Arc discharge self-detection method and system of photovoltaic intelligent arc discharge detection device
Through the CNN convolutional neural network model, the arc stretching faults are automatically judged in the photovoltaic system, which solves the problems of threshold adjustment and detection in complex environments in the existing technology, and achieves high accuracy and adaptability arc stretching detection, ensuring that the photovoltaic system is stable operation and reliability.
Patent Information
- Application Number
- CN202410920981.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-11-08
AI Technical Summary
The arc pull detection method in existing photovoltaic systems requires manual adjustment of thresholds to adapt to the characteristic values of different hardware, which is not universal enough, and complex components and variable environments make it impossible to collect fixed arc pull current values, making it difficult to use mathematical formulas or multiple The threshold determines whether there is an arc pull, and the software design is complex and easy to modify repeatedly.
Using the CNN convolutional neural network model, data cleaning and standardization is used to clean and standardize the data through closed-loop testing of data acquisition, processing and detection results, and the characteristics of voltage and current waveforms are extracted to realize automatic judgment of arc pulling faults. , avoiding threshold adjustment and additional algorithm development.
It improves the accuracy and reliability of detection, adapts to different photovoltaic system configurations and environments, reduces the risks caused by arc pull failures, ensures the continuous and stable operation of the system and overall reliability, and avoids the problem of frequent changes in thresholds.
Smart Images

Figure CN118921006A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of photovoltaics, and in particular relates to an arc self-detection method for a photovoltaic intelligent arc detection device. Background Art
[0002] During the use of photovoltaic products, an arc detection device will be designed to detect whether the product has an arc fault. When collecting the current or voltage entering the inverter, Hall current sensors or CT (current transformer) components are usually used. The purpose of arc self-test is to ensure the reliable connection of the Hall or CT to the chip link. For example, check whether the connector is firmly inserted, whether the internal wire of the Hall or CT is disconnected, and whether the board is poorly welded.
[0003] At present, the common method on the market is to extract characteristic values by sampling and analyzing the current or voltage, and then set the threshold method to compare the collected data with the threshold to judge the result. In the self-test stage, the characteristic values of the arcing current or voltage are sampled and analyzed, and the threshold is set through software methods to judge the detection results. For different hardware, the floating range of the collected current or voltage values may be inconsistent, and the threshold often needs to be adjusted to adapt to the different characteristic values generated by different hardware. The artificial setting of the characteristic value threshold may cause the characteristic value of the collected current to be different depending on the hardware, resulting in different characteristic value thresholds, which is not universal enough and needs to be adjusted manually. After collecting the known arcing data, the software needs to develop additional calculation methods, adjust the threshold and other secondary operations.
[0004] Since the photovoltaic inverter system is composed of complex components and the external environment is changeable, it is impossible to collect a fixed arc current value, and it is impossible to use mathematical formulas or set multiple thresholds to determine whether an arc occurs. As a result, the software design is easy to modify repeatedly and is not easy to solidify. Summary of the invention
[0005] In order to solve the above problems, the purpose of the present invention is to provide an arc self-detection method and system for a photovoltaic intelligent arc detection device.
[0006] The technical solution provided by the present invention is: an arc self-detection method of a photovoltaic intelligent arc detection device, comprising the following steps:
[0007] The program starts, obtains the pre-stored arc data, and outputs it through the first data port;
[0008] Collect output data of the first data port through the second data port, and input the data into the arc detection device after processing;
[0009] If the arc detection device outputs the result of an arc fault, the self-test passes; otherwise, the self-test fails and fault information is reported.
[0010] Preferably, when acquiring pre-stored arcing data, the sampling frequency is 96 kHz, the first data port outputs data at a frequency of 96 kHz, and the second data port collects data at a frequency of 96 kHz.
[0011] Preferably, the data output by the first data port is processed by a digital-to-analog converter and then output, and the data collected by the second data port is processed by an analog-to-digital converter and then input into the arc detection device.
[0012] Preferably, the first data port adopts an i2s interface, and the second data port collects data from the i2s interface and processes the data into a data type that is compatible with the arc detection device.
[0013] Preferably, the first data port is a PWM output port, and the second data port collects data output by the PWM output port and processes the data into a data type that is compatible with the arc detection device.
