KR102475374B1 - Apparatus and method for active failure predictive diagnosis of solar power generation system in string units using idle time - Google Patents

Abstract

An apparatus for active failure prediction and diagnosis of a solar power generation system in string units of the present invention comprises: a communication interface unit which provides a communication interface with a server of a meteorological agency and receives future weather information for a preset period in the future from the server of the meteorological agency; an operation data collection unit which, when a solar power generation system operates in a daytime operation mode, collects and stores power parameters generated in string units in real-time operation conditions and environmental parameters including operation environment information of the solar power generation system; a future driving state prediction unit which, when an operation mode of the solar power generation system is switched to a night rest mode due to a decrease in solar radiation, predicts a future driving state of the solar power generation system while generating at least one next-day operation prediction scenario by using the power parameters and environmental parameters for each string collected by the operation data collection unit and the future weather information; and a diagnostic unit which predicts and diagnoses presence of a failure of the solar power generation system by analyzing the at least one next-day operation prediction scenario in a state where the operation mode of the solar power generation system is the night rest mode. Therefore, the solar power generation system can be operated more efficiently, and system errors for various situations can be predicted and prepared in advance.

Description

String unit solar power generation system active fault prediction diagnosis device and method using idle time

The present invention relates to a photovoltaic power generation system, and more particularly, to a string-based active fault prediction diagnosis device and method using a solar power generation system idle time period.

A photovoltaic power generation system is a power generation system that converts energy received from the sun into electrical energy, and its value is increasing as environmental demands for reduction of fossil energy and non-pollution increase.

In addition, the importance of operation and maintenance (O&M) is gradually being emphasized for long-term development of photovoltaic power generation systems and stable profit generation. In the case of a photovoltaic power plant, it is a long-term investment project that takes 20 to 25 years to install once, but it cannot be guaranteed that high profits will be guaranteed continuously in the operation of the power plant due to the high initial cost investment.

Photovoltaic operation and maintenance (O&M) is an essential element for maintaining stable power supply in power plant operation, and includes all tasks ranging from simple facility maintenance and repair in a narrow sense to life extension and performance improvement in a broad sense. can do.

Therefore, the purpose of O&M is to identify in advance the factors that hinder the optimal power generation of the photovoltaic power generation system or to minimize the loss of power generation through quick response in the event of a failure.

However, the factors that hinder the amount of power generation are very diverse. For example, discoloration, cracking, and hot spots of solar cells, natural degradation of solar cells, short lifespan of inverters, and myriad natural environments such as snow, wind, and dust are examples. . In addition, an increase in system failures and system aging due to an increase in the cumulative installed amount of photovoltaic facilities are also one of the factors that hinder the amount of power generation.

As such, numerous factors appear without notice in various facilities, so steady maintenance is essential in the operation of a photovoltaic power generation system. However, it is very difficult to visually check all of these various factors.

In order to compensate for this problem, Korea Patent Registration No. 10-1728692 discloses a failure prediction monitoring system and method of a photovoltaic module. According to the above patent, it is possible to accurately determine whether or not the solar power module is out of order by determining whether the amount of solar power generation, which changes every moment, changes according to the appropriate theoretical amount of power generation according to the season or real-time weather change. In addition, according to the above patent, if a failure or abnormality is predicted, it is transmitted to the user terminal in real time.

However, the conventional technologies as described above are technologies for determining the operating state of the solar power generation system while in operation, and cannot utilize the operation characteristics of the solar power generation system that repeats operation and pause, resulting in low efficiency. there was.

In addition, conventionally, solar power generation according to various situations that may occur in the future (eg, sudden climate change, etc.) and secular change by predicting failure by simple comparison of theoretical power generation and real-time power generation Since the operating state of the photovoltaic power generation system reflecting the energy efficiency of the system cannot be predicted, there is a problem in which the accuracy of failure prediction and soundness judgment of the photovoltaic power generation system is lowered.

In addition, in the related art, since the operation state of the photovoltaic power generation system cannot be predicted for weather conditions that deviate from weather forecast information, there is a problem in that unexpected and sudden climate change cannot be coped with.

Korean Patent Registration No. 10-1728692

Therefore, the present invention reflects the operation characteristics of the photovoltaic power generation system that repeats operation and shutdown due to changes in solar radiation, and when the photovoltaic power generation system is operating in the night sleep mode, the previous daytime operation mode operation state of the photovoltaic power generation system By analyzing, it is intended to provide an active fault predictive diagnosis device and method for a solar power generation system in a string unit using the idle time that can operate the solar power system more efficiently.

In addition, the present invention analyzes the operating state of a string unit using real-time operating data and meteorological information for each string collected during the immediately preceding daytime operating mode in a state where the photovoltaic power generation system is in a night rest mode, thereby analyzing the corresponding solar power generation system. It is intended to provide an active fault predictive diagnosis device and method for a solar power generation system in a string unit using an idle time period capable of obtaining accurate judgment results on the operating state of the photovoltaic power generation system.

