Power generation anomaly warning system and method for photovoltaic power generation equipment

Abstract

The present invention provides a power generation anomaly warning system and method for photovoltaic power generation equipment, which relates to the technical field of photovoltaic power generation. It includes a partial discharge state monitoring system for power photovoltaic equipment that detects partial discharge signals with a spatial radio frequency partial discharge detection radar, and specifically includes the following steps: First, conduct preliminary preparations, collect basic information and construct a big data model. Then, conduct screening and verification of partial discharge signals, process, analyze, screen and determine the detected partial discharge signals. Finally, conduct actual monitoring operations, analyze the partial discharge signals, and adjust the monitoring intensity according to the partial discharge intensity, partial discharge frequency and basic information of the photovoltaic equipment. The present invention uses a spatial radio frequency partial discharge detection radar to detect partial discharge signals of photovoltaic equipment and photovoltaic power generation network facilities in urban environments, and combines the processing and monitoring of partial discharge events of facilities other than photovoltaic equipment in the photovoltaic power generation network.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and particularly to an abnormal power generation warning system and method for photovoltaic power generation equipment. Background Art

[0002] Photovoltaic power generation anomalies generally refer to faults or performance degradation that occur during the operation of a photovoltaic system. These anomalies may include the following situations: Abnormal insulation impedance: This is one of the most common faults in a photovoltaic power station and may cause a decrease in the power generation efficiency of the system; Overload and underload problems: A photovoltaic power station may encounter problems of load mismatch. Overload may damage equipment, while underload will result in energy waste. These problems can be solved by monitoring and analyzing the load conditions, as well as increasing the equipment capacity or optimizing the load distribution; Inter-phase fault: It refers to an accidental connection between two different potential points in a photovoltaic array, resulting in partial short-circuiting of some photovoltaic modules, thereby affecting the output power of the entire array; Component failure: The photovoltaic components themselves may also fail due to various reasons, such as aging, physical damage, or manufacturing defects, etc.; Environmental factors: Such as shadow occlusion, pollution, temperature changes, etc. will also affect the power generation efficiency of photovoltaic panels; Electrical connection problems: Including poor wiring, damaged connectors, etc., all of which may lead to a decrease in system efficiency; Inverter failure: The inverter is a key device for converting direct current into alternating current. If the inverter fails, it will directly affect the electrical energy output of the entire system; Grid connection problems: With the increase in the scale of photovoltaic power generation grid connection, the problems that occur during grid connection operation will also affect the stable operation of the photovoltaic power station.

[0003] Therefore, during the process of photovoltaic power generation, in order to ensure the normal operation of a photovoltaic power station, it is necessary to conduct regular maintenance and inspections to promptly discover and solve the above various potential problems. At the same time, adopting advanced monitoring and diagnostic technologies is also an important means to improve the reliability of a photovoltaic power station.

[0004] Partial discharge usually occurs between electrodes but does not penetrate the entire electrode. This is due to weaknesses existing inside the insulation or defects during the production process, and repeated breakdown and extinction phenomena occur under the action of a high electric field intensity. Each partial discharge will have an impact on the insulating medium. Slight partial discharge has a relatively small impact on the insulation of power equipment, while strong partial discharge will cause the insulation strength to rapidly decline. Partial discharge is one of the important reasons for the ultimate insulation breakdown of high-voltage electrical equipment and is also an important sign of insulation deterioration.

[0005] Currently, the detection of partial discharge in photovoltaic power equipment generally adopts high-speed signal acquisition technology (such as ADC + FPGA + ARM), combined with high-frequency pulse current, ultrasonic, transient earth overvoltage, and ultra-high-frequency partial discharge sensors to accurately monitor the partial discharge signals that may be generated inside the equipment, thereby preventing faults caused by partial discharge.

[0006] When a component failure occurs in a photovoltaic power generation device, it generally causes abnormalities in the local or entire power generation network due to the component failure. Among them, partial discharge of internal conductors caused by the aging of the insulation structure is a very common situation in component failures. In particular, photovoltaic power stations are generally built in locations with strong illumination, and long-term illumination and large day-night temperature differences are also likely to cause deterioration of the internal insulation structure.

[0007] The existing partial discharge monitoring of power photovoltaic devices is generally completed based on the joint detection of multiple sensors, which requires the acquisition and processing of multiple sets of detection information, resulting in a large amount of data. Moreover, this method only detects the partial discharge situation of the photovoltaic device itself. Although the data is fine enough, it still cannot analyze the partial discharge situation in combination with the photovoltaic power generation network environment where the photovoltaic device is located. The monitoring angle is relatively narrow. In the same photovoltaic power generation network, changes in the circuit voltage caused by partial discharge of other facilities may affect the operation of the photovoltaic device itself, which may lead to accelerated aging of the photovoltaic device hardware. In addition, this method cannot give an early warning of photovoltaic power generation anomalies, but can only display the abnormal state of the photovoltaic power generation device in real time, so there is room for improvement. Summary of the Invention

[0008] The purpose of the present invention is to solve the deficiencies in the prior art. By detecting the partial discharge phenomenon of the entire photovoltaic power generation network, while investigating the partial discharge events of the photovoltaic device itself, the partial discharge events of other devices in the photovoltaic power generation network are used to calculate the change in the power supply state of the photovoltaic power generation network, and then the state of the photovoltaic device is further monitored and predicted.