[0014] Preferably, the arc detection device adopts a CNN convolutional neural network model, wherein the input layer of the CNN convolutional neural network model receives electrical parameters from the photovoltaic inverter as the original input of the model, the convolution layer extracts features from the waveform data through a series of convolution operations, the pooling layer reduces the feature dimension, strengthens the key features extracted by the model, and reduces the amount of calculation, and further features are integrated and analyzed through the fully connected layer to make a final judgment, and the output layer determines whether there is an arcing phenomenon in the final detection result.
[0015] Preferably, the CNN convolutional neural network model input layer receives voltage and / or current waveform data from the photovoltaic inverter, and cleans and standardizes the collected data before inputting into the CNN convolutional neural network model to facilitate processing by the CNN convolutional neural network model.
[0016] Based on the same concept, the present invention also provides an arc self-detection system of a photovoltaic intelligent arc detection device, comprising:
[0017] The arc data acquisition module is used for program startup, acquiring pre-stored arc data, and outputting it through the first data port;
[0018] An arcing data conversion module, used for collecting output data of the first data port through a second data port, and inputting the data into an arcing detection device after processing;
[0019] The arcing self-checking module is used to report fault information if the arcing detection device outputs the result of an arcing fault and the self-checking passes; otherwise, the self-checking fails.
[0020] Based on the same concept, the present invention also provides an electronic device, comprising:
[0021] A memory, the memory being used to store a processing program;
[0022] A processor, wherein when executing the processing program, the processor implements the arc self-detection method of the photovoltaic intelligent arc detection device described in any one of the above items.
[0023] Based on the same concept, the present invention also provides a readable storage medium, on which a processing program is stored. When the processing program is executed by a processor, the arc self-detection method of the photovoltaic intelligent arc detection device described in any one of the above items is implemented.
[0024] Due to the adoption of the above technical solution, the present invention has the following advantages and positive effects compared with the prior art:
[0025] The technical solution of the present invention starts from the stored arcing data, and through a complete data processing link (including data output, collection, processing and generation of detection results), a closed-loop test is implemented to ensure the accuracy and reliability of the entire detection link. It is possible to confirm whether the arcing detection device can correctly identify arcing faults, thereby ensuring the actual operating performance of the device. Regular or irregular self-tests can ensure the continuous and stable operation of the photovoltaic system, reduce the risks caused by arcing faults, and improve the overall reliability of the system. The self-test method can adapt to different photovoltaic system configurations and different working environments, and has good versatility and adaptability. It avoids the problem in the prior art that an additional self-test algorithm is required and the threshold needs to be constantly changed due to the inverter system being composed of complex components and the changing external environment. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings, wherein:
[0027] Figure 1 The present invention is a flow chart of the arc self-detection method of the photovoltaic intelligent arc detection device. DETAILED DESCRIPTION
[0028] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise ratios, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.
[0029] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0030] First embodiment
[0031] like Figure 1 As shown, an arc self-detection method of a photovoltaic intelligent arc detection device comprises the following steps:
[0032] S100: The program starts, and the pre-stored arc data is obtained and outputted through the first data port;
[0033] S200: collecting output data of the first data port through the second data port, and inputting the processed data into an arc detection device;
[0034] S300: If the arc detection device outputs a result that an arc fault has occurred, the self-test passes; otherwise, the self-test fails and fault information is reported.
[0035] The technical solution of this embodiment starts from the stored arcing data, and through a complete data processing link (including data output, collection, processing and generation of detection results), a closed-loop test is implemented to ensure the accuracy and reliability of the entire detection link. It is possible to confirm whether the arc detection device can correctly identify arcing faults, thereby ensuring the actual operating performance of the device. Regular or irregular self-tests can ensure the continuous and stable operation of the photovoltaic system, reduce the risks caused by arcing faults, and improve the overall reliability of the system. The self-test method can adapt to different photovoltaic system configurations and different working environments, and has good versatility and adaptability. It avoids the problem in the prior art that an additional self-test algorithm is required and the threshold needs to be constantly changed due to the inverter system being composed of complex components and the changing external environment.
[0036] Preferably, when acquiring pre-stored arcing data, the sampling frequency is 96 kHz, the first data port outputs data at a frequency of 96 kHz, and the second data port collects data at a frequency of 96 kHz.
