Photovoltaic module fault diagnosis device based on characteristic signals
By using a photovoltaic module fault diagnosis device based on characteristic signals and employing FPGA and multispectral imaging technology, accurate diagnosis of photovoltaic module faults has been achieved. This solves the problems of low efficiency and high error in traditional diagnostic techniques, and improves the operating efficiency and reliability of photovoltaic modules.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- CHINA YANGTZE POWER
- Filing Date
- 2025-06-12
- Publication Date
- 2026-05-19
AI Technical Summary
Existing photovoltaic module fault diagnosis technologies rely on traditional manual inspection and basic equipment monitoring, which makes it difficult to quickly and accurately identify complex faults and cannot identify potential faults in real time, resulting in low diagnostic efficiency and high error rate.
A photovoltaic module fault diagnosis device based on feature signals is adopted, including a signal interface, a control module and a display module. FPGA is used to realize IV curve feature extraction. Combined with multispectral imaging and electroluminescence detection, the device achieves accurate diagnosis of high-dimensional imbalanced data through feature extraction unit and fault diagnosis unit.
It enables precise diagnosis of photovoltaic module faults, improves diagnostic efficiency and accuracy, prevents component failures caused by dust accumulation and localized overheating, and enhances the operating efficiency and reliability of photovoltaic modules.
Smart Images
Figure CN224264944U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of photovoltaic module fault diagnosis, and in particular to a photovoltaic module fault diagnosis device based on characteristic signals. Background Technology
[0002] With the transformation and upgrading of the global energy structure, renewable energy, as a clean and environmentally friendly form of green energy, is developing rapidly. Photovoltaic modules, as a crucial component of renewable energy, have become a key solution for addressing energy challenges and climate change. Among these, photovoltaic arrays and inverters, as core components of the system, are highly susceptible to failure, which directly leads to a decrease in system power generation efficiency and impacts the utilization of renewable energy. Therefore, establishing an automated fault diagnosis and intelligent operation and maintenance system for photovoltaic arrays and inverters is of great significance for ensuring the efficient and stable operation of photovoltaic modules.
[0003] Fault detection is a crucial technical means for the safe and reliable operation and maintenance of photovoltaic (PV) power generation systems. Detecting faults in key PV components, such as PV arrays and PV inverters, is particularly important. Furthermore, the high-dimensional and dynamically changing data from PV modules necessitates the establishment of an intelligent diagnostic system capable of processing high-dimensional unbalanced PV data and performing automated fault identification; this is a critical step in the PV diagnostic process.
[0004] However, most existing photovoltaic module fault diagnosis technologies rely on traditional manual inspection and basic equipment monitoring methods, which typically have significant limitations. On one hand, traditional manual inspection methods depend on the experience of on-site personnel and basic equipment diagnostics, making it difficult to accurately identify fault types in a short time. This is especially true for complex fault types, where manual troubleshooting often requires extensive experience accumulation and repeated testing. Furthermore, existing equipment monitoring methods, such as simple voltage and current sampling and periodic inspections, can usually only detect some surface or obvious faults, failing to identify potential, difficult-to-detect faults at an early stage.
[0005] In the field of photovoltaic (PV) module fault diagnosis, the high dimensionality and imbalance of data make traditional diagnostic methods increasingly inefficient. A complex interaction exists between PV modules and inverters, and faults can occur due to a combination of factors, such as module aging, hot spots, and cracks. These problems typically require high-precision data acquisition and analysis capabilities for accurate localization. However, existing technologies often fail to acquire sufficiently rich parameters in real time, and data analysis frequently relies on manual intervention, resulting in low diagnostic efficiency and a large error rate. Summary of the Invention
[0006] To address the aforementioned technical problems, this utility model proposes a photovoltaic module fault diagnosis device based on feature signals. This device can realize fault diagnosis and fault location of photovoltaic modules, improve the accuracy of photovoltaic module fault diagnosis, and increase the operating efficiency of photovoltaic modules.
[0007] The technical solution adopted by this utility model is as follows:
[0008] A photovoltaic module fault diagnosis device based on characteristic signals, the device comprising:
[0009] Signal interface, control module, display module;
[0010] The signal interface connects to the control module, and the control module connects to the display module.
[0011] The signal interface includes a photovoltaic module interface and a sensor interface, which are used to acquire the output characteristics of the photovoltaic module and environmental parameters.
[0012] The control module is used to extract the feature components of photovoltaic module fault data and determine the fault type of photovoltaic module;
[0013] The display module is used to display the fault type of the photovoltaic module.