[0009] To achieve the above purpose, the present invention adopts the following technical solutions: An abnormal power generation warning system and method for a photovoltaic power generation device, including a partial discharge state monitoring system for a power photovoltaic device that detects partial discharge signals with a spatial radio frequency partial discharge detection radar, specifically including:

[0010] Spatial radio frequency partial discharge detection radar: The spatial radio frequency partial discharge detection radar can detect partial discharge of the photovoltaic device in a live state and can achieve long-distance and non-contact detection;

[0011] Radar signal processing module: The radar signal processing module converts analog signals into digital signals through analog-to-digital conversion, and suppresses noise and interference through beamforming, pulse compression, clutter filtering, and Doppler processing technologies;

[0012] Server: The server is equipped with a big data processing model, a radar signal processing module, and a data storage module, and receives data from the spatial radio frequency partial discharge detection radar;

[0013] Big data processing model: The big data processing model is used to obtain the basic information of photovoltaic equipment as the basic reference factor for data processing, analyze the location and frequency of partial discharge signals, and combine the electricity consumption situation in the corresponding location area to analyze the reasons for the occurrence of partial discharge situations;

[0014] Display module: The display module is used to display the specific location where the partial discharge signal appears, and can intuitively observe the photovoltaic power generation system involved in the corresponding photovoltaic equipment;

[0015] Data storage module: The data storage module is used to store the basic data of each photovoltaic equipment, the signal data received by the detection radar, the processing records of partial discharge signals, the photovoltaic geographic information data, and the photovoltaic power generation network data.

[0016] As a preferred implementation manner, the spatial radio frequency partial discharge detection radar detects the partial discharge signal, and then sends the partial discharge signal to the radar signal processing module for data processing. The processed partial discharge signal data is transmitted to the big data processing model, and the big data processing model combines the basic data of the photovoltaic equipment, the photovoltaic power generation network data, and the power supply data to analyze the location and intensity of the partial discharge signal, and sends the analysis result to the display module for display.

[0017] As a preferred implementation manner, the data storage module receives the basic data of the photovoltaic equipment, copies the partial discharge signal data and sends it to the data storage module for storage backup when the spatial radio frequency partial discharge detection radar receives the partial discharge signal. After the big data processing model finishes processing the partial discharge signal data, it sends the processing record and result to the data storage module for storage backup.

[0018] A method for warning of abnormal power generation of photovoltaic power generation equipment. According to the above-mentioned system and method for warning of abnormal power generation of photovoltaic power generation equipment, it includes the following specific steps:

[0019] S1. First, carry out preliminary preparations, collect basic information and build a big data model;

[0020] S2. Then, carry out screening and verification of partial discharge signals, and process, analyze, screen and judge the detected partial discharge signals;

[0021] S3. Finally, carry out actual monitoring operations, analyze the partial discharge signals, and adjust the monitoring intensity according to the partial discharge intensity, partial discharge frequency and the basic data of photovoltaic equipment.

[0022] As a preferred implementation manner, the preliminary preparation stage specifically includes the following steps:

[0023] S1.1. First, select the installation location of the spatial radio frequency partial discharge detection radar according to the positions of various photovoltaic devices in the photovoltaic power station;

[0024] S1.2. Then, collect the basic information of the photovoltaic power generation network within the detection range of the radar, including: the geographical location, model, service life, rated power, real-time power, and maintenance log of the photovoltaic devices, as well as the photovoltaic power generation network data and the satellite map of the photovoltaic power station. Then, input the collected basic information into the data storage module for storage;

[0025] S1.3. Build a big data processing model based on the efficient adaptive online data stream clustering algorithm. Combine the basic data of photovoltaic devices, the photovoltaic power generation network data, and the photovoltaic geographical information data, and incorporate the Apriori algorithm into the big data processing model;

[0026] S1.4. Obtain the partial discharge signals detected by the spatial radio frequency partial discharge detection radar that has been put into use as the training set of the big data processing model, and use the training set to train the big data processing model;

[0027] S1.5. The big data processing model receives the basic data of the photovoltaic devices, and establishes a geographical location coordinate system with the photovoltaic power generation network where the photovoltaic devices are located. The origin of the coordinate system is the location of the spatial radio frequency partial discharge detection radar;

[0028] S1.6. Process the scale of the satellite map of the photovoltaic power station and merge it into the coordinate system, and verify the accuracy of the corresponding positions of the photovoltaic power generation network in the photovoltaic power generation network and the map;

[0029] S1.7. Set in the server: when a partial discharge signal is obtained, copy the signal and send it to the radar signal processing module and the data storage module respectively; when the big data processing model obtains the analysis result of the partial discharge signal, copy the result and send it to the display module and the data storage module respectively.