[0037] In this embodiment, the output, sampling, processing and other processes related to arc data processing all use a fixed frequency of 96kHz. This is because the CNN convolutional neural network model has a powerful feature extraction capability. The feature values extracted by the CNN convolutional neural network model are relatively fixed and do not need to be adjusted according to the hardware.
[0038] Preferably, the data output by the first data port is processed by a digital-to-analog converter and then output, and the data collected by the second data port is processed by an analog-to-digital converter and then input into the arc detection device.
[0039] The technical solution of this embodiment is essentially to output the arcing frequency band data through a digital-to-analog converter, and then sample it through an analog-to-digital converter and send it to the arc detection device deep neural network model arc detection algorithm to determine whether it is real arcing data, making full use of the principles and functions of digital-to-analog converters and analog-to-digital converters. There is no need to develop additional calculation methods for self-checking, and the existing arc detection device deep neural network model arc detection algorithm can be directly reused for self-checking. In addition, digital-to-analog converters and analog-to-digital converters are functions that conventional chips generally have. The above solution of this embodiment can make full use of the performance of the processing chip and simplify the peripheral interface of the product as much as possible, which is conducive to the universal and reliable design of the product.
[0040] Preferably, the first data port adopts an i2s interface, and the second data port collects data from the i2s interface and processes the data into a data type that is compatible with the arc detection device.
[0041] The technical solution of this embodiment also provides an embodiment in which the first data port adopts an i2s interface, making the concept of the present invention more universal.
[0042] Preferably, the first data port is a PWM output port, and the second data port collects data output by the PWM output port and processes the data into a data type that is compatible with the arc detection device.
[0043] The technical solution of this embodiment also provides an embodiment in which the first data port adopts a PWM output port, making the concept of the present invention more universal.
[0044] Preferably, the arc detection device adopts a CNN convolutional neural network model, wherein the input layer of the CNN convolutional neural network model receives electrical parameters from the photovoltaic inverter as the original input of the model, the convolution layer extracts features from the waveform data through a series of convolution operations, the pooling layer reduces the feature dimension, strengthens the key features extracted by the model, and reduces the amount of calculation, and further features are integrated and analyzed through the fully connected layer to make a final judgment, and the output layer determines whether there is an arcing phenomenon in the final detection result.
[0045] In the technical solution of this embodiment, the arc detection device adopts a CNN convolutional neural network model. Due to its powerful feature extraction capability, the CNN convolutional neural network model can identify subtle arc signals in complex inverter data, thereby improving the detection accuracy. The CNN convolutional neural network model is deployed in the inverter, which can monitor and respond to arc events in real time, reduce the time of fault diagnosis and response, and improve the safety and reliability of the system.
[0046] Preferably, the CNN convolutional neural network model input layer receives voltage and / or current waveform data from the photovoltaic inverter, and cleans and standardizes the collected data before inputting into the CNN convolutional neural network model to facilitate processing by the CNN convolutional neural network model.
[0047] The input layer of this embodiment receives voltage and / or current waveform data from the photovoltaic inverter. The powerful feature extraction capability of the CNN convolutional neural network model can identify subtle arcing signals in complex inverter data, thereby improving the detection accuracy, which is higher than the traditional methods.
[0048] Based on the same concept, the present invention also provides an arc self-detection system of a photovoltaic intelligent arc detection device, comprising:
[0049] The arc data acquisition module is used for program startup, acquiring pre-stored arc data, and outputting it through the first data port;
[0050] An arcing data conversion module, used for collecting output data of the first data port through a second data port, and inputting the data into an arcing detection device after processing;
[0051] The arcing self-checking module is used to report fault information if the arcing detection device outputs the result of an arcing fault and the self-checking passes; otherwise, the self-checking fails.
[0052] The technical solution of this embodiment starts from the stored arcing data, and through a complete data processing link (including data output, collection, processing and generation of detection results), a closed-loop test is implemented to ensure the accuracy and reliability of the entire detection link. It is possible to confirm whether the arc detection device can correctly identify arcing faults, thereby ensuring the actual operating performance of the device. Regular or irregular self-tests can ensure the continuous and stable operation of the photovoltaic system, reduce the risks caused by arcing faults, and improve the overall reliability of the system. The self-test method can adapt to different photovoltaic system configurations and different working environments, and has good versatility and adaptability. It avoids the problem in the prior art that an additional self-test algorithm is required and the threshold needs to be constantly changed due to the inverter system being composed of complex components and the changing external environment.
