data processing method based on liquid leakage detection according to claim 1 is characterized in that the signal strength change rate and duration are calculated according to the leakage basic data, and the leakage characteristic data identifying the leakage type and the impact range are obtained in combination with the topological structure of the building water supply and drainage system, including: Performing time series differentiation on the signal strength in the leakage basic data, calculating the signal strength change per unit time, and obtaining the signal strength change rate; Performing continuity analysis on the signal strength change rate, counting the length of the time period during which the signal strength continuously changes, and obtaining the leakage duration; Extract the topological structure of the building water supply and drainage system from the building information database, and mark the relative position relationship of each sensor in the water supply and drainage system; Compare the signal strength change rate with a preset change rate threshold, and identify it as a sudden leakage type when the change rate is higher than the threshold, and identify it as a slow leakage type when the change rate is lower than the threshold; Based on the leakage duration and the topological structure of the building water supply and drainage system, a fluid diffusion propagation algorithm is used to calculate the direction and area in which the liquid may flow, so as to form an impact range; The leakage type, the impact range and the original leakage basic data are integrated to construct the leakage feature data identifying the leakage type and the impact range. 4. The data processing method based on liquid leakage detection according to claim 1 is characterized in that the step of matching the risk threshold based on the identified leakage type and determining the leakage risk level by comparing the leakage feature data with the risk threshold comprises: Extracting a risk threshold parameter set corresponding to the identified leakage type from a leakage risk database, wherein the risk threshold parameter set includes critical values of liquid leakage amounts of different area types; Calculate the estimated value of leakage according to the influence range in the leakage characteristic data, and obtain the importance coefficient of the leakage area from the building information system; The leakage estimation value is compared with the liquid leakage critical value in the risk threshold parameter set, and divided into a primary leakage value, a secondary leakage value and a high leakage value; Conduct a value assessment of the facilities surrounding the leakage area and calculate the regional sensitivity index based on the importance of the facilities and the degree of possible losses; The regional sensitivity index is weightedly combined with the primary leakage value, the intermediate leakage value and the advanced leakage value to form a risk score; The leakage risk level is determined as a slight risk, a moderate risk, or a severe risk according to the score interval in which the risk score lies. 5. The data processing method based on liquid leakage detection according to claim 1 is characterized in that the correlation analysis of the leakage risk level with the historical leakage records in terms of time, location and type is performed to extract the periodic distribution of leakage occurrence, the correlation of equipment aging and the correlation of usage intensity to form leakage regularity data, including: Retrieve historical leakage records from the leakage event database, perform multi-dimensional matching with the current leakage risk level, and construct a leakage event time series table containing time and space tags; Applying Fourier time-frequency transform to the leakage event time series table to separate the leakage distribution characteristics of long period, medium period and short period, and extracting the periodic distribution by peak detection algorithm; According to the time difference between the equipment installation time and the leakage occurrence time in the historical leakage records, combined with the equipment material parameters and the operating environment factors, a nonlinear regression equation is constructed to obtain the equipment aging correlation; Collecting crowd density data and equipment operation frequency data in different areas of the building, determining the coupling coefficient between the usage intensity of each area and the leakage event by cross entropy calculation method, and obtaining the usage intensity correlation; The periodic distribution, the equipment aging correlation and the usage intensity correlation are multi-level integrated through an adaptive weight network to form a leakage risk quantification index with spatiotemporal prediction capability; The leakage risk quantification index is combined with the hydraulic characteristics of the geographical location and the failure mode of the equipment type to construct the leakage regularity data with prediction function. 