In addition, the present invention predicts the future driving state for each of at least one situation that can occur during the operation of the solar power generation system in a state where the solar power generation system is in a night rest mode, real-time operation data for each string, and future weather After generating at least one next-day driving prediction scenario using the information, failure prediction is performed based on the next-day driving prediction scenario to prevent various situations, including various unexpected situations (eg, system failure, sudden weather change, etc.) It is intended to provide an active failure predictive diagnosis device and method for a solar power generation system in a string unit using a down time period that can predict system errors in advance and prepare for them.

In addition, the present invention can safely and accurately measure the reference data by measuring reference data capable of analyzing the operating state of the photovoltaic power generation system during operation in a state where the photovoltaic power generation system is in a night rest mode, Therefore, it is intended to provide an active fault predictive diagnosis device and method for a solar power generation system in a string unit using an idle time zone that can accurately determine the operating state of the solar power generation system during operation.

In order to achieve the above object, the active failure predictive diagnosis device for a solar power generation system in a string unit provided by the present invention is an active failure predictive diagnosis of a solar power generation system operating in either a daytime operation mode or a night rest mode due to a change in solar radiation An apparatus comprising: a communication interface unit for providing a communication interface with the Korea Meteorological Administration server and receiving future weather information for a predetermined period in the future from the Korea Meteorological Administration server; When the photovoltaic power generation system is operating in daytime operation mode, a driving data collection unit that collects and stores power parameters generated in string units in real-time driving situations and environmental parameters including driving environment information of the photovoltaic power generation system. ; When the operation mode of the photovoltaic power generation system is switched to the night rest mode due to a decrease in solar radiation, the power parameters and environmental parameters for each string collected by the operation data collection unit and the future weather information are used to determine the performance of the photovoltaic power generation system. a future driving state prediction unit that predicts a future driving state and generates at least one next-day driving prediction scenario; and a diagnostic unit for predicting and diagnosing whether or not a failure of the photovoltaic power generation system will occur by analyzing the at least one next-day operation prediction scenario in a state in which the operation mode of the photovoltaic power generation system is a night rest mode.

On the other hand, in order to achieve the above object, the active failure prediction and diagnosis method of the solar power generation system in a string unit provided by the present invention is a solar power generation system that operates in any one of the daytime operation mode and the nighttime rest mode by the change in solar radiation An active failure prediction and diagnosis method comprising: a weather information receiving step of receiving future weather information for a predetermined period in the future from a server of the Korea Meteorological Administration; A driving data collection step of collecting and storing power parameters generated in string units in real-time driving situations and environmental parameters including operating environment information of the photovoltaic power generation system when the photovoltaic power generation system is operating in daytime operation mode ; When the operation mode of the photovoltaic power generation system is switched to the night rest mode due to a decrease in solar radiation, the future of the photovoltaic power generation system is determined by using the power parameters and environmental parameters for each string collected in the data collection step and the future weather information. a future driving state prediction step of predicting a driving state and generating at least one next-day driving prediction scenario; and a diagnosis step of predictively diagnosing whether or not a failure of the photovoltaic power generation system will occur by analyzing the at least one next-day operation prediction scenario in a state in which the operation mode of the photovoltaic power generation system is a night rest mode. .

As described above, the active fault prediction and diagnosis device and method of the string-based solar power generation system provided by the present invention reflect the driving characteristics of the solar power generation system that repeats operation and shutdown due to solar radiation changes, When the power generation system is operating in the nighttime sleep mode, the photovoltaic power generation system can be operated more efficiently by analyzing the driving state of the previous daytime operation mode of the photovoltaic power generation system.

In addition, the present invention analyzes the operation state of the photovoltaic power generation system using real-time driving data and meteorological information collected during the previous daytime driving mode in a state where the photovoltaic power generation system is in the night rest mode, thereby operating the corresponding photovoltaic power generation system. There is an effect of obtaining an accurate judgment result for the state.

In addition, the present invention predicts the future driving state for each of at least one situation that can occur during the operation of the solar power generation system in a state where the solar power generation system is in a night rest mode, but provides real-time operation data and future weather information. After generating at least one next-day driving prediction scenario using the next-day driving forecasting scenario, failure prediction is performed based on the next-day driving forecasting scenario, and various situations including generally unexpected situations (eg, unexpected accidents, system failures, sudden weather changes, etc.) It has the effect of predicting and preparing for system errors in advance.

In addition, the present invention can safely and accurately measure the reference data by measuring reference data capable of analyzing the operating state of the photovoltaic power generation system during operation in a state where the photovoltaic power generation system is in a night rest mode, Due to this, there is an effect of enabling to accurately determine the operating state of the photovoltaic power generation system during operation.

