A photovoltaic rack health monitoring system and method

Abstract

The application discloses a kind of photovoltaic support health monitoring system and method, and the vibration audio of photovoltaic support is collected by digital audio collector, and is converted into digital audio signal and is transmitted to audio terminal processing system, then the vibration voiceprint features of photovoltaic support are extracted;Meanwhile, the pressure between photovoltaic support connecting components and the pressure change are monitored by pressure voltage sensor, are converted into voltage signal and are transmitted to voltage terminal processing system, then voltage waveform features are extracted;Again, the vibration voiceprint features and voltage waveform features extracted are superimposed and integrated by double-wave signal processing subsystem, and double-mode monitoring data are obtained;Finally, the health status of photovoltaic support is analyzed based on the double-mode monitoring data.The present application can improve the inspection efficiency and monitoring accuracy of photovoltaic support, thereby improving the safety of photovoltaic power station.

Description

Technical Field

[0001] This invention relates to the field of new energy safety monitoring technology, and in particular to a photovoltaic support health monitoring system and method. Background Technology

[0002] During the long-term operation of a photovoltaic power station, the photovoltaic support system, as the core structure supporting the photovoltaic modules, directly affects the safe and stable operation of the power station. Photovoltaic support systems are exposed to the outdoor environment for extended periods, making them prone to stress deformation, loose bolts, loose modules, hidden cracks in the support system, and unexpected overloading. If these issues are not detected in time, they can create safety hazards for the entire photovoltaic power generation system.

[0003] Currently, health monitoring of photovoltaic (PV) supports mainly relies on manual inspections, robotic inspections, or single-parameter monitoring. Manual inspections depend on the experience and judgment of inspectors, making the results highly subjective and difficult to monitor subtle vibrations or pressure changes in connecting components in real time. This results in low inspection efficiency and cannot cover the continuous monitoring needs of large-scale PV power plants. While inspection robots can save manpower, they cannot provide precise monitoring and judgment, still requiring human intervention. Single-parameter monitoring primarily uses vibration sensors to monitor support vibration or pressure sensors to monitor pressure in connecting components, but it relies on a single signal (such as a simple vibration or pressure signal), making it susceptible to environmental interference and resulting in insufficient monitoring accuracy and a high false alarm rate. In summary, existing methods for PV support health monitoring suffer from low inspection efficiency and insufficient monitoring accuracy. Summary of the Invention

[0004] The purpose of this invention is to provide a photovoltaic support health monitoring system and method to improve the inspection efficiency and monitoring accuracy of photovoltaic supports.

[0005] To achieve the above objectives, the present invention provides a photovoltaic support health monitoring system, including a digital audio acquisition unit, an audio terminal processing subsystem, a pressure-sensitive voltage sensor, a voltage terminal processing subsystem, and a dual-wavelength signal processing subsystem; The digital audio acquisition device is used to acquire the vibration audio of the photovoltaic support and convert it into a digital audio signal, and transmit the digital audio signal to the audio terminal processing subsystem; The audio terminal processing subsystem is used to receive the digital audio signal, extract the vibration acoustic signature of the photovoltaic bracket based on the digital audio signal, and transmit the vibration acoustic signature to the dual-wavelength signal processing subsystem. The pressure-sensitive voltage sensor is used to monitor the pressure change signal between the various connecting components of the photovoltaic bracket, convert the pressure change signal into a voltage signal, and transmit the voltage signal to the voltage terminal processing subsystem. The voltage terminal processing subsystem is used to receive the voltage signal, extract the voltage waveform characteristics of the photovoltaic bracket based on the voltage signal, and transmit the voltage waveform characteristics to the dual-wavelength signal processing subsystem. The dual-wavelength signal processing subsystem is used to receive the vibration acoustic signature and the voltage waveform features, perform feature fusion on the vibration acoustic signature and the voltage waveform features to generate dual-mode monitoring data of the photovoltaic support, and analyze the health status of the photovoltaic support based on the dual-mode monitoring data.