[0030] As a preferred implementation manner, the screening and verification of the partial discharge signal specifically includes the following steps:

[0031] S2.1. Detect the partial discharge signal by the spatial radio frequency partial discharge detection radar;

[0032] S2.2. Then, the radar signal processing module converts the analog signal into a digital signal through analog-to-digital conversion, suppresses noise and interference through beamforming, pulse compression, clutter filtering, and Doppler processing technologies, and then sends the processed partial discharge signal to the big data processing model;

[0033] S2.3. The big data processing model analyzes the partial discharge signal, then extracts the position information of the partial discharge signal and the position information of the photovoltaic device, and makes a comparison;

[0034] S2.4. Determine the relative relationship between the position of the partial discharge signal and the position of the photovoltaic device according to the detection error of the spatial radio frequency partial discharge detection radar for geographical location information, make a determination, and then dispatch maintenance personnel to the site for investigation;

[0035] S2.5. After investigation, upload the inspection log to the data storage module, and the big data processing model assigns the inspection log to the maintenance log data in the basic data of the corresponding photovoltaic device.

[0036] As a preferred implementation manner, the specific monitoring operations include the following steps:

[0037] S3.1. The big data processing model receives the partial discharge signal, processes the partial discharge signal using an efficient adaptive online data stream clustering algorithm, extracts the position information of the partial discharge signal, and marks the partial discharge signal data in the coordinate system according to the position of the partial discharge signal;

[0038] S3.2. Use the Apriori algorithm to analyze the position distribution data of the partial discharge signal and the basic data of the photovoltaic device;

[0039] S3.3. Check the position of the partial discharge signal. When the signal error range detected by the radar is within the position of the photovoltaic device, it is regarded as the partial discharge signal of the photovoltaic device, otherwise it is regarded as the partial discharge signal of the photovoltaic power generation network facility;

[0040] S3.4. Then the big data processing model performs partial discharge analysis of the photovoltaic device and partial discharge analysis of other facilities for the partial discharge signal of the photovoltaic device and the partial discharge signal of the photovoltaic power generation network facility respectively;

[0041] S3.5. Then the big data processing model combines the partial discharge analysis of the photovoltaic device and the partial discharge analysis of other facilities to perform partial discharge analysis of the photovoltaic power generation network, and gives an early warning according to the partial discharge intensity, the partial discharge occurrence frequency and the possible abnormal conditions of the basic data of the local photovoltaic device of the photovoltaic device.

[0042] As a preferred implementation manner, in step S2.4, the specific determination method is:

[0043] 1) If the position error of the partial discharge signal is within the position range of the photovoltaic device, the big data processing model retrieves and exports the basic information of the corresponding photovoltaic device, and dispatches maintenance personnel to the site for inspection;

[0044] 2) If the position error of the partial discharge signal is outside the position range of the photovoltaic device, the big data processing model matches the position of the partial discharge signal according to the photovoltaic power generation network, retrieves the basic data of the photovoltaic devices in the corresponding network, analyzes the photovoltaic devices directly affected by the photovoltaic power generation network corresponding to the partial discharge position, and analyzes the weight relationship of the corresponding affected photovoltaic devices, and dispatches maintenance personnel to check the photovoltaic devices in descending order of weight.

[0045] As a preferred implementation manner, in the step S3.4, the specific processing methods for the partial discharge analysis of photovoltaic devices and the partial discharge analysis of other facilities are as follows:

[0046] 1) The big data processing model retrieves the circuit data in the partial discharge signal of the photovoltaic device, the photovoltaic power generation network data, and the basic data of the photovoltaic device for partial discharge analysis of the photovoltaic device, and then dispatches personnel to repair the photovoltaic device;

[0047] 2) The big data processing model retrieves the partial discharge signal of the photovoltaic power generation network facilities, the photovoltaic power generation network data, the geographical location of the photovoltaic device, and the real-time power of the photovoltaic device for partial discharge analysis of other facilities, and then dispatches personnel to investigate the partial discharge position.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0049] In the present invention, a spatial radio frequency partial discharge detection radar is used to detect the partial discharge signals of photovoltaic devices and photovoltaic power generation network facilities in the urban environment, and then the partial discharge data is processed and analyzed in combination with the basic information of the photovoltaic devices, the photovoltaic power generation network information, the satellite map of the photovoltaic power station, and the big data processing model. By combining the processing and monitoring of partial discharge events of facilities other than photovoltaic devices in the photovoltaic power generation network, the influence degree of this event on the operation state of the photovoltaic device itself is calculated. Compared with the prior art's partial discharge monitoring means that only targets the photovoltaic device itself, the monitoring of the influencing factors of the photovoltaic device state is more comprehensive and accurate, and early warnings can be made according to the partial discharge events based on the global monitoring of the photovoltaic power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the power generation anomaly early warning system and method for photovoltaic power generation equipment proposed by the present invention;

[0051] Figure 2 Flow chart of the power generation anomaly early warning method for photovoltaic power generation equipment proposed by the present invention;

[0052] Figure 3 Flow chart of the preliminary preparation stage of the power generation anomaly early warning method for photovoltaic power generation equipment proposed by the present invention;

[0053] Figure 4 This is the flowchart for screening and verification of partial discharge signals in the power generation anomaly warning method for photovoltaic power generation equipment proposed by the present invention;