1 is a schematic block diagram of a photovoltaic power generation system to which an active failure prediction and diagnosis device according to an embodiment of the present invention is applied.
2 is a schematic block diagram of an active fault predictive diagnosis device according to an embodiment of the present invention.
3 and 4 are schematic process flow charts for an active fault predictive diagnosis method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings, but will be described in detail so that those skilled in the art can easily practice the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. On the other hand, in order to clearly describe the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification. In addition, even if detailed descriptions are omitted, descriptions of parts that can be easily understood by those skilled in the art are omitted.

Throughout the specification and claims, when a part includes a certain component, it means that it may further include other components, not excluding other components unless otherwise stated.

1 is a schematic block diagram of a photovoltaic power generation system to which an active failure prediction and diagnosis device according to an embodiment of the present invention is applied.

Referring to FIG. 1 , a photovoltaic power generation system according to an embodiment of the present invention includes at least one solar cell string 10 composed of a plurality of solar cell modules 11, a solar connection board 20, It is configured to include an inverter 30 and an active fault predictive diagnosis device 100 .

The solar cell string 10 generates DC power by receiving sunlight, and the solar junction board 20 connects the DC power produced by each of the solar cell strings 10 in series and parallel to collect necessary power, and the inverter (30) receives DC power from the solar connection panel 20, converts it into AC power, and outputs it to the power system.

To this end, the connection board 20 is configured to include at least one string optima 21. The string optima 21 is connected to the output terminal of each of the solar cell strings, and converts the output voltage of the corresponding solar cell string to an inverter. (30), but outputs the optimal uniform voltage in line with the output voltage of other neighboring string optima. At this time, a known technology (eg, Korean Patent Registration No. 10-2242814, etc.) may be referred to for a method of equalizing voltage boosting for the string optima 21 to output equal voltage.

In addition, each of the string optima 21 is connected to an output terminal of each of the solar cell strings 10, and various real-time operation data output from the corresponding solar cell string can be transmitted to the active fault predictive diagnosis device 100.

The inverter 30 performs MPPT on the output voltage of the string optima 21, converts the result into AC power, and connects it to the grid.

The active failure prediction diagnosis device 100 analyzes the real-time operation data and predicts the operating state of the solar power generation system, but the operation of the solar power generation system operating in either a daytime operation mode or a night rest mode due to a change in solar radiation. Depending on the mode, the operation for failure prediction is also performed differently.

For example, the active fault predictive diagnosis device 100 analyzes the real-time operation state by applying the real-time operation data and real-time environment data when the photovoltaic power generation system is in daytime operation mode, and when the photovoltaic power generation system is in night rest mode. In this case, a future driving state may be predicted by applying the real-time driving data and future weather information.

When the photovoltaic power generation system is in the daytime operation mode, the operation of the active fault predictive diagnosis device 100 for analyzing the real-time operating state of the photovoltaic power generation system may employ various well-known technologies. For example, the failure prognosis diagnosis apparatus 100 receives measurement values measured in each of at least one string optima 21, and compares the measurement values with measurement values measured in other adjacent string optima 21. In comparison, the real-time operating state of each of the corresponding solar cell strings 11 may be analyzed. That is, it can be determined that an error has occurred in the solar cell string corresponding to the string optima indicating a measurement value significantly lower than the average value of measurement values measured in other surrounding string optima 21 .

On the other hand, when the photovoltaic power generation system is in the night rest mode, the active failure prediction diagnosis device 100 analyzes the driving state using real-time driving data and meteorological information collected during the daytime driving mode, and the real-time driving data, And future driving conditions can be predicted using future weather information. For example, after generating at least one next-day operation prediction scenario using the real-time operation data and future weather information, the failure prediction diagnosis apparatus 100 may perform failure prediction based on the next-day operation prediction scenario. Due to this, the present invention, in general, predicts and prepares system errors for various situations, including unexpected situations (eg, unexpected accidents, system failures, sudden climate change, etc.) in advance.

To this end, the active failure prediction diagnosis device 100 according to an embodiment of the present invention may be interlocked with, or further include, a diagnostic information management DB 200 that stores and manages various information for failure prediction diagnosis. Through an external communication network, it may communicate with the Korea Meteorological Administration server 400 providing meteorological information and may communicate with terminal devices 500 of managers or workers to notify managers or workers of failure prediction diagnosis results. At this time, the Korea Meteorological Administration server 400 provides weather information predicted from data observed at weather stations, weather station data, weather satellite data, which is weather information predicted from data collected from weather satellites, and learning results of past weather information. Weather information including numerical model data, which is weather information predicted from the generated weather model, may be generated and provided.

A schematic configuration example of the active fault predictive diagnosis device 100 is illustrated in FIG. 2 .

