Distributed photovoltaic power station fault diagnosis method and device based on AMI data

Abstract

The application provides a distributed photovoltaic power station fault diagnosis method and device based on AMI data, wherein the method comprises the following steps: constructing a training sample library and a test sample library of the distributed photovoltaic power station; establishing a power generation operation model of the distributed photovoltaic power station according to the training sample library; calculating a linear regression function of the power generation operation model; identifying a fault power station in the test sample library according to the linear regression function; and diagnosing the fault power station by using AMI data. The application can realize rapid identification, accurate positioning and fault cause diagnosis of the distributed photovoltaic power station fault only by using AMI data, so that the efficiency of fault troubleshooting can be improved, and the influence of the power station fault on photovoltaic power generation efficiency can be reduced.

Description

Technical Field

[0001] This invention relates to the field of power plant fault detection technology, specifically to a method for diagnosing faults in distributed photovoltaic power plants based on AMI data and a device for diagnosing faults in distributed photovoltaic power plants based on AMI data. Background Technology

[0002] In recent years, with the depletion of fossil fuels, ecological damage, and global energy security issues becoming increasingly prominent, the development of renewable energy has become an urgent priority. Solar energy, as a representative of new energy sources, possesses natural advantages such as being clean, environmentally friendly, renewable, and sustainable, enabling the rapid development of photovoltaic power generation technology.

[0003] However, photovoltaic (PV) power plants typically operate in harsh outdoor environments for extended periods, inevitably leading to malfunctions. Failure to detect and address these malfunctions promptly can not only disrupt normal operation but also potentially cause accidents, such as fires caused by hot spots in the PV array. Therefore, timely, comprehensive, and accurate monitoring and assessment of the PV power plant's operational status are crucial for optimizing maintenance strategies and achieving large-scale, safe, and efficient grid connection of distributed PV power generation, in order to prevent malfunctions and cascading failures. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for fault diagnosis of distributed photovoltaic power stations based on AMI data. This method enables rapid identification, accurate location, and fault cause diagnosis of distributed photovoltaic power stations using only AMI data, thereby improving the efficiency of fault diagnosis and reducing the impact of power station faults on photovoltaic power generation efficiency.

[0005] The technical solution adopted in this invention is as follows:

[0006] A method for fault diagnosis of distributed photovoltaic power plants based on AMI data includes the following steps: constructing a training sample library and a test sample library for distributed photovoltaic power plants; establishing a power generation operation model for the distributed photovoltaic power plants based on the training sample library; calculating a linear regression function of the power generation operation model; identifying faulty power plants in the test sample library based on the linear regression function; and using AMI data to diagnose the faulty power plants.

[0007] According to an embodiment of the present invention, the fault diagnosis of the faulty power station using AMI data specifically includes the following steps: obtaining voltage data, current data, associated meter power data, and voltage data of meters in the same meter box of the faulty power station based on the AMI data; determining whether the voltage data and current data of the faulty power station are abnormal; if the voltage data is normal and the current data is abnormal, then determining whether the associated meter power data is abnormal; if the associated meter power data is normal, then determining that the fault of the faulty power station is a meter fault; if the associated meter power data is abnormal, then determining that the fault of the faulty power station is an internal fault; if the voltage data is abnormal and the current data is normal, then determining whether the associated meter power data is abnormal. The system checks whether the voltage data is abnormal. If the voltage data of the meters in the same meter box is normal, the fault of the faulty power station is determined to be a measuring meter fault. If the voltage data of the meters in the same meter box is abnormal, the fault of the faulty power station is determined to be an external fault. If both the voltage data and the current data are abnormal, the system checks whether the power data of the associated meters is abnormal. If the power data of the associated meters is normal, the fault of the faulty power station is determined to be a measuring meter fault. If the power data of the associated meters is abnormal, the system checks whether the voltage data of the meters in the same meter box is abnormal. If the voltage data of the meters in the same meter box is normal, the fault of the faulty power station is determined to be an internal fault. If the voltage data of the meters in the same meter box is abnormal, the fault of the faulty power station is determined to be an external fault.

[0008] According to an embodiment of the present invention, the construction of the training sample library and test sample library for distributed photovoltaic power stations specifically includes the following steps: obtaining first-type power station operation information and first-type power station environmental information of normally operating distributed photovoltaic power stations; constructing the training sample library based on the first-type power station operation information and the first-type power station environmental information; obtaining second-type power station operation information and second-type power station environmental information of the distributed photovoltaic power station to be tested; and constructing the test sample library based on the second-type power station operation information and the second-type power station environmental information.