[0054] Figure 5 This is the flowchart for actual monitoring operation of the power generation anomaly warning method for photovoltaic power generation equipment proposed by the present invention;

[0055] Figure 6 This is the schematic diagram for processing partial discharge signals in the power generation anomaly warning method for photovoltaic power generation equipment proposed by the present invention. Detailed implementation mode

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] Embodiment 1

[0058] As Figure 1 shown, the present invention provides a technical solution: a power generation anomaly warning system and method for photovoltaic power generation equipment, including a partial discharge state monitoring system for power photovoltaic equipment that detects partial discharge signals with a spatial radio frequency partial discharge detection radar, specifically including:

[0059] Spatial radio frequency partial discharge detection radar: The spatial radio frequency partial discharge detection radar can detect partial discharge of photovoltaic equipment in a live state and can achieve long-distance and non-contact detection;

[0060] Radar signal processing module: The radar signal processing module converts analog signals into digital signals through analog-to-digital conversion, and suppresses noise and interference through beamforming, pulse compression, clutter filtering, and Doppler processing technologies;

[0061] Server: The server is equipped with a big data processing model, a radar signal processing module, and a data storage module, and receives data from the spatial radio frequency partial discharge detection radar;

[0062] Big data processing model: The big data processing model is used to obtain the basic information of photovoltaic equipment as the basic reference factor for data processing, analyze the location and frequency of the occurrence of partial discharge signals, and analyze the cause of the occurrence of partial discharge conditions in combination with the power consumption situation in the corresponding location area;

[0063] Display module: The display module is used to display the specific location where the partial discharge signal appears, and can intuitively observe the photovoltaic power generation system involved in the corresponding photovoltaic equipment;

[0064] Data storage module: The data storage module is used to store the basic data of each photovoltaic device, the signal data received by the detection radar, the processing records of partial discharge signals, the photovoltaic geographic information data, and the photovoltaic power generation network data.

[0065] In the above content:

[0066] The spatial radio frequency partial discharge detection radar has the function of wide-band coverage. Combining its strong anti-interference ability and high sensitivity characteristics, when performing long-distance non-contact detection on photovoltaic devices in the urban environment, it can avoid the detection interference of the complex urban steel structure environment on partial discharge signals, and the large-scale detection ability can also be realized when multiple radars are combined;

[0067] In addition, by using the characteristics of the spatial radio frequency partial discharge detection radar, multiple radars can be used to cooperate to accurately locate the partial discharge position, reduce the processing difficulty of position information during the analysis of partial discharge information, and improve the accuracy and efficiency;

[0068] The radar signal processing module performs basic processing on the partial discharge signals detected by the radar, converts the analog signals into digital signals, which is convenient for the big data processing model to process the data. And the data storage module stores the basic data of photovoltaic devices and partial discharge signals, which is convenient for the big data processing model to calculate the data changes, so that the big data processing model can increase the calculation of the correlation between various data as the data volume increases.

[0069] Furthermore, the spatial radio frequency partial discharge detection radar detects partial discharge signals, and then sends the partial discharge signals to the radar signal processing module for data processing. The processed partial discharge signal data is transmitted to the big data processing model. The big data processing model analyzes the position and intensity of the partial discharge signals in combination with the basic data of photovoltaic devices, the photovoltaic power generation network data, and the power supply data, and sends the analysis results to the display module for display;

[0070] Among them, the data storage module receives the basic data of photovoltaic devices. When the spatial radio frequency partial discharge detection radar receives partial discharge signals, it copies the partial discharge signal data and sends it to the data storage module for storage backup. After the big data processing model finishes processing the partial discharge signal data, it sends the processing records and results to the data storage module for storage backup.

[0071] In the above content, the partial discharge signal is detected by the spatial radio frequency partial discharge detection radar, and then the partial discharge signal is preliminarily processed by the radar signal processing module. Then, the big data processing model is used to analyze the partial discharge signal in combination with the basic data of the photovoltaic equipment, the processing records of the partial discharge signal, the photovoltaic geographic information data, and the photovoltaic power generation network data in the data storage module. Finally, the analysis results and the partial discharge information are jointly sent to the display module for display.

[0072] In this embodiment, the present invention detects the partial discharge phenomenon through the spatial radio frequency partial discharge detection radar. Compared with the prior art, this method can greatly reduce the number of detectors, and the partial discharge signals that the spatial radio frequency partial discharge detection radar can detect are not limited to the partial discharge signals of photovoltaic equipment. Therefore, the spatial radio frequency partial discharge detection radar can combine the partial discharges of other facilities in the photovoltaic power generation network and the basic data of photovoltaic equipment, which can achieve the effect of monitoring the environment of the photovoltaic power generation network. Since photovoltaic equipment is an important part of the photovoltaic power generation network, the partial discharge conditions of other facilities in the photovoltaic power generation network can also affect the operation of photovoltaic equipment. Therefore, monitoring the partial discharges of other facilities in the photovoltaic power generation network is beneficial to the monitoring of partial discharges of photovoltaic equipment.