FIG. 2 is a schematic block diagram of an active fault predictive diagnosis device according to an embodiment of the present invention. Referring to FIGS. 1 and 2, an aspect that operates in either a daytime operation mode or a night rest mode due to a change in solar radiation The active failure prediction diagnosis device 100 of the optical power generation system includes a driving data collection unit 110, a communication interface unit (I/F) 120, a future driving state prediction unit 130, a diagnosis unit 140, and a driving algorithm. It includes a management unit 150, a predictive model generation unit 160, a rest period insulation resistance measurement unit 170, and a leakage diagnosis unit 180. In addition, the diagnostic information management DB 200 and the predictive model management DB 300 may be interlocked or further included.

The driving data collection unit 110 collects various driving data generated when the photovoltaic power generation system is operating in the daytime driving mode. To this end, the driving data collection unit 110 is connected to a string optima 21 connected to an output terminal of each of the at least one solar cell string 10, and collects various driving data from the string optima 21.

At this time, the driving data collected by the driving data collection unit 110 includes power parameters (e.g., output voltage, current, power, etc. for each string) generated in units of strings in real-time driving conditions, and the driving environment of the photovoltaic power generation system. Environmental parameters (eg, temperature, humidity, fine dust, etc.) including information may be included.

The driving data collection unit 110 may collect the driving data in units of preset time and store them in the diagnostic information management DB 200 .

The communication interface unit (I/F) 120 provides an interface with a communication network. In particular, the communication interface unit (I/F) 120 provides a communication interface with the Korea Meteorological Administration server 400, and provides future weather information from the Korea Meteorological Administration server 400 for a predetermined period in the future (eg, one week in the future). can receive For example, if the current date is March 1, 2022, the communication interface unit (I/F) 120, from the Meteorological Administration server 400, from March 2, 2022 to March 8, corresponding to one week in the future. You can receive future weather information up to the date. This is for the future driving state prediction unit 130, which will be described later, to generate a general driving scenario based on the future weather information.

In addition, the communication interface unit (I/F) 120 may also receive past weather information for the past several years (eg, 10 years) corresponding to the predetermined period in the future from the Korea Meteorological Administration server 400 . In the above example, the communication interface unit (I/F) 120 may also receive past weather information corresponding to March 2nd to March 8th during the period from 2012 to 2021.

This is for the future driving state prediction unit 130, which will be described later, to predict the possibility of occurrence of an unexpected situation deviating from the prediction of the future weather information based on the past weather information, and to generate a driving scenario accordingly. For example, from March 2nd to March 8th, the temperature usually maintains the minimum morning temperature at 1°C to 10°C above zero, but as a result of analyzing the above weather information, the minimum morning temperature in the past 10 years has been below -5°C. This is to predict the occurrence of the abnormal low temperature and generate a driving scenario accordingly, when there is a case where an abnormal low temperature phenomenon of °C has occurred.

When the solar power generation system is in the night rest mode, that is, when the operation mode of the photovoltaic power generation system is switched from the daytime operation mode to the night rest mode due to a decrease in insolation, the future driving state prediction unit 130 collects data from the daytime operation mode. A future driving state is predicted using real-time driving data and meteorological information collected from the Korea Meteorological Administration server (400).

For example, the future driving state prediction unit 130 collects power parameters for each string (eg, output voltage, current, power, etc. for each string) and environmental parameters (eg, temperature, humidity, etc.) collected by the driving data collection unit 110. fine dust, etc.), that is, real-time operation data and future weather information collected from the Korea Meteorological Administration server 400 are used to predict the future operation state of the photovoltaic power generation system, but at least one next-day operation prediction scenario can be generated. have.

To this end, the future driving state prediction unit 130 may predict driving data corresponding to future weather information by matching the real-time driving data with an environmental parameter at a time when the real-time driving data is generated.

In particular, the future driving state prediction unit 130 may store a preset error range to predict an error of the future weather information and predict a future driving state by reflecting the error range.

For example, the future driving condition prediction unit 130 may generate at least one next-day driving prediction scenario by reflecting the error range on the next-day weather information collected from the Korea Meteorological Administration server 400 . That is, when the next day's weather information is collected from the Korea Meteorological Administration server 400, the future driving state prediction unit 130 derives at least one next day's weather information corresponding to the error range based on the error range, and the derived at least A next-day driving prediction scenario for each of the next-day weather information may be generated. If the next day's minimum temperature is minus 5 degrees Celsius and the error range is ±1, the future driving state prediction unit 130 calculates the next day's driving prediction scenario for the next day's minimum temperature of -5 degrees Celsius, and the next day's lowest temperature is below zero. A next-day driving prediction scenario for the case where it is 4 degrees Celsius and a next-day driving prediction scenario for the case where the next day's minimum temperature is -6 degrees Celsius may be respectively generated.

At this time, the case where the future driving state prediction unit 130 uses the lowest temperature as a standard to generate the next-day driving prediction scenario has been described as an example, but this is only one example, and the present invention is not limited by the above example. not. That is, the future driving state prediction unit 130 may generate the next day's driving prediction scenario by considering not only the minimum temperature, but also the humidity, the concentration of fine dust, and the like.