[0009] According to an embodiment of the present invention, the step of establishing the power generation operation model of the distributed photovoltaic power station based on the training sample library specifically includes the following steps: determining the photovoltaic module type of the normally operating distributed photovoltaic power station in the training sample library; classifying the normally operating distributed photovoltaic power stations in the training sample library according to the photovoltaic module type; and establishing the power generation operation model based on the classification results.

[0010] According to one embodiment of the present invention, the power generation operation model is as follows:

[0011] y = f(x1,x2,x3,x4,x5)

[0012] Where y represents the unit power generation of the normally operating distributed photovoltaic power station, x1 represents the average solar irradiance of the power station area, x2 represents the average ambient temperature of the power station, x3 represents the average ambient humidity of the power station, x4 represents the average ground air pressure of the power station, and x5 represents the average ambient wind speed of the power station.

[0013] According to one embodiment of the present invention, the photovoltaic module type includes monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules.

[0014] According to an embodiment of the present invention, the calculation of the linear regression function of the power generation operation model specifically includes the following steps: using linear regression to fit the power generation operation model to obtain the linear regression function of the power generation operation model; and defining the normal power generation threshold of the distributed photovoltaic power station based on the linear regression function.

[0015] According to one embodiment of the present invention, the linear regression function is:

[0016] y = ε + β1x1 + ... + β5x5

[0017] Wherein, ε represents the fitting intercept of the linear regression function fitted by the power generation operation model, β1 represents the regression coefficient of the average solar irradiance of the power station area, β2 represents the regression coefficient of the average ambient temperature of the power station, β3 represents the regression coefficient of the average ambient humidity of the power station, β4 represents the regression coefficient of the average ground air pressure of the power station, and β5 represents the regression coefficient of the average ambient wind speed of the power station.

[0018] According to one embodiment of the present invention, the step of identifying faulty power plants in the test sample library based on the linear regression function specifically includes the following steps: classifying the distributed photovoltaic power plants to be tested in the test sample library according to the linear regression function and the normal power generation threshold; and identifying the faulty power plants in the test sample library based on the classification results.

[0019] A fault diagnosis device for distributed photovoltaic power plants based on AMI data includes: a sample module for constructing a training sample library and a test sample library for distributed photovoltaic power plants; a modeling module for establishing a power generation operation model of the distributed photovoltaic power plant based on the training sample library; a calculation module for calculating a linear regression function of the power generation operation model; an identification module for identifying faulty power plants in the test sample library based on the linear regression function; and a diagnosis module for performing fault diagnosis on the faulty power plants using AMI data.

[0020] The beneficial effects of this invention are:

[0021] This invention enables rapid identification, accurate location, and fault diagnosis of distributed photovoltaic power station faults using only AMI data, thereby improving the efficiency of fault diagnosis and reducing the impact of power station faults on photovoltaic power generation efficiency. Attached Figure Description

[0022] Figure 1 This is a flowchart of a distributed photovoltaic power station fault diagnosis method based on AMI data according to an embodiment of the present invention;

[0023] Figure 2 This is a flowchart illustrating fault diagnosis of a faulty power plant using AMI data, according to an embodiment of the present invention.

[0024] Figure 3 This is a block diagram of a distributed photovoltaic power station fault diagnosis device based on AMI data according to an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Figure 1 This is a flowchart of a distributed photovoltaic power station fault diagnosis method based on AMI data, according to an embodiment of the present invention.

[0027] like Figure 1 As shown in the figure, the distributed photovoltaic power station fault diagnosis method based on AMI data according to an embodiment of the present invention includes the following steps:

[0028] S1, construct the training sample library and test sample library for distributed photovoltaic power stations.

[0029] Specifically, the system can first obtain the first type of power station operation information and the first type of power station environmental information of a normally operating distributed photovoltaic power station. Then, a training sample library can be constructed based on the first type of power station operation information and the first type of power station environmental information. Subsequently, the system can obtain the second type of power station operation information and the second type of power station environmental information of the distributed photovoltaic power station to be tested. Finally, a test sample library can be constructed based on the second type of power station operation information and the second type of power station environmental information.