[0073] Embodiment 2

[0074] As Figure 2 shown, the power generation anomaly warning method for photovoltaic power generation equipment, according to the power generation anomaly warning system and method for photovoltaic power generation equipment described in Embodiment 1, includes the following specific steps:

[0075] S1. First, carry out preliminary preparations, collect basic information and construct a big data model;

[0076] S2. Then, carry out screening and verification of partial discharge signals, and process, analyze, screen, and determine the detected partial discharge signals;

[0077] S3. Finally, carry out actual monitoring operations, analyze the partial discharge signals, and adjust the monitoring intensity according to the partial discharge intensity, partial discharge frequency, and basic data of photovoltaic equipment.

[0078] In this embodiment, during the preliminary preparation process, photovoltaic power generation network data, basic data of photovoltaic devices, and partial discharge signal data detected by other spatial radio frequency partial discharge detection radars are obtained, which serve as the basis for selecting algorithms and models during the construction of the big data model and can also be used as a training set to train the big data model. Subsequently, the screening and verification of partial discharge signals are carried out by using the constructed big data processing model to process the partial discharge signals detected by the spatial radio frequency partial discharge detection radar, as a preparatory work for monitoring operations. Finally, during the actual monitoring operation, based on the basic information of photovoltaic devices and the photovoltaic power generation network, the big data processing model is used to process the partial discharge signals. After obtaining the partial discharge location and processing method, the big data processing model is further used to analyze and process the partial discharge event, and an early warning is issued for abnormal photovoltaic power generation situations according to the results of the analysis and processing.

[0079] Embodiment 3

[0080] As Figure 3 shown, the specific steps of the preliminary preparation stage include the following:

[0081] S1.1. First, determine the installation quantity of spatial radio frequency partial discharge detection radars according to the positions of various photovoltaic devices in the photovoltaic power station and deploy them;

[0082] Among them, determine the deployment quantity of the spatial radio frequency partial discharge radar according to the urban scale and the layout of the photovoltaic power generation network, then select a location for the installation of the radar according to the coverage of urban buildings, and then implement the deployment of the spatial radio frequency partial discharge radar;

[0083] When carrying out the deployment site selection, a three-dimensional micro-scale model can be constructed based on the satellite map of the city combined with the urban building distribution information to analyze the interference of urban buildings on partial discharge signals, and avoid areas with severe interference for site selection. When the interference of urban buildings on partial discharge signals is severe, a combination of multiple groups of spatial radio frequency partial discharge detection radars can be used to improve the detection accuracy;

[0084] S1.2. Then collect the basic information of the photovoltaic power generation network within the detection range of the radar, and input the collected basic information into the data storage module for storage;

[0085] Among them, the basic information of the photovoltaic power generation network includes: the geographical location of photovoltaic devices, the model of photovoltaic devices, the service life of photovoltaic devices, the rated power of photovoltaic devices, the real-time power of photovoltaic devices, and the maintenance and repair logs of photovoltaic devices, as well as the photovoltaic power generation network data and the satellite map of the photovoltaic power station;

[0086] S1.3. Based on the efficient adaptive online data stream clustering algorithm, construct a big data processing model. Combine the basic data of photovoltaic equipment, the power generation network data of photovoltaic power generation, and the photovoltaic geographic information data, and incorporate the Apriori algorithm into the big data processing model;

[0087] Among them, the efficient adaptive online data stream clustering algorithm is used to effectively process and identify various partial discharge sources and on-site interference sources for partial discharge detection. It has great advantages in processing the partial discharge information of the photovoltaic power generation network. This algorithm is particularly suitable for scenarios where the data distribution is constantly changing, such as the multi-source interference problem in partial discharge detection;

[0088] Moreover, the Apriori algorithm is a key algorithm applied to association rule mining. In the big data processing model, using this algorithm can enable the big data processing model to centrally process the partial discharge signals, the basic signals of photovoltaic equipment, and the data of the photovoltaic power generation network, and has great advantages in analyzing and summarizing the data associations among the three, which is convenient for the big data processing model to analyze partial discharge events;

[0089] S1.4. Obtain the partial discharge signals detected by the space radio frequency partial discharge detection radar that has been put into use as the training set of the big data processing model, and use the training set to train the big data processing model;

[0090] Among them, since partial discharge events are not common, it is somewhat difficult to obtain the training set data of the big data processing model. However, using the partial discharge signals detected by other space radio frequency partial discharge detection radars can greatly reduce the difficulty of data acquisition in this system and facilitate the construction of the big data processing model;

[0091] S1.5. The big data processing model receives the basic data of photovoltaic equipment, establishes a geographical position coordinate system based on the photovoltaic power generation network where the photovoltaic equipment is located, and the origin of the coordinate system is the location of the space radio frequency partial discharge detection radar;

[0092] S1.6. Process the scale of the satellite map of the photovoltaic power station and merge it into the coordinate system, and verify the accuracy of the corresponding positions of the photovoltaic power generation network in the photovoltaic power generation network and the map;

[0093] Among them, set the satellite map of the photovoltaic power station and the coordinate system of the photovoltaic power generation network as two layers respectively. When the big data model inputs the position coordinates of the partial discharge signal, it is mainly marked in the coordinate system of the photovoltaic power generation network, and then the specific position of the partial discharge signal is found by overlapping the satellite map of the photovoltaic power station and the coordinate system of the photovoltaic power generation network, and then verified manually to determine the accuracy of the overlap between the satellite map of the photovoltaic power station and the coordinate system of the photovoltaic power generation network;

[0094] S1.7. Set in the server that when a partial discharge signal is obtained, the signal is copied and sent to the radar signal processing module and the data storage module respectively, and when the big data processing model obtains the analysis result of the partial discharge signal, the result is copied and sent to the display module and the data storage module respectively.