In addition, the future driving state prediction unit 130 may predict an unexpected weather situation in the future and generate a driving prediction scenario for the next day including a driving prediction scenario when the future unexpected weather situation occurs. To this end, the future driving state prediction unit 130 may predict at least one future sudden weather situation by using the past weather information received through the communication interface unit (I/F) 120 . At this time, the future driving state prediction unit 130 counts the number of occurrences of a sudden weather phenomenon that deviate from the weather forecast from the past weather information, and when the number is greater than a predetermined value, a response corresponding to the sudden weather situation Driving prediction scenarios may be generated, or when the sudden weather phenomenon occurs even once, driving prediction scenarios corresponding to all sudden weather situations may be generated. In addition, a sudden weather change error range may be set for specifying the sudden weather change situation, and the sudden weather change may be determined only when the sudden weather change error range is out of the range, and the number of occurrences may be counted.

The diagnosis unit 140 analyzes at least one next-day driving prediction scenario generated by the future driving condition predicting unit 130 in a state where the operation mode of the photovoltaic power generation system is the night rest mode, and based on the result, the diagnosis unit 140 analyzes the Predictive diagnosis of failure of the photovoltaic system. That is, the diagnosis unit 140 analyzes the at least one next-day operation prediction scenario based on a preset diagnosis algorithm, predicts and diagnoses whether a system operation error (eg, system overload, system shutdown, etc.) has occurred. can do. At this time, the diagnosis algorithm is an algorithm for diagnosing system operation errors based on driving data and meteorological information included in the next-day operation prediction scenario of the photovoltaic system, and generated by analyzing system error history information that occurred in the past may be information.

The operation algorithm management unit 150 updates the next day operation algorithm to prevent failure based on the diagnosis result of the diagnosis unit 140 in a state where the operation mode of the photovoltaic power generation system is the night sleep mode. For example, if system overload is predicted and diagnosed by the diagnosis unit 140 at a specific time point, the driving algorithm management unit 150 may update the driving algorithm for the next day to reduce the system load at the time point based on the diagnosis result. have.

The predictive model generation unit 160 generates a driving state prediction model for any weather forecast information. That is, the predictive model generation unit 160 generates at least one driving state prediction model reflecting the secular change of the current photovoltaic power generation system based on the power parameters and environmental parameters for each string collected by the driving data collection unit 110 can do. To this end, the predictive model generation unit 160 variously collects power parameters for each time period, climate, and string according to the secular change of the photovoltaic power generation system during a specific period in the past, and environmental parameter information corresponding thereto, , The driving state prediction model may be generated by learning the information.

The predictive model generator 160 may generate the driving state predictive model and store it in the predictive model management DB 300 .

The aforementioned future driving state prediction unit 130 may generate the at least one next-day driving prediction scenario by applying future weather information to the generated driving state prediction model.

The idle insulation resistance measuring unit 170 measures the idle insulation resistance value, which is the resistance value of the insulation resistance installed for each module of the photovoltaic power generation system, in a state where the operation mode of the photovoltaic power generation system is the night sleep mode. The idle insulation resistance measuring unit 170 can accurately measure the insulation resistance by measuring the insulation resistance while the photovoltaic power generation system is in the nighttime idle mode. The off-season insulation resistance value is used as a reference value for determining leakage when the photovoltaic power generation system operates in the daytime operation mode.

The leakage diagnosis unit 180 diagnoses leakage of the photovoltaic power generation system by monitoring the amount of change in the idle insulation resistance value in a state in which the operation mode of the photovoltaic power generation system is the daytime operation mode. That is, the leakage diagnosis unit 180 monitors the amount of change in the idle insulation resistance value in the daytime operation mode of the photovoltaic power generation system, and when the idle insulation resistance value suddenly becomes low, it is determined that leakage current has occurred in the photovoltaic power generation system. Diagnose.

The controller 190 controls the overall operation of the active fault predictive diagnosis apparatus 100 based on a preset control algorithm. For example, the control unit 190 applies real-time driving data collected by the driving data collection unit 110 and meteorological information received through the communication interface unit (I/F) 120 to the control algorithm to perform active failures. Operations of each unit of the prognostic diagnosis device 100 may be controlled. That is, the control unit 190, based on the control algorithm, includes a future driving state prediction unit 130, a diagnosis unit 140, an operation algorithm management unit 150, a predictive model generation unit 160, and an insulation resistance measuring unit ( 170), and each operation of the leak diagnosis unit 180 may be controlled.

The diagnostic information management DB 200 includes information necessary for the active failure prediction diagnosis device 100 to perform failure prediction (eg, real-time operation data, weather information by date and time, etc.) and information generated in the failure diagnosis process. (eg, next-day driving prediction scenario, etc.) is stored, and the predictive model management DB 300 stores the driving state prediction model generated by the predictive model generating unit 160 .