[0053] Based on the same concept, the present invention also provides an electronic device, comprising:
[0054] A memory, the memory being used to store a processing program;
[0055] A processor, wherein when executing the processing program, the processor implements the arc self-detection method of the photovoltaic intelligent arc detection device described in any one of the above items.
[0056] Based on the same concept, the present invention also provides a readable storage medium, on which a processing program is stored. When the processing program is executed by a processor, the arc self-detection method of the photovoltaic intelligent arc detection device described in any one of the above items is implemented.
[0057] If the arc self-check method of the photovoltaic intelligent arc detection device is implemented in the form of program instructions and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of software. The computer software is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.
[0058] Those skilled in the art can clearly understand that, for the sake of convenience and brevity in description, the identification content specifically executed by the above-described system and device can refer to the corresponding process in the aforementioned method embodiment.
[0059] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the protection scope of the present invention.
Claims
1. A self-detection method for arcing of a photovoltaic intelligent arcing detection device. It is characterized in that The following steps are involved: The program starts, obtains the pre-stored arc data, and outputs it through the first data port; Collect output data of the first data port through the second data port, and input the data into the arc detection device after processing; If the arc detection device outputs the result of an arc fault, the self-test passes; otherwise, the self-test fails and fault information is reported.
2. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 1, It is characterized in that When acquiring pre-stored arc data, the sampling frequency is 96kHz, the first data port outputs data at a frequency of 96kHz, and the second data port collects data at a frequency of 96kHz.
3. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 1, It is characterized in that The data output by the first data port is processed by a digital-to-analog converter and then output, and the data collected by the second data port is processed by an analog-to-digital converter and then input into the arc detection device.
4. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 1, It is characterized in that The first data port adopts an i2s interface, and the second data port collects data from the i2s interface and processes the data into a data type that is compatible with the arc detection device.
5. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 1, It is characterized in that The first data port adopts a PWM output port, and the second data port collects data output by the PWM output port and processes the data into a data type compatible with the arc detection device.
6. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 1, It is characterized in that The arc detection device adopts a CNN convolutional neural network model. The input layer of the CNN convolutional neural network model receives electrical parameters from a photovoltaic inverter as the original input of the model. The convolution layer extracts features from waveform data through a series of convolution operations. The pooling layer reduces feature dimensions, strengthens key features extracted by the model, and reduces the amount of calculation. Further feature integration and analysis are performed through the fully connected layer to make a final judgment. The output layer determines whether there is an arc phenomenon in the final detection result.
7. The arc self-detection method of the photovoltaic intelligent arc detection device according to claim 6, It is characterized in that The CNN convolutional neural network model input layer receives voltage and / or current waveform data from the photovoltaic inverter, and cleans and standardizes the collected data before inputting into the CNN convolutional neural network model to facilitate processing by the CNN convolutional neural network model.
8. An arc self-test system for a photovoltaic intelligent arc detection device. It is characterized in that include: The arc data acquisition module is used for program startup, acquiring pre-stored arc data, and outputting it through the first data port; An arcing data conversion module, used for collecting output data of the first data port through a second data port, and inputting the data into an arcing detection device after processing; The arcing self-checking module is used to report fault information if the arcing detection device outputs the result of an arcing fault and the self-checking passes; otherwise, the self-checking fails.
9. An electronic device, It is characterized in that include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the arc self-detection method of the photovoltaic intelligent arc detection device as described in any one of claims 1 to 7.
10. A readable storage medium, It is characterized in that The readable storage medium stores a processing program, and when the processing program is executed by the processor, the arc self-detection method of the photovoltaic intelligent arc detection device according to any one of claims 1 to 7 is implemented.
Citation Information
Cited By
Arc discharge detection equipment, detection system, arc discharge detection method and storage medium
CN119861271A
Multi-interruption time sequence adjustment method in arc discharge sampling process, arc discharge detection method and photovoltaic system
CN119921670A