[0095] In this embodiment, by collecting the basic information of various facilities in the photovoltaic power generation network, it is used as the basis for algorithm and model selection when constructing the big data processing model, so as to determine the accuracy and pertinence of the big data processing model for the analysis and processing of partial discharge signals. Then, the big data processing model is further trained with the partial discharge signals, and then verified with the new partial discharge signals detected by the spatial radio frequency partial discharge detection radar. The method of establishing a coordinate system is used to intuitively display the position of the partial discharge signal. This design is targeted at the monitoring of partial discharge events occurring in the photovoltaic power generation network where the photovoltaic equipment is located, and the analysis and processing of partial discharge signals can be used as the basis for monitoring and adjustment. The monitoring of the overall photovoltaic power generation network can make the environment where the photovoltaic equipment is located more controllable, that is, improve the reliability of photovoltaic equipment status monitoring.

[0096] Embodiment 4

[0097] As Figure 4 shown, the screening and verification of the partial discharge signal specifically includes the following steps:

[0098] S2.1. Detect the partial discharge signal by the spatial radio frequency partial discharge detection radar;

[0099] S2.2. Then, the radar signal processing module converts the analog signal into a digital signal through analog-to-digital conversion, suppresses noise and interference through beamforming, pulse compression, clutter filtering and Doppler processing technologies, and then sends the processed partial discharge signal to the big data processing model;

[0100] S2.3. The big data processing model analyzes the partial discharge signal, then extracts the position information of the partial discharge signal and the position information of the photovoltaic equipment, and makes a comparison;

[0101] S2.4. Determine the relative relationship between the position of the partial discharge signal and the position of the photovoltaic equipment according to the detection error of the spatial radio frequency partial discharge detection radar for geographical location information, and make a determination, and then dispatch maintenance personnel to the site for investigation;

[0102] Among them, the specific determination method is:

[0103] 1) If the position error of the partial discharge signal is within the range of the position of the photovoltaic equipment, the big data processing model retrieves the basic information of the corresponding photovoltaic equipment and exports it, and dispatches maintenance personnel to the site for inspection;

[0104] 2) If the position error of the partial discharge signal is outside the position range of the photovoltaic device, the big data processing model matches the position of the partial discharge signal according to the photovoltaic power generation network, retrieves the basic data of the photovoltaic devices in the corresponding network, analyzes the photovoltaic devices directly affected by the photovoltaic power generation network corresponding to the partial discharge position, and analyzes the weight relationship of the corresponding affected photovoltaic devices, and dispatches maintenance personnel to check the photovoltaic devices in descending order of weight;

[0105] Further, this determination method is based on the satellite map of the photovoltaic power station and the coordinate system of the photovoltaic power generation network, and determines whether the partial discharge signal belongs to the photovoltaic power generation network or the photovoltaic device itself by coordinate difference, relying on the accuracy of the spatial radio frequency partial discharge detection radar itself for detecting partial discharge signals;

[0106] Moreover, the method of dispatching personnel to the site for investigation or repair is to obtain the fault data of the photovoltaic power generation network while performing actual repair operations, and the big data processing model further improves the calculation of the association between the basic data of photovoltaic devices, the data of the photovoltaic power generation network, and the partial discharge information according to the manifestations and causes of the faults;

[0107] S2.5. After the investigation, upload the inspection log to the data storage module, and the big data processing model assigns the inspection log to the maintenance log data in the corresponding basic data of the photovoltaic device.

[0108] In this embodiment, since the partial discharge signals are all of the same signal form in terms of signal type, the partial discharge signals received by the spatial radio frequency partial discharge detection radar will also include the partial discharge signals of other devices. Since the power output by the photovoltaic device has little impact on the photovoltaic device itself, screening and verifying the partial discharge signals that do not belong to the photovoltaic power generation network part can eliminate unnecessary partial discharge signals, and this measure can reduce the pressure on the server during operation.

[0109] Embodiment 5

[0110] As Figure 5 and Figure 6 shown, the specific monitoring operation includes the following steps:

[0111] S3.1. The big data processing model receives the partial discharge signal, uses the efficient adaptive online data stream clustering algorithm to process the partial discharge signal, extracts the position information of the partial discharge signal, and marks the partial discharge signal data in the coordinate system according to the position of the partial discharge signal;

[0112] S3.2. Use the Apriori algorithm to analyze the partial discharge signal position distribution data and the basic data of the photovoltaic device;

[0113] S3.3. Check the position of the partial discharge signal. When the signal error range detected by the radar is within the position of the photovoltaic device, it is regarded as the partial discharge signal of the photovoltaic device; otherwise, it is regarded as the partial discharge signal of the photovoltaic power generation network facilities.