The diagnostic information management DB 200 and the prediction model management DB 300 may be installed in the active failure prediction diagnosis device 100 or may be implemented separately and interwork with the active failure prediction diagnosis device 100 .

3 and 4 are schematic process flow charts for an active fault predictive diagnosis method according to an embodiment of the present invention. Referring to FIGS. 1 to 4 , an active fault prediction and diagnosis method of a photovoltaic power generation system according to an embodiment of the present invention is as follows.

In step S110, the communication interface unit (I/F) 120 receives weather information from the Korea Meteorological Agency server 400. In particular, in step S110, the communication interface unit (I/F) 120 provides future weather information for a predetermined period in the future (eg, one week in the future) and past several years corresponding to the predetermined period in the future (eg, , 10 years) past weather information can be received together. In step S110, the specific example of receiving weather information by the communication interface unit (I/F) 120 is duplicated as described in the description of the communication interface unit (I/F) 120 with reference to FIG. 2. omit the explanation.

In step S120, the control unit 190 determines whether or not the photovoltaic power generation system starts to operate. For example, in step S120, the control unit 190 stores the information on the operation start time of the photovoltaic system that has been set in advance, counts the time, and determines that the operation has started when the operation start time arrives, or By storing insolation information at the start of operation and measuring real-time insolation, if the real-time insolation exceeds the predetermined amount of insolation at the start of operation, it is determined that operation has started, or by detecting that real-time operation data is collected through the driving data collection unit 110 It can be determined that driving has been initiated.

In step S130, the driving data collection unit 110 collects various driving data generated from the photovoltaic power generation system. That is, in step S130, the driving data collection unit 110, when the photovoltaic power generation system is operating in the daytime driving mode, generates power parameters in units of strings in real-time driving situations (eg, output voltage, current, and power for each string). etc.), and environmental parameters (eg, temperature, humidity, fine dust, etc.) including driving environment information of the photovoltaic power generation system are collected and stored. At this time, the driving data collection unit 110 may collect the driving data in units of preset time and store them in the diagnostic information management DB 200 .

In step S140, the control unit 190 monitors the operating state of the system while the solar power generation system is operating in the daytime operation mode. That is, in step S140, the controller 190 analyzes the real-time operating state of the photovoltaic power generation system when the photovoltaic power generation system is in the daytime operation mode. To this end, the controller 190 may monitor the operating state of the solar power generation system by applying various well-known technologies. For example, the control unit 190 receives measurement values measured in each of the at least one string optima 21, compares the measurement values with measurement values measured in other adjacent string optima 21, The real-time operating state of each of the corresponding solar cell strings 11 may be analyzed. That is, it can be determined that an error has occurred in the solar cell string corresponding to the string optima indicating a measurement value significantly lower than the average value of measurement values measured in other surrounding string optima 21 .

In particular, in step S140, the leakage diagnosis unit 180 diagnoses leakage of the photovoltaic power generation system. In step S141, the leakage diagnosis unit 180 measures the measured The amount of change in the insulation resistance value during the rest period is monitored, and in steps S142 and S143, when the insulation resistance value during the rest period decreases, leakage is determined and notified to the outside (eg, manager or operator). In particular, the leakage diagnosis unit 180 stores a preset insulation resistance threshold value during the idle period, in step S141, monitors the insulation resistance value during the idle period based on the insulation resistance threshold value during the idle period, and in steps S142 and S143, the insulation resistance value during the idle period When the insulation resistance value drops below the rest insulation resistance threshold value, it is determined that leakage occurs in the photovoltaic power generation system.

Meanwhile, in order to notify the outside of the leakage, the leakage diagnosis unit 180 transfers a preset alarm signal to the communication interface unit (I/F) 120, and the communication interface unit (I/F) 120 An alarm device (eg, a speaker, a warning light) may be further included to transmit the signal to the operator's or manager's terminal device 500 or to notify the active failure prediction diagnosis device 100 thereof.

In step S150, the control unit 190 determines whether the photovoltaic system has switched to the night sleep mode. For example, in step S150, the control unit 190 stores preset information about the idle mode switching time of the photovoltaic system, counts the time, and determines that the sleep mode has been switched to the night mode when the idle mode switching time is reached, or In order to switch to the idle mode, preset rest period solar radiation information is stored and real-time solar radiation is measured, and if the real-time solar radiation is less than the preset rest period solar radiation, it is determined that the sleep mode has been switched to the night, or real-time through the driving data collection unit 110 It may be determined that the collection of driving data is stopped to detect the transition to the night rest mode.