6. The data processing method based on liquid leakage detection according to claim 5 is characterized in that the failure probability of each pipeline segment and equipment is calculated according to the leakage law data, and the maintenance strategy data including inspection frequency, maintenance cycle and equipment update suggestion is generated in combination with maintenance cost and resource limitation parameters, including: The periodic distribution data in the leakage pattern data are sorted according to time nodes, the number of leakage events in each time period is counted, multiplied by the leakage severity weight, and divided by the time period length to obtain the leakage risk value per unit time, thereby calculating the leakage risk coefficient of each pipeline section; Extract the initial failure rate of the equipment according to the equipment aging correlation, multiply it by the square root of the equipment's service time, and add a linear combination of the environmental corrosion factor and the service pressure coefficient to calculate the failure probability of each pipeline section and equipment; The original price, installation time, and replacement material cost of each piece of equipment are extracted from the facility management system and multiplied by the inflation factor, plus the estimated economic loss during downtime to construct a maintenance cost matrix; Based on the enterprise resource management system, the total number of available working hours per month, the number of inventory equipment and the budget limit are extracted and set as the upper limit of human resources, the upper limit of material resources and the upper limit of cost resources respectively to form a resource restriction parameter set; The failure probability is multiplied by the fault loss assessment value, and divided by the single inspection and maintenance cost to obtain a cost-effectiveness ratio. When the ratio is greater than the resource allocation threshold, the inspection frequency and maintenance cycle are set according to three levels: high, medium and low. The failure probability of the equipment is multiplied by the cost-effectiveness ratio, and the equipment update difficulty coefficient is subtracted to obtain an update priority score. Equipment update suggestions are generated by sorting the scores to finally form the maintenance strategy data. 7. The data processing method based on liquid leakage detection according to claim 6 is characterized in that the liquid system operation status diagram is constructed based on the maintenance strategy data, the real-time monitoring data and maintenance execution status are integrated, and a management hierarchical leakage prevention and control data report is generated, including: Based on the inspection frequency, maintenance cycle and equipment update suggestions in the maintenance strategy data, combined with the building spatial structure information, a liquid system topology framework reflecting pipeline distribution and equipment location is constructed; The liquid system topology framework is superimposed on the failure probability, and divided into high-risk area, medium-risk area and low-risk area according to the risk degree, to generate a liquid system operation status diagram; Acquire the latest basic leakage data from the central monitoring system and merge it with the liquid system operation status diagram to form a real-time monitoring situation diagram; Extract maintenance execution records from the work order management system, including completion time, maintenance content and processing results, and calculate maintenance task completion rate and problem solving efficiency indicators; According to different levels of management personnel, the real-time monitoring situation map and the maintenance execution record are subjected to hierarchical data extraction to generate facility operation layer reports, management decision layer reports and strategic planning layer reports; The facility operation level report, the management decision level report and the strategic planning level report are integrated into a unified leakage prevention and control data report, and archived and stored in time series. 8. A data processing system based on liquid leakage detection, used to implement the data processing method based on liquid leakage detection according to any one of claims 1 to 7, characterized in that the data processing system based on liquid leakage detection comprises: The filtering module is used to collect the monitoring signal of the photoelectric liquid leakage sensor through the RS485 bus, perform signal denoising and outlier filtering on the monitoring signal, and obtain the basic leakage data including the sensor ID, location information, detection timestamp and signal strength; A calculation module, used to calculate the signal strength change rate and duration according to the basic leakage data, and obtain leakage characteristic data identifying the leakage type and impact range in combination with the topological structure of the building water supply and drainage system; A comparison module, used to match the risk threshold based on the identified leakage type, and determine the leakage risk level by comparing the leakage feature data with the risk threshold; An extraction module is used to analyze the correlation of the leakage risk level with the historical leakage records in terms of time, location and type, extract the periodic distribution of leakage occurrence, equipment aging correlation and usage intensity correlation, and form leakage regularity data; A generation module is used to calculate the failure probability of each pipeline segment and equipment according to the leakage law data, and generate maintenance strategy data including inspection frequency, maintenance cycle and equipment update suggestions in combination with maintenance cost and resource limitation parameters; The integration module is used to construct a liquid system operation status diagram based on the maintenance strategy data, integrate real-time monitoring data and maintenance execution status, and generate a management hierarchical leakage prevention and control data report. 9. A computer device, characterized in that it comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and wherein the processor implements the data processing method based on liquid leakage detection according to any one of claims 1 to 7 when executing the computer program. 10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is enabled to execute the data processing method based on liquid leakage detection according to any one of claims 1 to 7.