[0114] S3.4. Then, the big data processing model conducts partial discharge analysis of the photovoltaic device and partial discharge analysis of other facilities for the partial discharge signal of the photovoltaic device and the partial discharge signal of the photovoltaic power generation network facilities respectively.

[0115] Among them, the specific processing methods of the partial discharge analysis of the photovoltaic device and the partial discharge analysis of other facilities are as follows:

[0116] 1) The big data processing model retrieves the partial discharge signal of the photovoltaic device, the circuit data in the photovoltaic power generation network data, and the basic data of the photovoltaic device to conduct partial discharge analysis of the photovoltaic device, and then dispatches personnel to repair the photovoltaic device.

[0117] 2) The big data processing model retrieves the partial discharge signal of the photovoltaic power generation network facilities, the photovoltaic power generation network data, the geographical location of the photovoltaic device, and the real-time power of the photovoltaic device to conduct partial discharge analysis of other facilities, and then dispatches personnel to investigate the partial discharge position.

[0118] S3.5. Then, the big data processing model combines the partial discharge analysis of the photovoltaic device and the partial discharge analysis of other facilities to conduct partial discharge analysis of the photovoltaic power generation network, and issues early warnings according to the partial discharge intensity, the partial discharge occurrence frequency, and the possible abnormal conditions of the basic data of the local photovoltaic power generation equipment of the photovoltaic device.

[0119] In this embodiment, the big data processing model obtains the association between the changed facilities associated with the partial discharge event from the maintenance log data in the partial discharge event, adjusts the association data calculated by the Apriori algorithm. Under continuous training, when the big data processing model obtains the partial discharge signal and completes the analysis, it can directly calculate the photovoltaic power generation network facilities associated with the partial discharge event, or calculate the influence content of the compared photovoltaic device from the partial discharge event of the photovoltaic power generation network. This method enables the present invention to have extremely high intelligent perception and analysis capabilities when monitoring the state of the photovoltaic device.

[0120] The above is only a preferred embodiment of the present invention, and it is not a limitation to the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments according to the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A power generation abnormality warning system for photovoltaic power generation equipment, characterized by: The partial discharge state monitoring system of power photovoltaic equipment includes detecting partial discharge signals by using a space radio frequency partial discharge detection radar, specifically including: Space radio frequency partial discharge detection radar: The space radio frequency partial discharge detection radar can perform partial discharge detection on photovoltaic equipment in a charged state, and can realize long-distance and non-contact detection; Radar signal processing module: The radar signal processing module converts analog signals into digital signals through analog-to-digital conversion, and suppresses noise and interference through beam forming, pulse compression, clutter filtering and Doppler processing technology; Server: The server is equipped with a big data processing model, a radar signal processing module and a data storage module, and receives data from the space radio frequency local discharge detection radar; Big data processing model: The big data processing model is used to obtain basic information of photovoltaic equipment as a basic reference factor for data processing, analyze the location and frequency of partial discharge signals, and analyze the causes of partial discharge in combination with the power consumption of the corresponding location area; Display module: The display module is used to display the specific location where the partial discharge signal appears, so that the photovoltaic power generation system involved in the corresponding photovoltaic device can be visually observed; Data storage module: The data storage module is used to store basic data of each photovoltaic device, signal data received by the detection radar, partial discharge signal processing records, photovoltaic geographic information data, and photovoltaic power generation network data; The invention also includes a power generation abnormality early warning method for photovoltaic power generation equipment, and the power generation abnormality early warning method for photovoltaic power generation equipment includes the following specific steps: S1. First, make preliminary preparations, collect basic information and build a big data model. The preliminary preparation stage specifically includes the following steps: S1.

1. First, select the installation location of the spatial RF partial discharge detection radar according to the location of each photovoltaic device in the photovoltaic power station; S1.2, then collect the basic information of the photovoltaic power generation network within the radar detection range, including: the geographical location, model, service life, rated power, real-time power and maintenance log of the photovoltaic equipment, as well as the photovoltaic power generation network data and the satellite map of the photovoltaic power station, and then input the collected basic information into the data storage module for storage; S1.

3. Build a big data processing model based on an efficient and adaptive online data stream clustering algorithm, combine the basic data of photovoltaic equipment, photovoltaic power generation network data and photovoltaic geographic information data, and integrate the Apriori algorithm into the big data processing model; S1.

4. Obtain the partial discharge signals detected by the space radio frequency partial discharge detection radar that has been put into use as the training set of the big data processing model, and use the training set to train the big data processing model; S1.

5. The big data processing model receives the basic data of the photovoltaic equipment and establishes a geographical location coordinate system based on the photovoltaic power generation network where the photovoltaic equipment is located. The origin of the coordinate system is the location of the space radio frequency partial discharge detection radar. S1.

6. Proportionately process the satellite map of the PV power station and merge it into the coordinate system, and verify the accuracy of the PV power generation network in the PV power generation network and the corresponding location in the map; S1.

7. Set in the server: when a partial discharge signal is obtained, the signal is copied and sent to the radar signal processing module and the data storage module respectively; when the big data processing model obtains the partial discharge signal analysis result, the result is copied and sent to the display module and the data storage module respectively; S2. Then, a partial discharge signal screening test is performed to process, analyze, screen and determine the detected partial discharge signal. The partial discharge signal screening test specifically includes the following steps: S2.

