Photovoltaic fault diagnosis method based on partial IV curve
By using a micro inverter to scan part of the IV curve of the photovoltaic module in a photovoltaic power station and combining with the ResNet neural network, the problem of IV curve scanning in a centralized inverter is solved, and the photovoltaic is achieved under continuous connections. The fault diagnosis of components improves the accuracy and efficiency of diagnosis and is suitable for the intelligent operation and maintenance of photovoltaic power plants.
Patent Information
- Application Number
- CN202411071831.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-12
AI Technical Summary
In photovoltaic power stations with centralized inverters, IV curve scanning cannot be performed on a single photovoltaic module or string, resulting in fewer application of fault diagnosis methods based on IV curves, and the need to disconnect the component, resulting in fluctuations in power generation, which cannot be continuously carried out. Get the IV curve when the connection is open.
Using a photovoltaic fault diagnosis method based on partial IV curves, scanning part of the IV curve through a micro inverter, combining photovoltaic module model and environmental information, the curve features are extracted using ResNet neural network to achieve fault identification and positioning, and avoiding short-circuit current points and open circuit voltage points to reduce power loss and fluctuations.
It realizes IV curve scanning of photovoltaic modules while the module is connected continuously, improves the accuracy and efficiency of fault diagnosis, reduces power loss and power generation fluctuations, and is suitable for the intelligent operation and maintenance of photovoltaic power plants.
Smart Images

Figure CN118944596A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a photovoltaic fault diagnosis method based on a partial IV curve, and belongs to the technical field of photovoltaic power station fault diagnosis and intelligent operation and maintenance. Background Art
[0002] With the development of photovoltaic power generation, the demand for photovoltaic power generation operation and maintenance technology has also received strong attention. Due to the influence of factors such as the operating environment, production environment and transportation, photovoltaic modules may fail and be blocked, which will lead to a decrease in the power generation of photovoltaic modules, causing economic losses, and in serious cases, even fire and other safety issues. Therefore, timely detection of faults and confirmation of fault types are of great significance to the stability of photovoltaic power stations.
[0003] The current method of photovoltaic fault diagnosis is mainly based on electrical and image data, where electrical data includes the current, voltage and IV curve data of the photovoltaic module operation, and image data includes visible light images, infrared images, etc. Among the electrical data, the IV curve is one of the data that contains the most information about the status of the photovoltaic module. Compared with other electrical data, most faults can be diagnosed more accurately through the IV curve data.
[0004] Many photovoltaic fault diagnosis methods based on IV curves have been proposed, and most of them have high accuracy. However, since most of the current photovoltaic power stations are traditional centralized inverter photovoltaic power stations, it is impossible to detect individual components or strings. At the same time, scanning the complete IV curve requires the photovoltaic components to be disconnected from other photovoltaic components, resulting in drastic fluctuations in power generation and affecting power generation efficiency. Therefore, in photovoltaic power stations with centralized inverters, due to the difficulty in obtaining IV curves, photovoltaic fault diagnosis methods based on IV curves are rarely used.
[0005] However, with the development of micro inverters, their cost has gradually decreased, and the application of micro inverters has gradually increased. Micro inverters can perform component-level electrical monitoring of photovoltaic power stations and realize IV curve scanning of photovoltaic modules, solving the problem that centralized inverters have difficulty scanning the IV curve of photovoltaic modules. Therefore, the use of IV curves for fault diagnosis in photovoltaic power stations using micro inverters has a good application prospect. However, the problem of disconnecting the connection between photovoltaic modules has not been solved. Therefore, how to obtain the IV curve of photovoltaic modules without disconnecting the connection between modules and realize fault diagnosis of photovoltaic modules is a problem that needs to be considered. Summary of the invention
[0006] Purpose of the invention: The purpose of the present invention is to solve the deficiencies in the prior art. From the perspective of photovoltaic power station fault diagnosis, combined with a micro inverter, a partial IV curve scan of the photovoltaic module is performed to provide a photovoltaic module fault diagnosis method based on a partial IV curve without disconnecting the photovoltaic module.