[0006] Optionally, the dual-wavelength signal processing subsystem is further configured to: Acquire bimodal monitoring data from multiple runtime segments and various operating states, and establish a bimodal health database; Hazard modes are identified based on abnormal operating conditions of photovoltaic (PV) brackets to establish a hazard mode database; wherein, the abnormal operating conditions include loose bolts of PV brackets, loose components of PV brackets, deformation of PV brackets, and microcracks in PV brackets.

[0007] Optionally, the analysis of the health status of the photovoltaic support based on the dual-modal monitoring data includes: Acquire real-time generated dual-modal monitoring data; The real-time generated bimodal monitoring data is compared with the data in the bimodal health database and the hazard modality database to obtain the comparison results; Based on the comparison results, a safety status analysis and hazard warning for photovoltaic support structures are conducted.

[0008] Optionally, the step of performing safety status analysis and hazard warning for photovoltaic supports based on the comparison results includes: If the deviation between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is within a preset deviation threshold, then the health status of the photovoltaic support is determined to be safe. If the real-time generated dual-modal monitoring data matches any type of hazard mode in the hazard mode database more than a preset matching degree threshold, the health status of the photovoltaic support is determined to be abnormal and a hazard warning is triggered. If the deviation value between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is not within the preset deviation threshold, and the match between the real-time generated dual-modal monitoring data and any type of hazard mode in the hazard mode database does not exceed the preset matching degree threshold, then it is determined that there is a new hazard mode in the photovoltaic support and a review alarm is triggered.

[0009] Optionally, the digital audio acquisition device includes a digital microphone and a digital audio transmission device; The digital microphone is used to collect the vibration audio of the photovoltaic support and convert it into a digital audio signal, and then transmit the digital audio signal to the digital audio transmitting device. The digital audio transmitting device is used to transmit the digital audio signal to the audio terminal processing subsystem via wireless transmission.

[0010] Optionally, the audio terminal processing subsystem includes a digital audio receiving device and a voiceprint extraction module; The digital audio receiving device is used to receive the digital audio signal; The voiceprint extraction module is used to preprocess the digital audio signal to extract the vibration voiceprint features of the photovoltaic bracket, and transmit the vibration voiceprint features to the dual-wavelength signal processing subsystem.

[0011] Optionally, the preprocessing of the digital audio signal to extract the vibration acoustic signature features of the photovoltaic support based on the digital audio signal includes: The digital audio signal is subjected to noise reduction and segmentation processing; The audio feature spectrum is extracted from the processed digital audio signal to obtain the vibration acoustic characteristics of the photovoltaic support.

[0012] Optionally, the pressure-sensitive voltage sensor may be installed at the connection between the main beam and the column of the photovoltaic bracket, the connection between the main beam and the end purlin, and the connection between the end photovoltaic module and the purlin.

[0013] Optionally, the voltage terminal processing subsystem includes a voltage signal receiving device, a voltage signal amplification device, and a voltage waveform processing module; The voltage signal receiving device is used to receive voltage signals and transmit the voltage signals to the voltage signal amplifying device; The voltage signal amplification device is used to amplify the voltage signal and transmit the amplified voltage signal to the voltage waveform processing module; The voltage waveform processing module is used to extract voltage waveform features based on the amplified voltage signal and transmit the voltage waveform features to the dual-wave signal processing subsystem.

[0014] To achieve the above objectives, the present invention also provides a method for monitoring the health of photovoltaic supports, comprising: The vibration audio of the photovoltaic support is collected and converted into a digital audio signal; The vibration acoustic signature features of the photovoltaic bracket are extracted based on the digital audio signal; Monitor the pressure change signals between various connecting components of the photovoltaic support and convert the pressure change signals into voltage signals; The voltage waveform characteristics of the photovoltaic support are extracted based on the voltage signal; The vibration acoustic signature and the voltage waveform features are fused to obtain dual-mode monitoring data; The health status of the photovoltaic support is analyzed based on the dual-modal monitoring data.

