CN113589797A - Intelligent diagnosis method and system for coke oven vehicle operation fault - Google Patents

Abstract

The present invention provides a method and system for intelligent diagnosis of coke oven vehicle running faults. The method includes building a coke oven vehicle fault diagnosis knowledge model; collecting monitored equipment status information; and extracting features from the collected equipment status information. Then, using the coke oven vehicle fault diagnosis knowledge model, select some observation variables to form a symptom set; identify the abnormal state of the working condition on the symptom set, and then calculate the probability distribution of specific fault types, and the system includes a coke oven Vehicle fault diagnosis knowledge model, coke oven vehicle fault diagnosis data acquisition module and coke oven vehicle fault diagnosis inference module. The invention realizes the fault diagnosis of the four major car operations during coal charging, coke guiding and coke quenching, and establishes a reliable, safe, stable and efficient IFDC safety production monitoring system for the four major coke oven cars; realizes fault information in the production area. Real-time collection, processing, dissemination and control and early warning.

Description

Method and system for intelligent diagnosis of operating faults of coke oven vehicles

technical field

The invention relates to the technical field of coke oven vehicle operation and maintenance, in particular to a method and system for intelligent diagnosis of coke oven vehicle operation faults.

Background technique

Intelligent fault diagnosis and equipment maintenance are complex dynamic processes involving multiple disciplines and departments. With the support of multiple technologies such as condition monitoring, knowledge modeling, diagnostic reasoning, and auxiliary decision-making, equipment diagnosis and maintenance work is carried out collaboratively according to equipment operating status.

At present, the fault diagnosis of coke oven vehicle operation still relies on manual detection and judgment, so there is no real-time collection, processing, dissemination and control and early warning of new faults, and the manual judgment method is also reliable, safe and stable. , high efficiency can not meet the needs of current production, so there is a need for an intelligent diagnosis method and system for coke oven vehicle operation faults.

SUMMARY OF THE INVENTION

In view of the defects in the prior art, the purpose of the present invention is to provide a method and system for intelligent diagnosis of operating faults of a coke oven vehicle.

In the first aspect, the present invention provides an intelligent diagnosis method for operating faults of a coke oven vehicle, comprising the following steps:

Build a fault diagnosis knowledge model for coke oven vehicles;

Collect the monitored equipment status information;

After feature extraction is performed on the collected equipment state information, using the coke oven vehicle fault diagnosis knowledge model, some observed variables are selected to form a symptom set;

The abnormal state of the working condition is identified on the symptom set, and then the probability distribution of specific failure types is calculated.

Optionally, also include:

Combined with the equipment status information and the decision tool for optimizing resource utilization, an optimized intelligent diagnosis and maintenance group decision-making scheme is given.

Optionally, the calculating the probability distribution of specific failure types further includes:

Using the collected equipment state information, and then mapping the equipment state working condition space to the fault feature space based on the deep belief network model, and analyzing whether the state is a fault state;

In the case that the state is a fault state, the failure probability analysis is performed using the maximum possible explanation method, that is, the hypothesis with the largest posterior probability among all possible hypotheses is found according to the existing data.

Optionally, the knowledge model of coke oven vehicle fault diagnosis is constructed with knowledge processing as the core, and the knowledge content is evolved in terms of collection, modeling, reasoning, decision-making, etc. The complete processing flow of decision-making, the construction flow of the knowledge model for coke oven vehicle fault diagnosis further includes:

determine the boundaries of the area;

Establish and maintain a conceptual model of knowledge;

Entity verification of conceptual associations and constraints;

Maintenance process status and symptoms are mapped;

The fault and the symptom are matched.

Optionally, the equipment status information, maintenance process status and symptom mapping, fault and symptom matching data are all used for optimizing and updating the coke oven vehicle fault diagnosis knowledge model.

Further, the present invention also provides an intelligent diagnosis system for operating faults of a coke oven vehicle, which is used to implement the above-mentioned method, including:

Coke oven vehicle fault diagnosis knowledge model: used to provide data and knowledge content for coke oven vehicle fault diagnosis inference module and coke oven vehicle fault diagnosis decision support module;

Coke oven vehicle fault diagnosis data collection module: used to collect the monitored equipment status information;

Coke oven vehicle fault diagnosis and reasoning module: using the coke oven vehicle fault diagnosis knowledge model, select some observed variables to form a symptom set, identify abnormal working conditions on the symptom set, and then calculate the probability distribution of specific fault types. .

optionally, also includes,

Coke oven vehicle fault diagnosis decision support module: using the coke oven vehicle fault diagnosis knowledge model, combined with the equipment status information and the decision tool for optimizing resource utilization, to provide an optimized intelligent diagnosis and maintenance group decision-making scheme.