[0007] Technical solution: A photovoltaic fault diagnosis method based on a partial IV curve of the present invention comprises the following steps:
[0008] S1: Prepare to start scanning the IV curve and record the time when the scan starts, solar irradiance, and module backplane temperature data;
[0009] S2: Calculate the IV curve range that needs to be scanned based on the PV module information and environmental information;
[0010] S3: Scan the IV curve of the photovoltaic module according to the calculated IV curve range, and record the time, solar irradiance, and module backplane temperature data after the scanning is completed;
[0011] S4: Determine whether the scanned IV curve is usable based on the changes in environmental parameters before and after scanning, and discard the one with large changes in environmental parameters;
[0012] S5: Calculate the theoretical IV curve of the scanned component;
[0013] S6: Based on the scanned IV curve and theoretical IV curve environmental data, perform fault detection and fault classification of the component.
[0014] Furthermore, in step S2, the IV curve range to be scanned needs to meet certain constraints. Figure 2 It can be seen that the IV curve is a monotonically decreasing curve. The voltage value can be used to constrain the lower limit of the scanning range, and the current value can be used to constrain the upper limit of the scanning range. The voltage value of the lower limit is max(V / n,2), where V is the voltage value output by the PV module under the MPPT algorithm before scanning, n is the number of bypass diodes of the PV module, and 2 is the minimum value that the voltage can take, which can be adjusted according to the specifications of the PV module. The current value of the upper limit is βI, where β is a coefficient, which is generally recommended to be set to 0.3, and I is the current value output by the PV module under the MPPT algorithm before scanning. In summary, the scanning range of the IV curve is the part where the voltage is greater than max(V / n,2) and the current is greater than βI.
[0015] Furthermore, in step S3, the scanning of part of the IV curve is completed by the micro inverter. The inverter first adjusts the output voltage of the photovoltaic module to the lower limit calculated in S2, then starts to record data and gradually increases the voltage until the current exceeds the constraint condition.
[0016] Furthermore, in step S4, since the IV curve is sensitive to irradiance, the change in irradiance is used to determine the stability of the environment during the scanning period. The irradiance change is greater than 10w / m 2 The data is deprecated.
[0017] Furthermore, in step S5, a single diode model is generally used to describe the photovoltaic module, and an ideal curve of the photovoltaic module can be obtained through the model. The parameters of the single diode model can be determined by the maximum power point, open circuit voltage point and short circuit current point. During scanning, the maximum power point, open circuit voltage point and short circuit current point of the photovoltaic module can be obtained by measuring the environmental parameters:
[0018]
[0019]
[0020] Among them, I sc,ref 、V oc,ref ,I mpp,ref 、V mpp,ref , α I and α v They are the short-circuit current point current, open-circuit voltage point voltage, maximum power point current and voltage, current temperature coefficient and voltage temperature coefficient of the photovoltaic module. These parameters are provided by the photovoltaic module manufacturer. sc is the short-circuit current of the photovoltaic module under the current situation, V oc is the open circuit voltage of the PV module under the current condition, I mpp and V mpp is the maximum power point current and voltage of the photovoltaic module under the current situation, T ref is the standard temperature, the value is 25℃, n 0 is the diode ideal factor under standard conditions, T is the cell temperature of the photovoltaic module, k is the Boltzmann constant, and q is the elementary charge.
[0021] Furthermore, in the above step S6, the collected part of the IV curve is filtered, and the filtered curve is converted into a curve with a length of N and uniform distribution according to the voltage value through interpolation, and the theoretical IV curve is aligned with it.
[0022] Furthermore, in the above step S6, the ResNet method is used to implement fault diagnosis, because ResNet can better solve the gradient explosion problem, and the convolution in the network can automatically extract the eigenvalues in the curve, and finally achieve a good effect.