[0015] Compared with existing technologies, the photovoltaic support health monitoring system and method provided by this invention integrates two types of monitoring data: vibration acoustic signature and voltage waveform characteristics. This effectively filters out interference from other noises on the vibration acoustic signature of the photovoltaic support, reducing misjudgments caused by transient pressure changes between the connecting components of the photovoltaic support under sudden conditions. Furthermore, the two feature values ​​mutually verify each other, allowing for the superposition of effective information and improving the accuracy and precision of the detection data. In addition, this invention can digitize and process the safety and health status of the photovoltaic support, significantly improving the efficiency and capability of photovoltaic support safety and health monitoring. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a structural block diagram of a photovoltaic support health monitoring system provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the workflow of a photovoltaic support health monitoring system provided in an embodiment of the present invention. Figure 3 This is a flowchart of a photovoltaic support health monitoring method provided in an embodiment of the present invention. Detailed Implementation

[0018] 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.

[0019] See Figure 1 , Figure 1 This is a structural block diagram of a photovoltaic support health monitoring system provided in an embodiment of the present invention. Figure 1As shown, the photovoltaic support health monitoring system includes a digital audio acquisition unit 1, an audio terminal processing subsystem 2, a pressure-sensitive voltage sensor 3, a voltage terminal processing subsystem 4, and a dual-wavelength signal processing subsystem 5. The digital audio acquisition device 1 is used to acquire the vibration audio of the photovoltaic support and convert it into a digital audio signal, and transmit the digital audio signal to the audio terminal processing subsystem 2; In one alternative embodiment, the digital audio acquisition device 1 includes a digital microphone 101 and a digital audio transmission device 102; The digital microphone 101 is used to collect the vibration audio of the photovoltaic bracket and convert it into a digital audio signal, and transmit the digital audio signal to the digital audio transmitting device 102; The digital audio transmitting device 102 is used to transmit the digital audio signal to the audio terminal processing subsystem 2 via wireless transmission.

[0020] It should be noted that in this embodiment of the invention, the digital audio acquisition device is installed on the main part of the photovoltaic support. The connection part between the main column and the main beam of the photovoltaic support, the vibration part at the far end of the main beam, or the main stress part of the photovoltaic support can be selected. In specific implementation, the part with typical audio characteristics or large audio signal is selected as the main focus, so that the vibration audio of the photovoltaic support can be collected more effectively.

[0021] For example, the digital microphone can be fixed to the photovoltaic support column at a height of 1.2m above the ground, at the midpoint of the main beam, and 5cm from the connection node between the end purlin and the component using stainless steel clips. The high-sensitivity microphone built into the digital microphone can capture the vibration sound waves of the support column in a specific frequency band and convert them into analog electrical signals. These signals can be converted into digital audio signals by a digital-to-analog converter. Then, the digital audio signal data packet containing the device number and timestamp can be transmitted to the digital audio transmitting device via an RS485 interface. The digital audio transmitting device can be installed in a waterproof box on the top of the column and is ultimately received by the digital audio receiving module of the audio terminal processing subsystem.

[0022] It is worth noting that, in this embodiment of the invention, the range of the sound spectrum collected by the digital microphone and the range of the digital spectrum when converted into a digital signal can determine the accuracy of voiceprint feature extraction.

[0023] The audio terminal processing subsystem 2 is used to receive the digital audio signal, extract the vibration acoustic features of the photovoltaic bracket based on the digital audio signal, and transmit the vibration acoustic features to the dual-wave signal processing subsystem 5. In one optional embodiment, the audio terminal processing subsystem 2 includes a digital audio receiving device 201 and a voiceprint extraction module 202; The digital audio receiving device 201 is used to receive the digital audio signal; The voiceprint extraction module 202 is used to preprocess the digital audio signal to extract the vibration voiceprint features of the photovoltaic bracket and transmit the vibration voiceprint features to the dual-wave signal processing subsystem 5.

[0024] Specifically, the preprocessing of the digital audio signal to extract the vibration acoustic signature features of the photovoltaic support based on the digital audio signal includes: The digital audio signal is subjected to noise reduction and segmentation processing; The audio feature spectrum is extracted from the processed digital audio signal to obtain the vibration acoustic characteristics of the photovoltaic support.

[0025] It should be noted that there can be one or more audio terminal processing subsystems.