Optionally, the device status information is provided by four major vehicle fault monitoring devices.

Optionally, the offline inspection system of the four major vehicle fault monitoring equipment includes:

Patrol instrument: used to collect equipment status information;

Real-time display of inspection data: used for real-time display of equipment status information data collected by the inspection instrument;

System server, management workstation and mobile workstation: receive the equipment status information data sent by the real-time display screen of the inspection data through the communication network.

Optionally, the online monitoring system of the four major vehicle fault monitoring equipment includes:

Furnace state monitoring room: used for monitoring furnace state information;

Central control state monitoring room: used for monitoring the central control state information of fault monitoring equipment;

Information center: used for the collection of monitoring data from the furnace state monitoring room;

On-site control room: collect monitoring data sent by the central control state monitoring room and the information center through the communication network.

Compared with the prior art, the present invention has the following beneficial effects:

The invention discloses a complete set of intelligent fault diagnosis method and system for the four major vehicles of coke oven production, realizes fault diagnosis of the four major vehicles during coal charging and coke pushing, coke guiding and coke quenching, and establishes a reliable, safe, stable and efficient operation. The IFDC safety production monitoring system for the four major coke ovens; realize the real-time collection, processing, dissemination and control and early warning of fault information in the production area; realize the safety production and comprehensive management of coke ovens such as personnel and vehicles, and can quickly And effectively find out the cause of the failure, so as to predict the remaining service life of the device or firmware.

Description of drawings

Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings:

1 is a schematic structural diagram of an off-line inspection system of four major vehicle fault monitoring devices provided in an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of an online monitoring system of four major vehicle fault monitoring devices provided by an embodiment of the present invention;

3 is a flowchart of modeling a knowledge model for coke oven vehicle fault diagnosis provided by an embodiment of the present invention;

4 is a schematic structural diagram of an abnormal symptom of an abnormal increase in the fundamental frequency amplitude of a coke oven vehicle transmission provided by an embodiment of the present invention;

5 is a schematic structural diagram of a system for diagnosing operating faults of a coke oven vehicle provided by an embodiment of the present invention;

FIG. 6 is a flowchart of a method for diagnosing operating faults of a coke oven vehicle according to an embodiment of the present invention.

Detailed ways

The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the inventive concept. These all belong to the protection scope of the present invention.

1 is a schematic structural diagram of an off-line inspection system of the four major vehicle fault monitoring equipment provided by an embodiment of the present invention, FIG. 2 is a schematic structural diagram of an online monitoring system of the four major vehicle fault monitoring equipment provided by an embodiment of the present invention; FIG. 3 is the present invention Fig. 4 is an effect diagram of the judgment of abnormal signs of coal vehicle transmission; Fig. 5 is a schematic structural diagram of a fault diagnosis system for coke oven vehicle operation provided by an embodiment of the present invention; Fig. 6 A flowchart of a method for diagnosing operating faults of a coke oven vehicle provided by an embodiment of the present invention; referring to FIGS. 1-6 , the intelligent diagnosis system for operating faults of a coke oven vehicle in this embodiment includes:

Coke oven vehicle fault diagnosis knowledge model 1: Provide data and knowledge content for coke oven vehicle fault diagnosis reasoning module 3 and coke oven vehicle fault diagnosis decision support module 4;

Coke oven vehicle fault diagnosis data collection module 2: collect equipment status information;

In this implementation, the equipment status information is the data monitored by the four major vehicle fault monitoring equipments 5 .

Coke oven vehicle fault diagnosis and reasoning module 3: Using coke oven vehicle fault diagnosis knowledge model 1, select some observed variables to form a symptom set, identify the abnormal state of the symptom set, and then calculate the probability distribution of specific fault types.

In this embodiment, the identification of the abnormal state of the working condition is realized by the Bayes classifier, and then the probability distribution when a certain fault occurs is calculated by probabilistic reasoning in the most probable explanation MPE (Most Probable Explanation, MPE) method.

Coke oven vehicle fault diagnosis decision support module 4: Using coke oven vehicle fault diagnosis knowledge model 1, combined with equipment status information and decision tools for optimizing resource utilization, an optimized intelligent diagnosis and maintenance group decision-making scheme is given.