[0023] Beneficial effects: The present invention provides a photovoltaic fault diagnosis method based on partial IV curves, which is based on scanning partial IV curves with micro-inverters, combined with prior knowledge such as photovoltaic module models, and uses ResNet neural networks to extract curve features to achieve fault identification and location of photovoltaic modules. The method uses micro-inverters to achieve module-level data collection, while only scanning IV curves in partial intervals, avoiding short-circuit current points and open-circuit voltage points, and achieving IV curve scanning without disconnecting the module, reducing power loss and fluctuations caused by IV curve scanning. The method can be used for intelligent operation and maintenance of photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 It is the overall flow chart of the present invention;
[0025] Figure 2 This is the IV and PV curve of a normal photovoltaic module;
[0026] Figure 3 It is the IV curve of the photovoltaic module under shading and the schematic diagram of the scanned part. Specific implementation methods
[0028] The following is a detailed description of the technical solution of the present invention, but the protection scope of the present invention is not limited to the embodiments. The described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0029] like Figure 1 As shown, the present invention provides a photovoltaic fault diagnosis method based on a partial IV curve, comprising the following steps:
[0030] S1: Prepare to start scanning the IV curve and record the time when the scan starts, solar irradiance, and module backplane temperature data.
[0031] S2: Calculate the IV curve range that needs to be scanned based on the PV module information and environmental information.
[0032] In this embodiment, the IV curve range to be scanned needs to meet certain constraints. Figure 2It can be seen that the IV curve is a monotonically decreasing curve. The voltage value can be used to constrain the lower limit of the scanning range, and the current value can be used to constrain the upper limit of the scanning range. The voltage value of the lower limit is max(V / n,2), where V is the voltage value output by the PV module under the MPPT algorithm before scanning, n is the number of bypass diodes of the PV module, and 2 is the minimum value that the voltage can take, which can be adjusted according to the specifications of the PV module. The current value of the upper limit is βI, where β is a coefficient, which is generally recommended to be set to 0.3, and I is the current value output by the PV module under the MPPT algorithm before scanning. In summary, the scanning range of the IV curve is the part where the voltage is greater than max(V / n,2) and the current is greater than βI.
[0033] S3: Scan the IV curve of the photovoltaic module according to the calculated IV curve range, and record the time, solar irradiance, and module backplane temperature data after the scanning is completed.
[0034] In this embodiment, the scanning of part of the IV curve is completed by the micro inverter. The inverter first adjusts the output voltage of the photovoltaic module to the lower limit calculated in S2, then starts to record data and gradually increases the voltage until the current exceeds the constraint condition. The scanning result is as follows Figure 3 shown.
[0035] S4: Determine whether the scanned IV curve is usable based on the changes in environmental parameters before and after the scan. The one with large changes in environmental parameters will be discarded.
[0036] In this embodiment, since the IV curve is sensitive to irradiance, the change in irradiance is used to determine the stability of the environment during the scanning period. The irradiance change is greater than 10w / m 2 The data is deprecated.
[0037] S5: Calculate the theoretical IV curve of the scanned component.
[0038] In this embodiment, a single diode model is used to describe the photovoltaic module, and the ideal curve of the photovoltaic module can be obtained through the model. The parameters of the single diode model can be determined by the maximum power point, open circuit voltage point and short circuit current point. During scanning, the maximum power point, open circuit voltage point and short circuit current point of the photovoltaic module can be obtained by measuring the environmental parameters:
[0039]
[0040]
[0041] Among them, I sc,ref 、V oc,ref ,I mpp,ref 、V mpp,ref , α I and α vThey are the short-circuit current point current, open-circuit voltage point voltage, maximum power point current and voltage, current temperature coefficient and voltage temperature coefficient of the photovoltaic module. These parameters are provided by the photovoltaic module manufacturer. sc is the short-circuit current of the photovoltaic module under the current situation, V oc is the open circuit voltage of the PV module under the current condition, I mpp and V mpp is the maximum power point current and voltage of the photovoltaic module under the current situation, T ref is the standard temperature, the value is 25℃, n 0 is the diode ideal factor under standard conditions, T is the cell temperature of the photovoltaic module, k is the Boltzmann constant, and q is the elementary charge.
[0042] S6: Based on the scanned IV curve and theoretical IV curve environmental data, perform fault detection and fault classification of the component.
[0043] In this embodiment, part of the collected IV curve is filtered, and the filtered curve is converted into a curve with a length of N and uniform distribution according to voltage values through interpolation, and the theoretical IV curve is aligned with it.
[0044] Finally, the ResNet method is used to implement fault diagnosis because ResNet can better solve the gradient explosion problem, and the convolution in the network can automatically extract the eigenvalues in the curve, ultimately achieving a good effect.