In step S160, the predictive model generation unit 160 generates a driving state prediction model for any weather forecast information. That is, in step S160, the predictive model generation unit 160 generates at least one driving state prediction model in which the secular change of the current photovoltaic power generation system is reflected, based on the power parameters and environmental parameters for each string collected in step S130. can To this end, in step S160, the prediction model generation unit 160 generates power parameters for each time period, climate, and string according to the secular change of the photovoltaic power generation system during a specific period in the past, and environmental parameter information corresponding thereto. The driving state prediction model may be generated by variously collecting and learning the information.

In step S170, the future driving state prediction unit 130 predicts the future driving state of the photovoltaic power generation system operating in the night rest mode. That is, in step S170, the future driving state prediction unit 130 uses the meteorological information received in step S110 and the power parameters and environmental parameters for each string collected in step S130 when the photovoltaic power generation system is in the daytime operation mode to generate sunlight. The future operation state of the power generation system is predicted, but at least one next-day operation prediction scenario is generated. At this time, in step S170, the specific example in which the future driving state predicting unit 130 predicts the future driving state is as described in the description of the future driving state predicting unit 130 with reference to FIG. omit

In step S180, the idle insulation resistance measuring unit 170 measures the idle insulation resistance value, which is the resistance value of the insulation resistance installed for each module of the photovoltaic power generation system, in a state where the operation mode of the photovoltaic power generation system is the night sleep mode. . At this time, the measured idle insulation resistance value is used as a reference value for determining whether or not leakage of the photovoltaic power generation system in operation is performed in step S140.

In step S190, the control unit 190 determines whether or not to end the photovoltaic power generation system, and if not, the process after step S110 is repeatedly performed.

On the other hand, in the active fault prediction and diagnosis method of the photovoltaic power generation system of the present invention, in a state where the operation mode of the photovoltaic power generation system is the night sleep mode, the at least one next-day operation prediction scenario is analyzed to determine the level of the photovoltaic power generation system. It may further include a diagnosis step (not shown) of predictively diagnosing whether or not a failure has occurred, and a driving algorithm update step (not shown) of updating the driving algorithm for the next day to prevent a failure based on the diagnosis result of the diagnosis step. .

At this time, the diagnosis step and the driving algorithm update step are the same as those mentioned in the description of the diagnosis unit 140 and the driving algorithm management unit 150 with reference to FIG. 2 , so duplicate descriptions are omitted.

As such, the present invention reflects the driving characteristics of the photovoltaic power generation system, which repeats operation and shutdown due to changes in solar radiation, and operates in the previous daytime operation mode of the photovoltaic power generation system when the photovoltaic power generation system is operating in the night sleep mode. By analyzing the state, there is a feature that can operate the photovoltaic power generation system more efficiently.

In addition, since there are various known technologies for diagnosing whether or not the daytime operation mode of the photovoltaic power generation system is out of order, the active failure prediction and diagnosis device and method of the present invention are combined with the above known technologies to diagnose the photovoltaic power generation system. There is an effect that can maximize efficiency.

In the above, the embodiments of the present invention have been described, but the scope of the present invention is not limited thereto, and it is recognized that the present invention is easily changed from the embodiments to those of ordinary skill in the art to which the present invention belongs and is equivalent. including all changes and modifications within the scope of

100: Active fault prediction diagnosis device 110: Operation data collection unit
120: communication interface unit 130: future driving state prediction unit
140: diagnosis unit 150: driving algorithm management unit
160: prediction model generating unit 170: resting insulation resistance measuring unit
180: leak diagnosis unit 190: control unit
200: Diagnostic information management DB 300: Prediction model management DB
400: Meteorological Administration server 500: Manager/worker terminal device

Claims (12)