In this embodiment, the decision tool for optimizing resource utilization is provided by the coke oven vehicle fault diagnosis and reasoning module 3 .

Among them, the intelligent diagnosis system for coke oven vehicle operation faults consists of a coke oven vehicle fault diagnosis data acquisition module 2 responsible for on-site detection, a coke oven vehicle fault diagnosis knowledge model 1 responsible for remote monitoring and diagnosis, a coke oven vehicle fault diagnosis inference module 3 and a coke oven vehicle fault diagnosis inference module 3. The furnace vehicle fault diagnosis and decision support module is composed of 4. The developed system fault diagnosis and maintenance guidance software is based on the PLC monitoring of the four major vehicles and the information fed back by the intelligent sensors. The fault information of process, mechanical and electrical information such as running, positioning, etc., establishes the relevant mechanism and intelligent model and analysis algorithm, and realizes the functions of unified monitoring, control and information exchange.

Among them, the four major vehicle fault monitoring devices 5 described above are composed of inspection instruments, sensors, management software, and background databases, etc. The structure of the offline inspection system is shown in Figure 1. include:

Inspection instrument: collect equipment status information;

Real-time display of inspection data: Real-time display of equipment status information data collected by the inspection instrument;

System servers, management workstations, mobile workstations and other information systems: receive equipment status information data sent by the real-time display of inspection data through the intranet;

The online monitoring system structure of the four major vehicle fault monitoring equipment is shown in Figure 2, including:

Furnace state monitoring room: used for monitoring the furnace state;

Central control state monitoring room: used for monitoring the central control state of fault monitoring equipment;

Information center: used for the collection of monitoring data from the furnace state monitoring room;

On-site control room: collect the monitoring data sent by the central control state monitoring room and the information center through the enterprise LAN.

The four major vehicle fault monitoring devices described above integrate a variety of condition monitoring sensors and monitoring instruments, and have a unified human-computer interaction interface and a general database structure. The coke oven vehicle fault diagnosis data acquisition module 2 also provides a variety of industry standard interfaces. (such as RS-485, Modbus, Ethernet, etc.) to interact with the company's existing external control and automation systems.

The online monitoring system of the above-mentioned four major vehicle fault monitoring equipment can set the monitoring interface of the application layer app in the on-site control room, and the maintenance personnel can monitor the diagnostic information of each vehicle in the on-site control room, and share the diagnostic information with each vehicle. The layer software completes the matching of the information and the meaning of the information, and displays it on the operation interface of each vehicle and the monitoring interface of the on-site control room. According to the inspection system and the information situation, the maintenance personnel perform manual intervention on board, troubleshoot or dispatch the action of replacing the vehicle. , to realize the centralized control function based on the on-site control room.

As shown in FIG. 6 , this embodiment also provides an intelligent diagnosis method for operating faults of a coke oven vehicle. The method is implemented by using the above system and includes the following steps:

S1, construct a fault diagnosis knowledge model for coke oven vehicles;

S2, collect the monitored equipment status information;

S3, after feature extraction is performed on the collected equipment status information, using the coke oven vehicle fault diagnosis knowledge model 1, some observed variables are selected to form a symptom set;

S4, identify the abnormal state of the working condition on the symptom set, and then calculate the probability distribution of specific failure types.

In this implementation, the equipment status information is the information monitored by the four major vehicle fault monitoring equipments 5 .

In this embodiment, a fault diagnosis knowledge model 1 for coke oven vehicles is constructed, and an intelligent diagnosis system for vehicle running faults is constructed based on this model. When the model is constructed, knowledge processing is the core, and knowledge is evolved in the aspects of acquisition, modeling, reasoning, and decision-making. The content, combined with the state semantics in the extraction and diagnosis process, completes the complete processing flow of decision-making from data information and knowledge, and builds a reliable and complete maintenance and diagnosis knowledge modeling model. The basic flow is shown in Figure 3, which specifically includes:

determine the boundaries of the area;

Establish and maintain a conceptual model of knowledge;

Entity verification of conceptual associations and constraints;

The maintenance process status and symptoms are mapped, and the generated data is also used to update the knowledge model 1 of coke oven vehicle fault diagnosis;

The faults and symptoms are matched, and the generated data is used to update the knowledge model 1 of the fault diagnosis of coke oven vehicles at the same time.