[0026] For example, the acquired digital audio signal is transmitted via a digital audio transmitter and then received by a digital audio receiver, and can be transmitted to the audio terminal processing subsystem. The received digital audio signal is then preprocessed by a voiceprint extraction module to extract vibration voiceprint features. First, noise reduction is performed using an adaptive wavelet threshold denoising algorithm. A filter template is constructed based on the inherent frequency range of photovoltaic support vibration (e.g., 20Hz-5kHz) to filter out environmental interference such as wind noise and equipment noise, thereby improving the signal-to-noise ratio to over 30dB. Next, segmentation is performed, dividing the denoised audio signal into segments with each second as a time window. At the same time, an energy detection method is used to identify and retain valid segments with energy values ​​exceeding the baseline threshold (the baseline is the signal energy value when there is no vibration), while silent or invalid interference periods are eliminated. Then, audio feature spectrum extraction is performed on the segmented valid signal. The time-domain signal is converted into a frequency-domain signal using a fast Fourier transform, and parameters such as peak frequency, spectral bandwidth, energy ratio, and frequency difference between adjacent frequency bands are calculated. Stable features with high repeatability (such as the main peak frequency of the main beam vibration, the frequency correlation between the column and purlin vibration, etc.) are selected to form vibration acoustic features containing multi-dimensional parameters. Finally, the voiceprint extraction module transmits the extracted vibration voiceprint features to the dual-wave signal processing subsystem in the form of data frames for subsequent superposition and integration with voltage waveform features.

[0027] The pressure-sensitive voltage sensor 3 is used to monitor the pressure change signal between the various connecting components of the photovoltaic bracket, convert the pressure change signal into a voltage signal, and transmit the voltage signal to the voltage terminal processing subsystem 4. In one optional embodiment, the pressure-sensitive voltage sensor 3 is installed at the connection between the main beam and the column of the photovoltaic bracket, the connection between the main beam and the end purlin, and the connection between the end photovoltaic module and the purlin.

[0028] It should be noted that the pressure-sensitive voltage sensor can be either diaphragm-type or contact-type, and can be installed at locations such as the connection between the main beam and the column of the photovoltaic support, the connection between the main beam and the end purlins, and the connection between the end photovoltaic modules and the purlins. The diaphragm-type sensor can be directly press-fitted between the two connecting parts, while the contact-type sensor requires a fixing device to be added at the connection points. The pressure-sensitive voltage sensor is mainly used to monitor the pressure value between the connecting parts and the pressure changes caused by vibration, and converts these changes into voltage signals, which are then transmitted to the voltage terminal processing subsystem.

[0029] The voltage terminal processing subsystem 4 is used to receive the voltage signal, extract the voltage waveform characteristics of the photovoltaic bracket based on the voltage signal, and transmit the voltage waveform characteristics to the dual-wavelength signal processing subsystem 5. In one optional embodiment, the voltage terminal processing subsystem 4 includes a voltage signal receiving device 401, a voltage signal amplification device 402, and a voltage waveform processing module 403. The voltage signal receiving device 401 is used to receive voltage signals and transmit the voltage signals to the voltage signal amplifying device 402; The voltage signal amplification device 402 is used to amplify the voltage signal and transmit the amplified voltage signal to the voltage waveform processing module 403; The voltage waveform processing module 403 is used to extract voltage waveform features based on the amplified voltage signal and transmit the voltage waveform features to the dual-wave signal processing subsystem 5.

[0030] It should be noted that there can be one or more voltage terminal processing systems. After receiving the voltage signal, it needs to be amplified by a voltage signal amplification device to increase the voltage signal value to a processable value. Then, the characteristics of the voltage fluctuation frequency, amplitude, and waveform of the photovoltaic support during normal operation are extracted.

[0031] For example, the voltage signal receiving device can use a multi-channel analog signal acquisition card, connected to each pressure-sensitive voltage sensor (diaphragm type or contact type) via shielded cables, to receive the voltage signals output by the sensors; the voltage signal amplification device can use a differential amplifier circuit composed of high-precision operational amplifiers to amplify the received voltage signal to 0-5V (so that it is within the standard range for subsequent processing); the voltage waveform processing module is a software module running on the industrial control host, which processes the amplified voltage signal. For example, it can first divide the time domain into segments every 0.5 seconds, then obtain the voltage signal time domain parameters in each segment through waveform sampling, and then convert the time domain signal into a frequency domain signal through Fourier transform to extract the core features of the voltage such as frequency, amplitude, waveform, and waveform duration, forming a voltage waveform feature dataset; finally, the voltage waveform processing module packages the voltage waveform feature dataset in a standardized format and transmits it to the dual-wavelength signal processing subsystem through an Ethernet interface for superposition and integration with vibration acoustic features.