1. Detecting partial discharge signals by using a space radio frequency partial discharge detection radar; S2.2, the radar signal processing module converts the analog signal into a digital signal through analog-to-digital conversion, suppresses noise and interference through beam forming, pulse compression, clutter filtering and Doppler processing technology, and then sends the processed partial discharge signal to the big data processing model; S2.3, the big data processing model analyzes the partial discharge signal, then extracts the location information of the partial discharge signal and the location information of the photovoltaic device, and compares them; S2.

4. Determine the relative relationship between the location of the partial discharge signal and the location of the photovoltaic equipment according to the detection error of the spatial radio frequency partial discharge detection radar for the geographical location information, make a judgment, and then dispatch maintenance personnel to the site for investigation; S2.

5. After the inspection, the inspection log is uploaded to the data storage module, and the inspection log is assigned as the inspection and maintenance log data in the basic data of the corresponding photovoltaic equipment by the big data processing model; S3. Finally, the actual monitoring operation is performed to analyze the partial discharge signal and adjust the monitoring intensity according to the partial discharge intensity, partial discharge frequency and basic data of the photovoltaic equipment. The actual monitoring operation specifically includes the following steps: S3.

1. The big data processing model receives the partial discharge signal, processes the partial discharge signal using an efficient adaptive online data stream clustering algorithm, extracts the location information of the partial discharge signal, and marks the partial discharge signal data in a coordinate system according to the location of the partial discharge signal; S3.

2. Use the Apriori algorithm to analyze the location distribution data of partial discharge signals and the basic data of photovoltaic equipment; S3.

3. Check the location of the partial discharge signal. When the signal error range detected by the radar is within the location of the photovoltaic equipment, it is regarded as a partial discharge signal of the photovoltaic equipment. Otherwise, it is regarded as a partial discharge signal of the photovoltaic power generation network facility. S3.4, then the big data processing model performs photovoltaic equipment partial discharge analysis and other facilities partial discharge analysis on the photovoltaic equipment partial discharge signal and the photovoltaic power generation network facility partial discharge signal respectively; S3.

5. Then the big data processing model combines the local discharge analysis of photovoltaic equipment and the local discharge analysis of other facilities to conduct local discharge analysis of the photovoltaic power generation network, and issues early warnings based on the local discharge intensity, local discharge frequency and basic data of photovoltaic equipment for possible abnormal conditions of local photovoltaic power generation equipment.

2. The abnormal power generation warning system for photovoltaic power generation equipment according to claim 1 is characterized in that: The spatial radio frequency partial discharge detection radar detects the partial discharge signal, and then sends the partial discharge signal to the radar signal processing module for data processing. The processed partial discharge signal data is transmitted to the big data processing model. The big data processing model combines the basic data of photovoltaic equipment, photovoltaic power generation network data, and power supply data to analyze the partial discharge signal position and partial discharge signal strength, and sends the analysis results to the display module for display.

3. The power generation abnormality warning system for photovoltaic power generation equipment according to claim 1 is characterized in that: The data storage module receives basic data of the photovoltaic equipment, and when the space radio frequency partial discharge detection radar receives the partial discharge signal, copies the partial discharge signal data and sends it to the data storage module for storage backup. After the big data processing model completes the processing of the partial discharge signal data, the processing record and result are sent to the data storage module for storage backup.

4. The power generation abnormality warning system for photovoltaic power generation equipment according to claim 1 is characterized in that: In step S2.4, the specific method of determination is: 1) If the position error of the partial discharge signal is within the position range of the photovoltaic equipment, the big data processing model retrieves and exports the basic information of the corresponding photovoltaic equipment, and dispatches maintenance personnel to the site for inspection; 2) If the position error of the partial discharge signal is outside the position range of the photovoltaic equipment, the big data processing model will match the position of the partial discharge signal according to the photovoltaic power generation network, retrieve the basic data of the photovoltaic equipment in the corresponding network, analyze the photovoltaic equipment directly affected by the photovoltaic power generation network corresponding to the partial discharge position, and analyze the weight relationship of the corresponding photovoltaic equipment affected, and dispatch maintenance personnel to check the photovoltaic equipment in descending order of weight.

5. The power generation abnormality warning system for photovoltaic power generation equipment according to claim 1 is characterized in that: In step S3.4, the specific processing method of the partial discharge analysis of photovoltaic equipment and the partial discharge analysis of other facilities is: 1) The big data processing model retrieves the partial discharge signal of photovoltaic equipment, the circuit data in the photovoltaic power generation network data and the basic data of photovoltaic equipment to perform partial discharge analysis on photovoltaic equipment, and then dispatches personnel to repair the photovoltaic equipment; 2) The big data processing model retrieves the partial discharge signals of photovoltaic power generation network facilities, photovoltaic power generation network data, the geographical location of photovoltaic equipment and the real-time power of photovoltaic equipment to analyze the partial discharge of other facilities, and then dispatches personnel to investigate the location of the partial discharge.

Images