[0030] The first category of power station operation information for multiple normally operating distributed photovoltaic (PV) power stations may include power station profile information and power station meter information for these multiple normally operating distributed PV power stations; the first category of power station environmental information for multiple normally operating distributed PV power stations may include meteorological environmental information for these multiple normally operating distributed PV power stations; in addition, the second category of power station operation information for the distributed PV power station under test may include power station profile information, power station meter information, associated meter information, and meter information in the same meter box of the distributed PV power station under test in its AMI system; the second category of power station environmental information for the distributed PV power station under test may include meteorological environmental information in the AMI system of the distributed PV power station under test.

[0031] More specifically, the power station profile information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding power station account number, installed capacity, and photovoltaic module type information; the power station meter information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding daily power generation, high-density voltage, current, and power information; the associated meter information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding grid-connected power generation, high-density voltage, current, and power information; the meter information for the same meter box for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding high-density voltage, current, and power information; the meteorological environment information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding average solar irradiance, average ambient temperature, average ambient humidity, average ground air pressure, and average ambient wind speed of the power station area.

[0032] S2, establish a power generation and operation model for distributed photovoltaic power stations based on the training sample library.

[0033] Specifically, the photovoltaic module types of normally operating distributed photovoltaic power stations in the training sample library can be determined first. Then, the normally operating distributed photovoltaic power stations in the training sample library can be classified according to the photovoltaic module types, and a power generation operation model can be established based on the classification results. Among them, the photovoltaic module types can include monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules.

[0034] More specifically, the distributed photovoltaic power stations operating normally in the training sample library can be classified according to the type of photovoltaic module, namely monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules, and a power generation operation model can be established for each type of operating distributed photovoltaic power station.

[0035] The power generation operation model for each type of normally operating distributed photovoltaic power station can be represented as follows:

[0036] y = f(x1,x2,x3,x4,x5)

[0037] Where y represents the unit power generation of each type of normally operating distributed photovoltaic power station, x1 represents the average solar irradiance of the power station area, x2 represents the average ambient temperature of the power station, x3 represents the average ambient humidity of the power station, x4 represents the average ground air pressure of the power station, and x5 represents the average ambient wind speed of the power station.

[0038] S3 is the linear regression function for calculating the power generation operation model.

[0039] Specifically, a linear regression model can be used to fit the power generation operation model to obtain a linear regression function for the power generation operation model, and the normal power generation threshold of the distributed photovoltaic power station can be defined based on the linear regression function.

[0040] More specifically, we can first define X as the matrix composed of meteorological environmental information of various types of normally operating distributed photovoltaic power stations in the training sample library, and Y as the matrix composed of the unit power generation of various types of normally operating distributed photovoltaic power stations (unit power generation of the distributed photovoltaic power station to be tested = power generation of the power station / power station capacity). Then we can obtain:

[0041] X = [x ij ] 5×5

[0042] Where i represents the type of photovoltaic module, and i = 1, 2, ..., 5; j represents the type of meteorological environmental information of the photovoltaic power station, and j = 1, 2, ..., 5;

[0043] Furthermore, the formula can be expanded as follows:

[0044]

[0045] Furthermore, the meteorological environment information matrix X and its inverse matrix Σ of various normally operating distributed photovoltaic power stations can be calculated. -1 :

[0046]

[0047] Furthermore, the regression coefficient β of the linear regression function fitted to the power generation operation model can be calculated:

[0048]

[0049] Furthermore, the intercept ε of the linear regression function fitted to the power generation operation model can be calculated:

[0050]

[0051]

[0052]

[0053] Therefore, the linear regression function of the power generation operation model can be obtained as follows:

[0054] y = ε + β1x1 + ... + β5x5

[0055] Where ε represents the fitting intercept of the linear regression function fitted to the power generation operation model, β1 represents the regression coefficient of the average solar irradiance in the power station area, β2 represents the regression coefficient of the average ambient temperature of the power station, β3 represents the regression coefficient of the average ambient humidity of the power station, β4 represents the regression coefficient of the average ground air pressure of the power station, and β5 represents the regression coefficient of the average ambient wind speed of the power station.

[0056] S4. Identify faulty power plants in the test sample library based on a linear regression function.

[0057] Specifically, the distributed photovoltaic power stations to be tested in the test sample library can be classified and processed according to the linear regression function and the normal power generation threshold, and then the faulty power stations in the test sample library can be identified according to the classification results.