In the active failure predictive diagnosis device of a photovoltaic power generation system operating in either a daytime operation mode or a night rest mode due to a change in solar radiation,
a communication interface unit for providing a communication interface with the Korea Meteorological Administration server and receiving future weather information for a predetermined period in the future from the Korea Meteorological Administration server;
When the photovoltaic power generation system is operating in daytime operation mode, a driving data collection unit that collects and stores power parameters generated in string units in real-time driving situations and environmental parameters including driving environment information of the photovoltaic power generation system. ;
When the photovoltaic power generation system switches to the night rest mode, the photovoltaic power generation system uses the power parameters and environmental parameters for each string collected by the driving data collection unit in the immediately preceding daytime driving mode and the future meteorological information Predicting the driving state of the next day, predicting driving data corresponding to future weather information by matching the environmental parameter at the time when the power parameter value for each string occurred, and based on the prediction result, a future driving state prediction unit generating at least one next-day driving prediction scenario; and
In a state in which the operation mode of the photovoltaic power generation system is a night rest mode, a diagnostic unit for predicting and diagnosing whether or not a failure of the photovoltaic power generation system will occur by analyzing the at least one next-day operation prediction scenario;
The communication interface unit
Receive together past weather information for the past several years corresponding to the predetermined period in the future,
The future driving state prediction unit
At least one future abrupt weather situation is predicted using the past weather information, and a next-day operation prediction scenario including an operation prediction scenario when the future abrupt weather situation occurs Active type of string-based solar power generation system, characterized in that Fault prediction diagnosis device. According to claim 1,
In a state where the operation mode of the photovoltaic power generation system is a night rest mode, based on the diagnosis result of the diagnosis unit, a driving algorithm management unit for updating a next-day operation algorithm to prevent failure Photovoltaic power generation system active fault prediction and diagnosis device. According to claim 1 or 2,
Based on the power parameters and environmental parameters for each string collected by the driving data collection unit, a prediction model generation unit for generating at least one driving state prediction model in which the secular change of the current photovoltaic power generation system is reflected;
The future driving state prediction unit
The solar power generation system active fault predictive diagnosis device of a string unit, characterized in that for generating the at least one next-day driving prediction scenario by applying the future weather information to the driving state prediction model. The method of claim 3, wherein the future driving state prediction unit
A string-based photovoltaic power generation system active fault prediction characterized by storing a preset error range to predict an error of the future weather information and generating the at least one next-day operation prediction scenario by reflecting the error range diagnostic device. delete According to claim 1,
In a state in which the operation mode of the photovoltaic power generation system is a night sleep mode, a rest insulation resistance measurement unit for measuring a rest insulation resistance value, which is a resistance value of insulation resistance installed for each module of the photovoltaic power generation system; and
In a state where the operation mode of the photovoltaic power generation system is a daytime operation mode, a leakage diagnosis unit for diagnosing leakage of the photovoltaic power generation system by monitoring the amount of change in the idle insulation resistance value of the string unit, characterized in that it further comprises Photovoltaic power generation system active fault prediction and diagnosis device. In the active fault prediction and diagnosis method of a string-based photovoltaic power generation system operating in any one of the daytime operation mode and the nighttime sleep mode due to changes in solar radiation,
A weather information receiving step of receiving future weather information for a predetermined period in the future from a server of the Korea Meteorological Administration;
A driving data collection step of collecting and storing power parameters generated in string units in real-time driving situations and environmental parameters including operating environment information of the photovoltaic power generation system when the photovoltaic power generation system is operating in daytime operation mode ;
When the photovoltaic power generation system switches to the night rest mode, the photovoltaic power generation system uses the power parameters and environmental parameters for each string collected in the driving data collection step in the immediately preceding daytime driving mode and the future meteorological information Predicting the next day's driving state, predicting driving data corresponding to future weather information by matching the environmental parameter at the time when the power parameter value for each string occurs, and based on the prediction result, at least one next day driving prediction scenario a future driving state prediction step of generating; and
In a state in which the operation mode of the photovoltaic power generation system is a night rest mode, a diagnosis step of predicting and diagnosing whether or not a failure of the photovoltaic power generation system will occur by analyzing the at least one next-day operation prediction scenario,
The weather information receiving step is
Receive together past weather information for the past several years corresponding to the predetermined period in the future,
The future driving state prediction step is
At least one future abrupt weather situation is predicted using the past weather information, and a next-day operation prediction scenario including an operation prediction scenario when the future abrupt weather situation occurs Active type of string-based solar power generation system, characterized in that Failure prognosis diagnosis method. According to claim 7,
In a state in which the operation mode of the photovoltaic power generation system is a night rest mode, a driving algorithm update step of updating a next-day driving algorithm to prevent a failure based on the diagnosis result of the diagnosis step Characterized in that it further comprises a string A method for predicting and diagnosing active failures of photovoltaic power generation systems in units. According to claim 7 or 8,
Based on the power parameters and environmental parameters for each string collected in the operation data collection step, a predictive model generation step of generating at least one driving state prediction model in which the secular change of the current photovoltaic power generation system is reflected;
The future driving state prediction step is
The method of predicting and diagnosing an active failure of a solar power generation system in units of strings, characterized in that generating the at least one next-day driving prediction scenario by applying the future weather information to the driving state prediction model. The method of claim 9, wherein the predicting the future driving state
A method for predicting and diagnosing an active failure of a solar power generation system in units of strings, characterized in that generating the at least one next-day operation prediction scenario by reflecting a preset error range to predict an error of the future weather information. delete According to claim 7,
In a state in which the operation mode of the photovoltaic power generation system is a night sleep mode, a rest insulation resistance measurement step of measuring a rest insulation resistance value, which is a resistance value of insulation resistance installed for each module of the photovoltaic power generation system; and
In a state where the operation mode of the photovoltaic power generation system is a daytime operation mode, a leakage diagnosis step of diagnosing leakage of the photovoltaic power generation system by monitoring the amount of change in the idle insulation resistance value String unit characterized in that it further comprises A Method for Prognosing and Diagnosing Active Failures of Photovoltaic Power Generation Systems.

Images