[0014] The control module includes:
[0015] The feature extraction unit extracts multiple feature components from photovoltaic module fault data;
[0016] The fault diagnosis unit determines the type of photovoltaic module fault;
[0017] Storage unit, storing historical data of photovoltaic modules.
[0018] The feature extraction unit uses an FPGA to extract IV curve features.
[0019] The sensor interface includes:
[0020] Multispectral imaging unit acquires multispectral images of hot spot areas and defects on the surface of photovoltaic modules;
[0021] An electroluminescence (EL) detection unit acquires EL images of the surface of a photovoltaic module.
[0022] The multispectral imaging unit includes a short-wave infrared sensor and a visible light CCD.
[0023] The electroluminescence (EL) detection unit includes a pulse generator and an InGaAs detector;
[0024] A pulse generator is used to apply DC voltage pulses to photovoltaic (PV) modules. Its function is to generate specific voltage pulses and apply these pulse signals to the PV modules. By exciting the electroluminescence phenomenon of the PV modules, the pulse generator causes visible electroluminescent signals to be produced in defective areas (such as cracks or damage) of the PV modules. These signals can be captured and analyzed by subsequent InGaAs detectors for fault detection and location.
[0025] An InGaAs detector is used to capture electroluminescence signals. Its function is to receive the light signal emitted by the photovoltaic module after it is excited by a pulse generator during electroluminescence (EL) detection. This detector is specifically designed to capture electroluminescence images within a wide wavelength range, particularly for detecting microcracks and other surface defects. It can sensitively capture the weak light signal generated by low-voltage pulse excitation. Therefore, it can provide high-precision images for analyzing the module's condition. The hardware components of the sensor interface are as follows... Figure 3 As shown.
[0026] It also includes an outer casing, inside which a control module is installed, and outside which a signal interface and a display module are installed.
[0027] It also includes a flexible photovoltaic charging panel, which is set on the top of the outer shell. The outer shell is equipped with a supercapacitor energy storage system, which is connected to the flexible photovoltaic charging panel and the control module.
[0028] The outer casing is provided with dust removal holes, and a dust guide groove is provided at the bottom of the dust removal holes, extending to the outside of the outer casing. This, together with the cooling fan located on the outside of the outer casing, forms a directional airflow channel.
[0029] The cooling fan is located on the outside of the outer casing, and the air outlet of the cooling fan is equipped with a 45° deflector.
[0030] This utility model discloses a photovoltaic module fault diagnosis device based on characteristic signals, with the following technical advantages:
[0031] 1) The photovoltaic module fault diagnosis device of this utility model can process high-dimensional unbalanced photovoltaic data in photovoltaic modules and realize accurate diagnosis of photovoltaic module faults.
[0032] 2) The cooling fan of the photovoltaic module fault diagnosis device of this utility model forms a directional airflow channel to prevent poor contact of electrical components caused by dust accumulation.
[0033] 3) The cooling fan outlet of the photovoltaic module fault diagnosis device of this utility model is equipped with a 45° guide plate to ensure that the airflow evenly covers the internal heat-generating components and avoids local overheating that could lead to component failure.
[0034] 4) The photovoltaic module fault diagnosis device of this utility model aims to solve the problems of inefficiency and inaccuracy in the current photovoltaic module fault diagnosis technology. By adopting more intelligent signal acquisition and feature extraction technology, it can realize accurate diagnosis and real-time monitoring of photovoltaic module faults, thereby improving the operating efficiency and reliability of photovoltaic modules. Attached Figure Description
[0035] The present invention will be further described below with reference to the accompanying drawings and examples;
[0036] Figure 1 This utility model presents a schematic diagram of the structure of a photovoltaic module fault diagnosis device.
[0037] Figure 2 This is a schematic diagram of the photovoltaic module fault diagnosis of this utility model.
[0038] Figure 3 This is a schematic diagram of the sensor interface structure of the photovoltaic module fault diagnosis device of this utility model.
[0039] The components are: 1-outer shell, 2-shell door, 3-photovoltaic module interface, 4-sensor interface, 5-control module, 6-display module, 7-dust removal hole; 8-handle; 9-dust guide groove; 10-guide plate. Detailed Implementation
[0040] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.
[0041] A photovoltaic module fault diagnosis device based on characteristic signals, the device comprising:
[0042] Signal interface, control module 5, display module 6; the signal interface connects to control module 5, and control module 5 connects to display module 6;
[0043] The signal interface is connected to the display module 6 via the control module 5. The signal interface includes a photovoltaic module interface 3 and a sensor interface 4. The output characteristics of the photovoltaic module and environmental parameters are obtained through the signal interface.