[0058] More specifically, data from the test sample library can be substituted into a linear regression function to obtain the test value of the unit power generation of the distributed photovoltaic power station under test. This test value can then be compared with the actual unit power generation, and the distributed photovoltaic power stations in the test sample library can be classified based on the normal power generation threshold. For example, if the ratio of the test power generation value to the actual power generation value is greater than the normal power generation threshold, the distributed photovoltaic power station under test is judged to be a high-efficiency operating station; if the ratio is less than the normal power generation threshold, it is judged to be a normally operating station; and if the ratio equals the normal power generation threshold, it is judged to be a faulty operating station. This enables differentiated collaborative troubleshooting of power stations in power outage areas, thereby improving the efficiency of fault diagnosis.

[0059] S5 uses AMI data to diagnose faulty power plants.

[0060] Specifically, such as Figure 2 As shown, step S5 above may further include the following steps:

[0061] S501, based on AMI data, obtain voltage data, current data, associated meter power data and meter voltage data in the same box for the faulty power station. Among them, the voltage data and current data of the faulty power station can be obtained from the power station measurement meter data of the faulty power station.

[0062] S502, determine whether the voltage and current data of the faulty power station are abnormal. If the voltage data is normal and the current data is abnormal, proceed to step S503. If the voltage data is abnormal and the current data is normal, proceed to step S506. If both the voltage and current data are abnormal, proceed to step S509.

[0063] S503, determine whether the associated meter power data is abnormal. If the associated meter power data is normal, proceed to step S504; if the associated meter power data is abnormal, proceed to step S505.

[0064] S504, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0065] S505, the fault at the power station is determined to be an internal fault.

[0066] S506, determine whether the voltage data of the meters in the same meter box is abnormal. If the voltage data of the meters in the same meter box is normal, proceed to step S507. If the voltage data of the meters in the same meter box is abnormal, proceed to step S508.

[0067] S507, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0068] S508 indicates that the fault at the power station is an external fault, i.e., a fault in the upstream power grid.

[0069] S509, determine whether the associated meter power data is abnormal. If the associated meter power data is normal, proceed to step S510; if the associated meter power data is abnormal, proceed to step S511.

[0070] S510, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0071] S511, determine whether the voltage data of the meters in the same meter box is abnormal. If the voltage data of the meters in the same meter box is normal, proceed to step S512. If the voltage data of the meters in the same meter box is abnormal, proceed to step S513.

[0072] S512, the fault at the faulty power station is determined to be an internal fault.

[0073] S513 determines that the fault of the faulty power station is an external fault, that is, a fault of the upstream power grid.

[0074] The following will use the training sample library shown in Table 1 and the test sample library shown in Table 2 as examples to illustrate the practical application process of the distributed photovoltaic power station fault diagnosis method based on AMI data of the present invention.

[0075] Table 1

[0076]

[0077]

[0078]

[0079] Table 2

[0080]

[0081] In a specific embodiment of the present invention, the data in Table 1 can be substituted into step S3 above, i.e., in the process of calculating the linear regression function of the power generation operation model, the fitting intercept ε of the fitted linear regression function of the power generation operation model is 0.125, the regression coefficient β1 of the average solar irradiance of the power station area is 0.605, the regression coefficient β2 of the average ambient temperature of the power station is 0.153, the regression coefficient β3 of the average ambient humidity of the power station is 0.336, the regression coefficient β4 of the average ground air pressure of the power station is 0.129, and the regression coefficient β5 of the average ambient wind speed of the power station is 0.146. Therefore, the linear regression function of the power generation operation model can be calculated as follows:

[0082] y = 0.125 + 0.605 × 10 -2 x1+0.153×10 -1 x2+0.336×10 -2 x3+0.129×10 -1 x4+0.146x5.

[0083] Furthermore, the normal power generation threshold of a distributed photovoltaic power station can be defined based on the linear regression function of the power generation operation model. For example, the normal power generation threshold ρ can be defined in the range of [0.3, 0.8].

[0084] Furthermore, it can be based on the normal power generation threshold. ρ To identify whether the distributed photovoltaic power stations under test in Table 2 are faulty power stations, specifically, you can determine whether the distributed photovoltaic power stations under test 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 in Table 2 are within the normal power generation threshold range. ρ The relationship is as follows: The measured power generation value of distributed photovoltaic power stations 3, 5, and 10 / the actual power generation value = normal power generation threshold. ρ The ratios of the tested distributed photovoltaic power stations 3, 5, and 10, which are 0.562, 0.389, and 0.618, all fall within the normal power generation threshold ρ range [0.3, 0.8]. Therefore, the tested distributed photovoltaic power stations 3, 5, and 10 can be identified as faulty power stations.