[0032] The dual-wavelength signal processing subsystem 5 is used to receive the vibration acoustic signature and the voltage waveform features, perform feature fusion on the vibration acoustic signature and the voltage waveform features to generate dual-mode monitoring data of the photovoltaic support, and analyze the health status of the photovoltaic support based on the dual-mode monitoring data.

[0033] It should be noted that the audio terminal processing subsystem, voltage waveform processing subsystem, and dual-wave signal processing subsystem in the embodiments of the present invention can use any open-source operating system, programming system, or storage system, without affecting the feasibility and uniqueness of the embodiments of the present invention.

[0034] In an optional embodiment, the dual-wavelength signal processing subsystem 5 is further configured to: Acquire bimodal monitoring data from multiple runtime segments and various operating states, and establish a bimodal health database; Hazard modes are identified based on abnormal operating conditions of photovoltaic (PV) brackets to establish a hazard mode database; wherein, the abnormal operating conditions include loose bolts of PV brackets, loose components of PV brackets, deformation of PV brackets, and microcracks in PV brackets.

[0035] For example, statistical analysis can be performed on multiple bimodal monitoring data across various operating periods and states to compile a bimodal health database for photovoltaic (PV) supports during normal operation. Alternatively, during operational monitoring, hazard modes can be automatically or manually calibrated based on on-site issues such as loose bolts, loose components, support deformation, hidden cracks, and other unforeseen circumstances, establishing a hazard mode library. Furthermore, after large-scale data filtering of the bimodal health database and hazard mode library, data models of health change trends and typical hazard problems of PV supports can be extracted. This enables the digitization and data processing of the safety and health status of PV supports, transforming previous fuzzy inspections and experience-based monitoring into data standards.

[0036] It should be noted that external data sources, such as stored historical data or generated standardized data, can also be used as data for the bimodal health database and hazard modality database in this embodiment of the invention.

[0037] Furthermore, the analysis of the health status of the photovoltaic support based on the dual-modal monitoring data includes: Acquire real-time generated dual-modal monitoring data; The real-time generated bimodal monitoring data is compared with the data in the bimodal health database and the hazard modality database to obtain the comparison results; Based on the comparison results, a safety status analysis and hazard warning for photovoltaic support structures are conducted.

[0038] In one optional embodiment, the step of performing safety status analysis and hazard warning for the photovoltaic support based on the comparison results includes: If the deviation between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is within a preset deviation threshold, then the health status of the photovoltaic support is determined to be safe. If the real-time generated dual-modal monitoring data matches any type of hazard mode in the hazard mode database more than a preset matching degree threshold, the health status of the photovoltaic support is determined to be abnormal and a hazard warning is triggered. If the deviation value between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is not within the preset deviation threshold, and the match between the real-time generated dual-modal monitoring data and any type of hazard mode in the hazard mode database does not exceed the preset matching degree threshold, then it is determined that there is a new hazard mode in the photovoltaic support and a review alarm is triggered.

[0039] Preferably, the health status of the photovoltaic support system can be statistically managed. As the data increases and improves, it can provide clear data support for the subsequent improvement, upgrading, acceptance and decommissioning of the photovoltaic support system.

[0040] See Figure 2, Figure 2 This is a flowchart illustrating the workflow of a photovoltaic support health monitoring system provided in an embodiment of the present invention. Figure 2 As shown, firstly, the vibration audio of the photovoltaic support can be collected by a digital audio acquisition device, converted into a digital audio signal, and transmitted to the audio terminal processing system. Then, the vibration acoustic features of the photovoltaic support are extracted. Simultaneously, the pressure and pressure changes between the connecting components of the photovoltaic support are monitored by a pressure-sensitive voltage sensor, converted into a voltage signal, and transmitted to the voltage terminal processing system. Then, the voltage waveform features are extracted. Next, the extracted vibration acoustic features and voltage waveform features are superimposed and integrated by a dual-wavelength signal processing subsystem to obtain dual-modal monitoring data. Based on the dual-modal monitoring data during the Togo operation period and under various operating conditions, a dual-modal health database is established. Based on the on-site emergency situations, hazard modes are calibrated, and a hazard mode database is established. The health monitoring of the photovoltaic support is carried out based on the real-time acquired dual-modal monitoring data, the dual-modal health database, and the hazard mode database.