[0044] The control module 5 includes:
[0045] The feature extraction unit extracts multiple feature components from photovoltaic module fault data;
[0046] The fault diagnosis unit determines the type of photovoltaic module fault;
[0047] Storage unit, storing historical data of photovoltaic modules.
[0048] The feature extraction unit extracts features from the sampled signals of the test set and training set in the storage unit, and then sends the processed feature signals to the fault diagnosis unit.
[0049] The feature extraction unit uses an FPGA to extract IV curve features, mainly extracting 12 feature parameters such as Voc, Isc, and Pmax.
[0050] The feature extraction unit extracts the temperature fluctuation features of the hot spot region from the spectral imaging using wavelet transform.
[0051] The feature extraction unit extracts crack shape features from EL imaging through edge detection and corrosion expansion operations.
[0052] The fault diagnosis unit uses the feature signals of historical time windows and the corresponding fault types as inputs from the feature extraction unit to obtain a photovoltaic module fault diagnosis model.
[0053] The display module 6 is used to display the fault type of the photovoltaic module.
[0054] In each time window, the photovoltaic module interface 3 samples the output characteristics of the photovoltaic module, sends the sampling results to the storage unit, and divides them into test set and training set.
[0055] In each time window, the sensor interface 4 samples the data captured by the sensor, sends the sampling results to the storage unit, and divides them into test set and training set.
[0056] The sensor interface 4 includes a multispectral imaging unit, which uses a 0.9-1.7μm short-wave infrared sensor and a visible light CCD to achieve synchronous acquisition, with a spatial resolution of 0.1mm / pixel and a hot spot detection sensitivity ΔT=0.3K.
[0057] The sensor interface 4 includes an electroluminescence (EL) detection unit equipped with a 1200V DC pulse generator and an InGaAs detector, supports adjustable bias voltage from 0-1000V, and has a minimum crack detection length ≤2mm.
[0058] The outer casing 1 is made of aluminum alloy with an anodized surface. The outer casing 1 houses the control module 5, and externally houses the signal interface and display module 6.
[0059] The control module 5 uses an STM32H743 chip, which integrates a feature extraction unit, a fault diagnosis unit, and a storage unit. The feature extraction unit uses an FPGA.
[0060] The flexible photovoltaic charging panel is a SunPower X22-380 ultra-thin double-glass module with dimensions of 380mm×280mm×2mm.
[0061] The flexible photovoltaic charging panel is installed on the top of the outer shell 1, with a conversion efficiency of >23%. The outer shell 1 is equipped with a supercapacitor energy storage system, which is connected to the flexible photovoltaic charging panel and the control module 5.
[0062] The outer casing 1 is provided with a dust removal hole 7 with a diameter of 3mm. A dust guide groove 9 is provided at the bottom of the dust removal hole 7, and the dust guide groove 9 extends to the outside of the outer casing 1. Together with the cooling fan arranged on the outside of the outer casing 1, it forms a directional airflow channel to prevent poor contact of electrical components caused by dust accumulation.
[0063] The cooling fan is located on the outside of the outer casing 1, and the air outlet of the cooling fan is provided with a 45° guide plate 10 to ensure that the airflow evenly covers the internal heat-generating components and avoids local overheating that could lead to component failure.
[0064] The outer housing 1 has a cabinet door 2 installed on its left side; a photovoltaic module interface 3 and a sensor interface 4 are respectively installed on the upper side of the outer housing 1. The photovoltaic module interface 3 uses an MC4 quick connector and is a DC input port. The sensor interface 4 uses a standard M12 waterproof connector and is an analog / digital signal input port. A handle made of glass fiber reinforced plastic is configured in the middle of the upper side of the outer housing 1.
[0065] This utility model discloses a photovoltaic module fault diagnosis device based on characteristic signals, the working process of which is as follows:
[0066] Power switch: Press briefly to start the device, press and hold for 3 seconds to turn it off;
[0067] 1) Deploy the photovoltaic module fault diagnosis device, connect the voltage and current output of the photovoltaic module to the photovoltaic module interface 3, and connect the sensor output to the sensor interface 4 at the same time;
[0068] 2) Sampling: The signal interface completes the acquisition of IV characteristics, multispectral imaging, and EL images;
[0069] 3) Feature extraction: The feature extraction unit uses FPGA to extract IV curve features (extracting 12 feature parameters such as Voc / Isc / Pmax), extracts temperature fluctuation features of hot spot region through wavelet transform for multispectral imaging, and extracts crack shape features of EL imaging through edge detection (Canny operator) and corrosion expansion operation.