[0085] Furthermore, the corresponding power meter data (voltage data and current data), associated power meter data, and voltage data of meters in the same box can be obtained from the AMI data of the distributed photovoltaic power stations 3, 5, and 10 under test, as shown in Table 3.

[0086] Table 3

[0087]

[0088] Furthermore, the data in Table 3 can be substituted into step S3 above. That is, in the process of using AMI data to diagnose the fault of the power station, the fault of the distributed photovoltaic power station 3 under test can be diagnosed as a meter fault, the fault of the distributed photovoltaic power station 5 under test is an external fault, that is, a fault of the upstream power grid, and the fault of the distributed photovoltaic power station 10 under test is an internal fault.

[0089] The beneficial effects of this invention are as follows:

[0090] This invention enables rapid identification, accurate location, and fault diagnosis of distributed photovoltaic power station faults using only AMI data, thereby improving the efficiency of fault diagnosis and reducing the impact of power station faults on photovoltaic power generation efficiency.

[0091] Corresponding to the above embodiments of the distributed photovoltaic power station fault diagnosis method based on AMI data, the present invention also proposes a distributed photovoltaic power station fault diagnosis device based on AMI data.

[0092] like Figure 3 As shown, the distributed photovoltaic power station fault diagnosis device based on AMI data according to an embodiment of the present invention includes a sample module 10, a modeling module 20, a calculation module 30, an identification module 40, and a diagnosis module 50. The sample module 10 is used to construct a training sample library and a test sample library for distributed photovoltaic power stations; the modeling module 20 is used to establish a power generation operation model for the distributed photovoltaic power station based on the training sample library; the calculation module 30 is used to calculate the linear regression function of the power generation operation model; the identification module 40 is used to identify faulty power stations in the test sample library based on the linear regression function; and the diagnosis module 50 is used to perform fault diagnosis on the faulty power stations using AMI data.

[0093] In one embodiment of the present invention, the sample module 10 can be specifically used to obtain the first type of power station operation information and the first type of power station environment information of a normally operating distributed photovoltaic power station, and then construct a training sample library based on the first type of power station operation information and the first type of power station environment information, thereby obtaining the second type of power station operation information and the second type of power station environment information of the distributed photovoltaic power station to be tested, and finally constructing a test sample library based on the second type of power station operation information and the second type of power station environment information.

[0094] The first category of power station operation information for multiple normally operating distributed photovoltaic (PV) power stations may include power station profile information and power station meter information for these multiple normally operating distributed PV power stations; the first category of power station environmental information for multiple normally operating distributed PV power stations may include meteorological environmental information for these multiple normally operating distributed PV power stations; in addition, the second category of power station operation information for the distributed PV power station under test may include power station profile information, power station meter information, associated meter information, and meter information in the same meter box of the distributed PV power station under test in its AMI system; the second category of power station environmental information for the distributed PV power station under test may include meteorological environmental information in the AMI system of the distributed PV power station under test.

[0095] More specifically, the power station profile information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding power station account number, installed capacity, and photovoltaic module type information; the power station meter information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding daily power generation, high-density voltage, current, and power information; the associated meter information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding grid-connected power generation, high-density voltage, current, and power information; the meter information for the same meter box for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding high-density voltage, current, and power information; the meteorological environment information for the aforementioned multiple groups of normally operating distributed photovoltaic power stations and distributed photovoltaic power stations under test can all include the corresponding average solar irradiance, average ambient temperature, average ambient humidity, average ground air pressure, and average ambient wind speed of the power station area.

[0096] In one embodiment of the present invention, the modeling module 20 may be specifically used to determine the photovoltaic module types of normally operating distributed photovoltaic power stations in the training sample library, and then classify the normally operating distributed photovoltaic power stations in the training sample library according to the photovoltaic module types, thereby establishing a power generation operation model based on the classification results. The photovoltaic module types may include monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules.

[0097] More specifically, the distributed photovoltaic power stations operating normally in the training sample library can be classified according to the type of photovoltaic module, namely monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules, and a power generation operation model can be established for each type of operating distributed photovoltaic power station.

[0098] The power generation operation model for each type of normally operating distributed photovoltaic power station can be represented as follows:

[0099] y = f(x1,x2,x3,x4,x5)

[0100] Where y represents the unit power generation of each type of normally operating distributed photovoltaic power station, x1 represents the average solar irradiance of the power station area, x2 represents the average ambient temperature of the power station, x3 represents the average ambient humidity of the power station, x4 represents the average ground air pressure of the power station, and x5 represents the average ambient wind speed of the power station.