[0045] Figure 3 This is a partial IV curve scanning effect diagram provided in an embodiment of the present invention. According to the method described in steps S1-S4, combined with the information provided by the photovoltaic module manufacturer and the current and voltage output during normal operation of the photovoltaic module, a partial IV curve scanning of the photovoltaic module is achieved.
Claims
1. A photovoltaic fault diagnosis method based on partial IV curves, Features: The following steps are included in sequence: S1: Prepare to start scanning the IV curve and record the time when the scan starts, solar irradiance, and module backplane temperature data; S2: Calculate the IV curve range that needs to be scanned based on the PV module information and environmental information; S3: Scan the IV curve of the photovoltaic module according to the calculated IV curve range, and record the time, solar irradiance, and module backplane temperature data after the scanning is completed; S4: Determine whether the scanned IV curve is usable based on the changes in environmental parameters before and after scanning, and discard the one with large changes in environmental parameters; S5: Calculate the theoretical IV curve of the scanned PV module; S6: Based on the scanned IV curve and theoretical IV curve environmental data, perform fault detection and fault classification of the component.
2. A photovoltaic fault diagnosis method based on a partial IV curve according to claim 1, Features: In step S2, the characteristics of the IV curve of the photovoltaic module are used to constrain the lower limit of the scanning range with a voltage value, and the upper limit of the scanning range with a current value; the voltage value of the lower limit is max(V / n,2), where V is the voltage value output under the MPPT algorithm before the photovoltaic module is scanned, n is the number of bypass diodes of the photovoltaic module, and 2 is the minimum value that the voltage can take; the current value of the upper limit is βI, where β is a coefficient, set to 0.3, and I is the current value output under the MPPT algorithm before the photovoltaic module is scanned; the IV curve scanning range is the part where the voltage is greater than max(V / n,2) and the current is greater than βI.
3. A photovoltaic fault diagnosis method based on partial IV curve according to claim 2, Features: In step S3, part of the IV curve of the photovoltaic module is completed by adjusting the output voltage of the photovoltaic module by the micro inverter; the inverter first adjusts the output voltage of the photovoltaic module to the lower limit calculated in step S2, and then starts to record data and gradually increases the voltage until the current exceeds the constraint condition.
4. A photovoltaic fault diagnosis method based on a partial IV curve according to claim 1, Features: In step S4, the change in irradiance is used to determine the stability of the environment during the scanning period. The irradiance change is greater than 10w / m 2 The data is deprecated.
5. A photovoltaic fault diagnosis method based on a partial IV curve according to claim 1, Features: In step S5, a single diode model is used to describe the photovoltaic module, and an ideal curve of the photovoltaic module is obtained through the model; the parameters of the single diode model are determined by the maximum power point, the open circuit voltage point and the short circuit current point; during scanning, the maximum power point, the open circuit voltage point and the short circuit current point of the photovoltaic module are obtained by measuring the environmental parameters: Among them, I sc,ref 、V oc,ref ,I mpp,ref 、V mpp,ref , α I and α v I is the short-circuit current point current, open-circuit voltage point voltage, maximum power point current and voltage, current temperature coefficient and voltage temperature coefficient of the photovoltaic module. sc is the short-circuit current of the photovoltaic module under the current situation, V oc is the open circuit voltage of the PV module under the current condition, I mpp and V mpp is the maximum power point current and voltage of the photovoltaic module under the current situation, T ref is the standard temperature, the value is 25℃, n 0 is the diode ideal factor under standard conditions, T is the cell temperature of the photovoltaic module, k is the Boltzmann constant, and q is the elementary charge.
6. A photovoltaic fault diagnosis method based on partial IV curve according to claim 1, Features: In the step S6, the collected part of the IV curve is filtered, and the filtered curve is converted into a curve with a length of N and uniform distribution according to the voltage value through interpolation, and the theoretical IV curve is aligned with it.
7. A photovoltaic fault diagnosis method based on partial IV curve according to claim 1, Features: In step S6, ResNet is selected to implement fault diagnosis of photovoltaic components.
Citation Information
Cited By
Photovoltaic power generation abnormity early warning method and system based on time sequence prediction model
CN122293032A
Photovoltaic power generation abnormality early warning method and system based on time series prediction model
CN122293032B