In this embodiment, step S4 specifically includes selecting part of the observed variables to form a symptom set, using the Bayes classifier to identify abnormal working conditions, and then using the Most Probable Explanation (MPE) method to calculate a certain probability through probabilistic reasoning. The probability distribution when the fault occurs, and the specific calculation is shown in the following abnormal working condition state identification method and fault diagnosis probability inference algorithm:

Among them, the S41 abnormal working condition state identification method includes:

In some cases, the abnormal working condition state identification method can directly determine the occurrence of the fault from the working condition of the equipment, while most of the working conditions require the use of various data acquisition hardware to obtain operating data, and then perform data analysis to determine whether the state is It is a fault symptom; use data acquisition hardware to obtain operating data, and then map the equipment state and working condition space to the fault feature space based on the Onto DBN model (deep belief network model), and analyze whether the state is a fault state. The analysis process is as follows:

When estimating the probability of the occurrence of the working condition state (that is, the prior probability) according to the prior knowledge, the state space of the equipment working condition is Ω _i =(ω ₁ ... ω _i ... ω _n ), ω _i (i=1,2, ...n) represents a mode point in the state space;

The normal and abnormal operating conditions can be represented by P(ω ₁ ) and P(ω ₂ ), respectively, and

P(ω ₁ )+P(ω ₂ )=1

Assuming that x is a discrete random variable representing the state of the working condition, combined with the probability distribution function P(xω _i ) of x when the working condition state is ω _i and the Bayes formula, we can get:

Let {α ₁ , α _{, 2} ..., α _k } represent a limited set of k possible judgment behaviors, and the risk function λ(α _i ω _j ) represents the risk of judging behavior α _i when the working condition is ω _j j;

Conditional risk is defined as:

According to the Bayes decision rule (Bayes decision rule), the problem of abnormal state identification is to select the appropriate abnormal state judgment behavior α _i to minimize the conditional risk value, that is, α ^* = arg minR(α _i x), where α ^* The risk of abnormal state judgment behavior when it is the minimum risk value;

_Suppose the feature vector ^X =(x ₁ ,x ₂ .

in,

p _i and q _i are the conditional probability values when the equipment is in the normal state ω ₁ and the abnormal state ω ₂ respectively ( _xi = 1);

Among them, ψ _i is the intermediate parameter calculated based on the conditional probabilities p and q in the normal state; ψ ₀ is the intermediate parameter calculated based on the conditional probability p and q in the abnormal state, and the state eigenvector of xi _i , The subscript _i of xi indicates different working conditions.

After the above-mentioned abnormal working condition state identification method determines that an abnormal state occurs, the following will infer the probability of each type of abnormal state occurrence: The specific steps are:

S42, the fault diagnosis probability inference algorithm:

The entity relationships in the fault problem domain are organized into a graph structure to describe the possible dependencies between entities in the problem domain. The conditional probability represents the uncertainty of the problem domain, and the conditional probability distribution indicates the reliability value of the strength of the causal relationship. In this way, the qualitative part of the Bayesian network describes the dependence of the problem domain, and the quantitative part describes the dependence. The reliability value of the relationship; the failure probability analysis is carried out by using the MPE reasoning method, that is, the hypothesis with the largest posterior probability among all the possible hypotheses is found according to the existing data, as shown in Figure 4, and the calculation steps are as follows:

Among them: P(H|C,S) represents the probability of failure in the fault hypothesis subset H under the condition of given equipment operating state C and fault symptom S.

According to the judgment method of fault symptoms (that is, the identification of abnormal working conditions), it can be known from Bayes' theorem that:

P(HC,S)∞P(C,SH)×P(H)

then there is,

In the above formula, P(H) is the probability of fault occurrence in the fault hypothesis subset H, and P(C|H) and P(S|H) are the conditional probabilities of the working condition and fault symptoms in H when the fault occurs.

If the prior failure probability of a certain failure f is P(f), and f∈{0,1}, then:

表示当故障假设H成立时，状态c与H中故障f无关的概率。in,

represents the probability that the state c is independent of the fault f in H when the fault hypothesis H holds.

Among them, P(c) refers to the failure probability of state c, P(cH) is the failure probability of state c independent of H; P(cf) is the failure probability of state c independent of failure f in H.

When the fault symptom set is empty, it is considered that all possible faults have been added to the subset of fault hypotheses, and the inference loop is exited and the maximum possible fault explanation is obtained.