[0101] In one embodiment of the present invention, the calculation module 30 may be specifically used to obtain the linear regression function of the power generation operation model by using linear regression fitting, and the normal power generation threshold of the distributed photovoltaic power station may be defined according to the linear regression function.

[0102] More specifically, we can first define X as the matrix composed of meteorological environmental information of various types of normally operating distributed photovoltaic power stations in the training sample library, and Y as the matrix composed of the unit power generation of various types of normally operating distributed photovoltaic power stations (unit power generation of the distributed photovoltaic power station to be tested = power generation of the power station / power station capacity). Then we can obtain:

[0103] X = [x ij ] 5×5

[0104] Where i represents the type of photovoltaic module, and i = 1, 2, ..., 5; j represents the type of meteorological environmental information of the photovoltaic power station, and j = 1, 2, ..., 5;

[0105] Furthermore, the formula can be expanded as follows:

[0106]

[0107] Furthermore, the meteorological environment information matrix X and its inverse matrix Σ of various normally operating distributed photovoltaic power stations can be calculated. -1 :

[0108]

[0109] Furthermore, the regression coefficient β of the linear regression function fitted to the power generation operation model can be calculated:

[0110]

[0111] Furthermore, the intercept ε of the linear regression function fitted to the power generation operation model can be calculated:

[0112]

[0113]

[0114]

[0115] Therefore, the linear regression function of the power generation operation model can be obtained as follows:

[0116] y = ε + β1x1 + ... + β5x5

[0117] Where ε represents the fitting intercept of the linear regression function fitted to the power generation operation model, β1 represents the regression coefficient of the average solar irradiance in the power station area, β2 represents the regression coefficient of the average ambient temperature of the power station, β3 represents the regression coefficient of the average ambient humidity of the power station, β4 represents the regression coefficient of the average ground air pressure of the power station, and β5 represents the regression coefficient of the average ambient wind speed of the power station.

[0118] In one embodiment of the present invention, the identification module 40 can be specifically used to classify the distributed photovoltaic power stations under test in the test sample library according to the linear regression function and the normal power generation threshold, and then identify the faulty power stations in the test sample library according to the classification results.

[0119] More specifically, data from the test sample library can be substituted into a linear regression function to obtain the test value of the unit power generation of the distributed photovoltaic power station under test. This test value can then be compared with the actual unit power generation, and the distributed photovoltaic power stations in the test sample library can be classified based on the normal power generation threshold. For example, if the ratio of the test power generation value to the actual power generation value is greater than the normal power generation threshold, the distributed photovoltaic power station under test is judged to be a high-efficiency operating station; if the ratio is less than the normal power generation threshold, it is judged to be a normally operating station; and if the ratio equals the normal power generation threshold, it is judged to be a faulty operating station. This enables differentiated collaborative troubleshooting of power stations in power outage areas, thereby improving the efficiency of fault diagnosis.

[0120] In one embodiment of the present invention, such as Figure 2 As shown, the working process of the diagnostic module 50 may include the following steps:

[0121] S501, based on AMI data, obtain voltage data, current data, associated meter power data and meter voltage data in the same box for the faulty power station. Among them, the voltage data and current data of the faulty power station can be obtained from the power station measurement meter data of the faulty power station.

[0122] S502, determine whether the voltage and current data of the faulty power station are abnormal. If the voltage data is normal and the current data is abnormal, proceed to step S503. If the voltage data is abnormal and the current data is normal, proceed to step S506. If both the voltage and current data are abnormal, proceed to step S509.

[0123] S503, determine whether the associated meter power data is abnormal. If the associated meter power data is normal, proceed to step S504; if the associated meter power data is abnormal, proceed to step S505.

[0124] S504, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0125] S505, the fault at the power station is determined to be an internal fault.

[0126] S506, determine whether the voltage data of the meters in the same meter box is abnormal. If the voltage data of the meters in the same meter box is normal, proceed to step S507. If the voltage data of the meters in the same meter box is abnormal, proceed to step S508.

[0127] S507, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0128] S508 indicates that the fault at the power station is an external fault, i.e., a fault in the upstream power grid.

[0129] S509, determine whether the associated meter power data is abnormal. If the associated meter power data is normal, proceed to step S510; if the associated meter power data is abnormal, proceed to step S511.

[0130] S510, the fault of the faulty power station is determined to be a faulty measuring instrument;

[0131] S511, determine whether the voltage data of the meters in the same meter box is abnormal. If the voltage data of the meters in the same meter box is normal, proceed to step S512. If the voltage data of the meters in the same meter box is abnormal, proceed to step S513.