[0070] 4) Fault Diagnosis: Features are used as input for fault diagnosis, and the XGBoost network model is used to identify the fault category.
[0071] 5) Result display and transmission: Fault categories and their real-time display via display module 6;
[0072] The diagnostic results are stored synchronously in the storage unit and uploaded to the cloud management platform via a 4G module.
[0073] Figure 2 This is a schematic diagram illustrating the fault diagnosis of the photovoltaic module according to this utility model. From... Figure 2 The workflow of this utility model can be clearly defined. The photovoltaic module interface 3 samples the I / V characteristic curve of the photovoltaic module, and the sensor interface 4 uses a multispectral imaging unit and an electroluminescent EL detection unit to acquire multispectral imaging and EL images. The feature extraction unit uses FPGA to extract the features of the I / V curve. After extracting the temperature fluctuation features of the hot spot area from the multispectral imaging through wavelet transform and extracting the crack shape features from the EL imaging through edge detection and corrosion expansion operation, the features are input into the XGBoost network model in batches. The model can obtain the fault category based on the features. Finally, the fault diagnosis is completed after being displayed by the display module 6.
Claims
1. A photovoltaic module fault diagnosis device based on characteristic signals, characterized in that... The device includes: Signal interface, control module (5), display module (6); The signal interface is connected to the control module (5), and the control module (5) is connected to the display module (6). The signal interface includes a photovoltaic module interface (3) and a sensor interface (4), which are used to obtain the output characteristics of the photovoltaic module and environmental parameters; The control module (5) is used to extract the feature components of photovoltaic module fault data and determine the fault type of photovoltaic module; The display module (6) is used to display the fault type of the photovoltaic module.
2. The photovoltaic module fault diagnosis device based on feature signals according to claim 1, characterized in that: The control module (5) includes: The feature extraction unit extracts multiple feature components from photovoltaic module fault data; Storage unit, storing historical data of photovoltaic modules.
3. The photovoltaic module fault diagnosis device based on feature signals according to claim 2, characterized in that: The feature extraction unit uses an FPGA to extract IV curve features.
4. The photovoltaic module fault diagnosis device based on feature signals according to claim 1, characterized in that: The sensor interface (4) includes: Multispectral imaging unit acquires multispectral images of hot spot areas and defects on the surface of photovoltaic modules; An electroluminescence (EL) detection unit acquires EL images of the surface of a photovoltaic module.
5. The photovoltaic module fault diagnosis device based on feature signals according to claim 4, characterized in that: The multispectral imaging unit includes a short-wave infrared sensor and a visible light CCD.
6. The photovoltaic module fault diagnosis device based on feature signals according to claim 4, characterized in that: The electroluminescence (EL) detection unit includes a pulse generator and an InGaAs detector; A pulse generator is used to apply DC voltage pulses to the photovoltaic module, causing visible electroluminescent signals to be generated in the defective areas of the photovoltaic module; these signals are captured and analyzed by a subsequent InGaAs detector to detect and locate faults. InGaAs detectors are used to receive light signals emitted by photovoltaic modules after being excited by a pulse generator.
7. The photovoltaic module fault diagnosis device based on feature signals according to claim 1, characterized in that: The device also includes an outer housing (1), inside which a control module (5) is installed, and outside the outer housing (1) a signal interface and a display module (6) are installed.
8. The photovoltaic module fault diagnosis device based on feature signals according to claim 7, characterized in that: The device also includes a flexible photovoltaic charging panel, which is set on the top of the outer shell (1). The outer shell (1) is equipped with a supercapacitor energy storage system, which is connected to the flexible photovoltaic charging panel and the control module (5).
9. The photovoltaic module fault diagnosis device based on feature signals according to claim 8, characterized in that: The outer casing (1) is provided with a dust removal hole (7), and a dust guide groove (9) is provided at the bottom of the dust removal hole (7). The dust guide groove (9) extends to the outside of the outer casing (1); it forms a directional airflow channel with the cooling fan arranged on the outside of the outer casing (1).
10. The photovoltaic module fault diagnosis device based on feature signals according to claim 9, characterized in that: The cooling fan is located on the outside of the outer casing (1), and the air outlet of the cooling fan is provided with a 45° guide plate.