The method and system of the present application are described above, and the overall principles and beneficial effects of the present application are described below:

Among them, in the coke oven vehicle fault diagnosis and reasoning module 3, the present invention proposes an intelligent fault diagnosis model IFDM (Intelligent Fault Diagnosis Modeling), that is, the above-mentioned intelligent diagnosis system for coke oven vehicle operation faults, which can detect potential faults or faults of key equipment or key components. Accurately estimate the failure start time, quickly and effectively find out the cause of the failure when the failure or abnormality occurs, so as to predict the remaining service life (Remaining Useful Life, RUL) of the equipment or firmware, and take further maintenance measures according to the given failure prediction. , diagnostic models based on condition monitoring data and probabilistic reasoning provide decision-making tools to optimize resource utilization.

Among them, the coke oven vehicle fault diagnosis decision support module 4, under the support of multiple technologies such as condition monitoring, knowledge modeling and auxiliary decision-making, conducts equipment diagnosis work in coordination with the equipment operating status, and formulates reasonable and effective equipment maintenance strategies according to the diagnosis results. Through the knowledge representation, diagnostic reasoning, decision support and other links in the production process, to ensure the effective implementation of diagnostic maintenance work, thereby improving the efficiency and economic benefits of equipment maintenance.

Among them, the intelligent diagnosis and maintenance platform uses the coke oven vehicle fault diagnosis knowledge model 1, in the condition monitoring system (that is, the monitoring system of the four major vehicle fault monitoring equipment 5), the equipment inspection subsystem, the coke oven vehicle fault diagnosis inference module 3 and the coke oven vehicle fault diagnosis and reasoning module 3. With the support of application systems such as the fault diagnosis and decision support module of the coke oven vehicle, the equipment diagnosis and maintenance work is carried out collaboratively according to the operation status of the equipment. The intelligent fault diagnosis system for the operation of the coke oven vehicle includes an on-site monitoring module and a remote monitoring and diagnosis module. The operation logic of each module of the fault intelligent diagnosis system is shown in Figure 5.

The coke oven vehicle fault diagnosis data collection module 2 collects the equipment status information monitored by the four major vehicle fault monitoring equipment. In this embodiment, the status information includes status parameters and process variables, and the collected equipment status information is also used for coke oven vehicle faults. Optimization and update of Diagnostic Knowledge Model 1.

In this embodiment, in the process of CDQ production, the cable communication speed between the machine workshop and the upper computer is slow (using the enterprise local area network and the enterprise intranet will speed up the communication speed) and the lack of intelligent safety monitoring methods and facilities, a coke This paper proposes an intelligent fault diagnosis (IFDM) model based on the knowledge of four-vehicle operation, which is based on the intelligent fault diagnosis (IFDM) model based on the knowledge of four-vehicle operation. The intelligent diagnosis system for furnace vehicle operation faults integrates and infers the fault diagnosis and maintenance knowledge of four vehicles, and forms an intelligent maintenance mode centered on the diagnosis and maintenance process; establishes a prototype of the equipment condition monitoring system based on wireless sensor network, and integrates the signal analysis of the embedded processor Ability and local data, realize distributed state detection, and form a state maintenance sensor network with self-analysis and diagnosis ability; introduce the ontology semantic expression method of diagnosis and maintenance knowledge, establish an ontology-driven fault diagnosis and reasoning model, and complete static maintenance knowledge and dynamic diagnosis The unification of the process; proposes an ontology semantics to express the fault probability inference framework, constructs an ontology-based fault diagnosis network, and gives a fault probability inference algorithm based on the most probable explanation (MPE), according to the operating conditions, fault symptoms and Evidence information reasoning is used to obtain fault diagnosis explanation. Aiming at the uncertainty existing in the process of fault diagnosis and maintenance decision-making, a group decision-making method for equipment maintenance is proposed. Based on the integration and modeling of multi-source heterogeneous manufacturing process knowledge, diagnostic reasoning is carried out. Based on the analysis of the cause of the failure, combined with the experience and knowledge of the diagnostic experts, an optimized maintenance decision-making plan is given. The prediction results show that the method proposed by the present invention can realize the fault diagnosis and maintenance guidance of the four major vehicle operations during coal charging and coke pushing, coke guiding and coke quenching. , establish a reliable, safe, stable and efficient IFDC safety production monitoring system for the four major coke ovens; realize the real-time collection, processing, dissemination, control and early warning of fault information in the production area; realize the safety production and comprehensive management of coke ovens such as personnel and vehicles , the present invention has feasibility and superiority.

The above description describes the diagnosis method, system and principle. The following Figure 4 further illustrates the judgment of abnormal symptoms of coal truck transmission. When the new state data E (the vibration amplitude value is 8.9 μm) is newly monitored, according to the characteristics of the fault symptoms The decision function E should be regarded as an abnormal symptom of the fundamental frequency amplitude growth rate.

Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above-mentioned specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essential content of the present invention. The embodiments of the present application and features in the embodiments may be combined with each other arbitrarily, provided that there is no conflict.

Claims (10)

1. a method for intelligent diagnosis of coke oven vehicle operation failure, it is characterized in that comprising the steps: Build a fault diagnosis knowledge model for coke oven vehicles; Collect the monitored equipment status information; After feature extraction is performed on the collected equipment state information, using the coke oven vehicle fault diagnosis knowledge model, some observed variables are selected to form a symptom set; The abnormal state of the working condition is identified on the symptom set, and then the probability distribution of specific failure types is calculated. 2. The intelligent method for diagnosing operating faults of a coke oven vehicle according to claim 1, further comprising: Combined with the equipment status information and the decision tool for optimizing resource utilization, an optimized intelligent diagnosis and maintenance group decision-making scheme is given. 3. The intelligent method for diagnosing operating faults of a coke oven vehicle according to claim 1, wherein the calculating the probability distribution of specific fault types further comprises: Using the collected equipment state information, and then mapping the equipment state working condition space to the fault feature space based on the deep belief network model, and analyzing whether the state is a fault state; In the case that the state is a fault state, the failure probability analysis is performed using the maximum possible explanation method, that is, the hypothesis with the largest posterior probability among all possible hypotheses is found according to the existing data. 4. The method for intelligent diagnosis of operating faults of coke oven vehicles according to claim 1, characterized in that, when the knowledge model for fault diagnosis of coke oven vehicles is constructed, knowledge processing is taken as the core, and in the process of collection, modeling, reasoning, decision-making, etc. Aspects of evolving knowledge content, combined with state semantics in the extraction and diagnosis process, complete the complete processing flow of decision-making from data information knowledge, the coke oven vehicle fault diagnosis knowledge model construction process further includes: determine the boundaries of the area; Establish and maintain a conceptual model of knowledge; Entity verification of conceptual associations and constraints; Maintenance process status and symptoms are mapped; The fault and the symptom are matched. 5 . The method for intelligently diagnosing operating faults of a coke oven vehicle according to claim 4 , wherein the equipment status information, maintenance process status and symptoms are mapped, and the data for matching faults with the symptoms are used for the Optimization update of the knowledge model for fault diagnosis of coke oven vehicles. 6. An intelligent diagnosis system for coke oven vehicle operation faults, for implementing the method according to any one of claims 1-5, characterized in that, comprising: Coke oven vehicle fault diagnosis knowledge model: used to provide data and knowledge content for coke oven vehicle fault diagnosis inference module and coke oven vehicle fault diagnosis decision support module; Coke oven vehicle fault diagnosis data collection module: used to collect the monitored equipment status information; Coke oven vehicle fault diagnosis and reasoning module: using the coke oven vehicle fault diagnosis knowledge model, select some observation variables to form a symptom set, identify abnormal working conditions on the symptom set, and then calculate the probability distribution of specific failure types . 7. The intelligent diagnosis system for coke oven vehicle operation faults according to claim 6, characterized in that, further comprising: Coke oven vehicle fault diagnosis decision support module: using the coke oven vehicle fault diagnosis knowledge model, combined with the equipment status information and the decision tool for optimizing resource utilization, to provide an optimized intelligent diagnosis and maintenance group decision-making scheme. 8 . The intelligent diagnosis system for coke oven vehicle running faults according to claim 6 , wherein the equipment status information is provided by four major vehicle fault monitoring equipments. 9 . 9. The intelligent diagnosis system for coke oven vehicle operation faults according to claim 8, wherein the offline inspection system of the four major vehicle fault monitoring equipment comprises: Patrol instrument: used to collect equipment status information; Real-time display of inspection data: used for real-time display of equipment status information data collected by the inspection instrument; System server, management workstation and mobile workstation: receive the equipment status information data sent by the real-time display screen of the inspection data through the communication network. 10. The intelligent diagnosis system for coke oven vehicle operation faults according to claim 8, wherein the online monitoring system of the four major vehicle fault monitoring equipment comprises: Furnace state monitoring room: used for monitoring furnace state information; Central control state monitoring room: used for monitoring the central control state information of fault monitoring equipment; Information center: used for the collection of monitoring data from the furnace state monitoring room; On-site control room: collect monitoring data sent by the central control state monitoring room and the information center through the communication network.

Images