[0132] S512, the fault at the faulty power station is determined to be an internal fault.

[0133] S513 determines that the fault of the faulty power station is an external fault, that is, a fault of the upstream power grid.

[0134] The following will use the training sample library shown in Table 1 and the test sample library shown in Table 2 as examples to illustrate the practical application process of the distributed photovoltaic power station fault diagnosis device based on AMI data of the present invention.

[0135] Table 1

[0136]

[0137]

[0138] Table 2

[0139]

[0140] In a specific embodiment of the present invention, the data in Table 1 can be substituted into the calculation module 30 to calculate the linear regression function of the power generation operation model. The fitting intercept ε of the linear regression function of the power generation operation model is 0.125, the regression coefficient β1 of the average solar irradiance of the power station area is 0.605, the regression coefficient β2 of the average ambient temperature of the power station is 0.153, the regression coefficient β3 of the average ambient humidity of the power station is 0.336, the regression coefficient β4 of the average ground air pressure of the power station is 0.129, and the regression coefficient β5 of the average ambient wind speed of the power station is 0.146. Therefore, the linear regression function of the power generation operation model can be calculated as follows:

[0141] y = 0.125 + 0.605 × 10 -2 x1+0.153×10 -1 x2+0.336×10 -2 x3+0.129×10 -1 x4+0.146x5.

[0142] Furthermore, the normal power generation threshold of a distributed photovoltaic power station can be defined based on the linear regression function of the power generation operation model. For example, the normal power generation threshold ρ can be defined in the range of [0.3, 0.8].

[0143] Furthermore, the identification module 40 can identify whether the distributed photovoltaic power stations under test in Table 2 are faulty power stations based on the normal power generation threshold ρ. Specifically, the relationship between the distributed photovoltaic power stations under test 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 in Table 2 and the normal power generation threshold ρ can be determined. Among them, the power generation test value / actual power generation value of distributed photovoltaic power stations 3, 5, and 10 is equal to the normal power generation threshold ρ. That is, the ratios of distributed photovoltaic power stations 3, 5, and 10, which are 0.562, 0.389, and 0.618, all fall within the normal power generation threshold ρ range [0.3, 0.8]. Therefore, distributed photovoltaic power stations 3, 5, and 10 can be identified as faulty power stations.

[0144] Furthermore, the corresponding power meter data (voltage data and current data), associated power meter data, and voltage data of meters in the same box can be obtained from the AMI data of the distributed photovoltaic power stations 3, 5, and 10 under test, as shown in Table 3.

[0145] Table 3

[0146]

[0147]

[0148] Furthermore, by substituting the data in Table 3 into the diagnostic module 50, it can be determined that the fault of the distributed photovoltaic power station 3 under test is a meter fault, the fault of the distributed photovoltaic power station 5 under test is an external fault, i.e., a fault of the upstream power grid, and the fault of the distributed photovoltaic power station 10 under test is an internal fault.

[0149] The beneficial effects of this invention are as follows:

[0150] This invention enables rapid identification, accurate location, and fault diagnosis of distributed photovoltaic power station faults using only AMI data, thereby improving the efficiency of fault diagnosis and reducing the impact of power station faults on photovoltaic power generation efficiency.

[0151] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. "A plurality of" means two or more, unless otherwise explicitly specified.

[0152] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0153] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0154] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

Claims

1. A method for fault diagnosis of distributed photovoltaic power plants based on AMI data, characterized in that, Includes the following steps: Construct training and testing sample libraries for distributed photovoltaic power plants; A power generation and operation model for the distributed photovoltaic power station is established based on the training sample library. Calculate the linear regression function of the power generation operation model; The faulty power plants in the test sample library are identified based on the linear regression function. AMI data was used to diagnose the faulty power station. The fault diagnosis of the faulty power station using AMI data specifically includes the following steps: obtaining voltage data, current data, associated meter power data, and voltage data of meters in the same meter box from the faulty power station based on the AMI data; determining whether the voltage and current data of the faulty power station are abnormal; if the voltage data is normal but the current data is abnormal, then determining whether the associated meter power data is abnormal; if the associated meter power data is normal, then determining that the fault of the faulty power station is a meter malfunction; if the associated meter power data is abnormal, then determining that the fault of the faulty power station is an internal fault; if the voltage data is abnormal but the current data is normal, then determining whether the voltage data of meters in the same meter box is abnormal. Abnormalities: If the voltage data of the meters in the same meter box is normal, the fault of the faulty power station is determined to be a measuring meter fault; if the voltage data of the meters in the same meter box is abnormal, the fault of the faulty power station is determined to be an external fault; if both the voltage data and the current data are abnormal, the power data of the associated meters is determined to be abnormal; if the power data of the associated meters is normal, the fault of the faulty power station is determined to be a measuring meter fault; if the power data of the associated meters is abnormal, the voltage data of the meters in the same meter box is determined to be abnormal; if the voltage data of the meters in the same meter box is normal, the fault of the faulty power station is determined to be an internal fault; if the voltage data of the meters in the same meter box is abnormal, the fault of the faulty power station is determined to be an external fault.

2. The method for fault diagnosis of distributed photovoltaic power plants based on AMI data according to claim 1, characterized in that, The construction of the training and testing sample libraries for distributed photovoltaic power stations specifically includes the following steps: Obtain operational information and environmental information of Class I distributed photovoltaic power stations that are operating normally; The training sample library is constructed based on the first type of power plant operation information and the first type of power plant environmental information. Obtain the second-type power station operation information and the second-type power station environmental information of the distributed photovoltaic power station under test; The test sample library is constructed based on the second type of power plant operation information and the second type of power plant environmental information.

3. The method for fault diagnosis of distributed photovoltaic power plants based on AMI data according to claim 2, characterized in that, The process of establishing the power generation and operation model of the distributed photovoltaic power station based on the training sample library specifically includes the following steps: Determine the types of photovoltaic modules in the normally operating distributed photovoltaic power stations in the training sample library; Classify the normally operating distributed photovoltaic power stations in the training sample library according to the photovoltaic module type; The power generation operation model is established based on the classification results.

4. The method for fault diagnosis of distributed photovoltaic power stations based on AMI data according to claim 3, characterized in that, The power generation operation model is as follows: in, y This represents the unit power generation of the normally operating distributed photovoltaic power station. x 1 represents the average solar irradiance in the power station area. x 2 represents the average ambient temperature of the power plant. x 3 indicates the average humidity of the power station environment. x 4 represents the average air pressure at the ground level of the power station. x 5 represents the average wind speed in the power station environment.

5. The method for fault diagnosis of distributed photovoltaic power stations based on AMI data according to claim 4, characterized in that, The photovoltaic module types include monocrystalline silicon photovoltaic modules, polycrystalline silicon photovoltaic modules, amorphous silicon photovoltaic modules, bifacial monocrystalline silicon photovoltaic modules, and bifacial polycrystalline silicon photovoltaic modules.

6. The method for fault diagnosis of distributed photovoltaic power stations based on AMI data according to claim 3, characterized in that, The calculation of the linear regression function of the power generation operation model specifically includes the following steps: The power generation operation model is fitted using linear regression to obtain the linear regression function of the power generation operation model; The normal power generation threshold of the distributed photovoltaic power station is defined based on the linear regression function.

7. The method for fault diagnosis of distributed photovoltaic power stations based on AMI data according to claim 6, characterized in that, The linear regression function is: in, This represents the intercept of the linear regression function fitted to the power generation operation model. The regression coefficient represents the average solar irradiance in the power plant area. The regression coefficient represents the average ambient temperature of the power plant. The regression coefficient represents the average humidity of the power plant environment. The regression coefficient represents the average air pressure at the ground level of the power station. The regression coefficient represents the average wind speed in the power plant environment.

8. The method for fault diagnosis of distributed photovoltaic power plants based on AMI data according to claim 6, characterized in that, The step of identifying faulty power plants in the test sample library based on the linear regression function specifically includes the following steps: The distributed photovoltaic power stations to be tested in the test sample library are classified and processed according to the linear regression function and the normal power generation threshold. Based on the classification results, faulty power plants in the test sample library are identified.

9. A distributed photovoltaic power station fault diagnosis device based on AMI data, implementing the fault diagnosis method for distributed photovoltaic power stations based on AMI data as described in claim 1, characterized in that, include: The sample module is used to construct training and testing sample libraries for distributed photovoltaic power stations. A modeling module, which is used to establish a power generation and operation model of the distributed photovoltaic power station based on the training sample library; A calculation module, which is used to calculate the linear regression function of the power generation operation model; The identification module is used to identify faulty power plants in the test sample library based on the linear regression function; A diagnostic module is used to diagnose the faulty power station using AMI data.

Images