[0041] In summary, the photovoltaic support health monitoring system provided in this embodiment of the invention integrates two types of monitoring data: vibration acoustic signature and voltage waveform characteristics. This effectively filters out interference from other noises on the vibration acoustic signature of the photovoltaic support, reducing misjudgments caused by transient pressure changes between the connecting components of the photovoltaic support under sudden conditions. Furthermore, the two feature values ​​mutually verify each other, allowing for the superposition of effective information and improving the accuracy and precision of the detection data. Additionally, this embodiment of the invention can digitize and process the safety and health status of the photovoltaic support, significantly improving the efficiency and capability of photovoltaic support safety and health monitoring.

[0042] Based on the above system items, the present invention provides corresponding embodiments of the method items.

[0043] See Figure 3 , Figure 3 This is a flowchart of a photovoltaic support health monitoring method provided in an embodiment of the present invention. The photovoltaic support health monitoring method includes steps S1 to S6: S1. Collect the vibration audio of the photovoltaic support and convert the vibration audio into a digital audio signal; S2. Extract the vibration acoustic signature features of the photovoltaic bracket based on the digital audio signal; S3. Monitor the pressure change signals between various connecting components of the photovoltaic bracket and convert the pressure change signals into voltage signals; S4. Extract the voltage waveform characteristics of the photovoltaic support based on the voltage signal; S5. Perform feature fusion on the vibration acoustic signature and the voltage waveform to obtain dual-mode monitoring data; S6. Analyze the health status of the photovoltaic support based on the dual-modal monitoring data.

[0044] It should be noted that the photovoltaic support health monitoring method provided in this embodiment of the invention is not limited to the device for collecting vibration audio of the photovoltaic support and monitoring pressure change signals between various connecting components of the photovoltaic support. It can be the digital audio acquisition device and pressure-sensitive voltage sensor in the above embodiment, or any other device can be selected for acquisition, processing and transmission. This does not affect the feasibility and uniqueness of this embodiment of the invention.

[0045] It should be noted that, in fact, the photovoltaic support health monitoring system provided in this embodiment of the invention is used to execute all the process steps of the photovoltaic support health monitoring method in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0046] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A photovoltaic support health monitoring system, characterized in that, It includes a digital audio acquisition unit, an audio terminal processing subsystem, a pressure-sensitive voltage sensor, a voltage terminal processing subsystem, and a dual-wavelength signal processing subsystem; The digital audio acquisition device is used to acquire the vibration audio of the photovoltaic support and convert it into a digital audio signal, and transmit the digital audio signal to the audio terminal processing subsystem; The audio terminal processing subsystem is used to receive the digital audio signal, extract the vibration acoustic signature of the photovoltaic bracket based on the digital audio signal, and transmit the vibration acoustic signature to the dual-wavelength signal processing subsystem. The pressure-sensitive voltage sensor is used to monitor the pressure change signal between the various connecting components of the photovoltaic bracket, convert the pressure change signal into a voltage signal, and transmit the voltage signal to the voltage terminal processing subsystem. The voltage terminal processing subsystem is used to receive the voltage signal, extract the voltage waveform characteristics of the photovoltaic bracket based on the voltage signal, and transmit the voltage waveform characteristics to the dual-wavelength signal processing subsystem. The dual-wavelength signal processing subsystem is used to receive the vibration acoustic signature and the voltage waveform features, perform feature fusion on the vibration acoustic signature and the voltage waveform features to generate dual-mode monitoring data of the photovoltaic support, and analyze the health status of the photovoltaic support based on the dual-mode monitoring data.

2. The photovoltaic support health monitoring system as described in claim 1, characterized in that, The dual-wavelength signal processing subsystem is also used for: Acquire bimodal monitoring data from multiple runtime segments and various operating states, and establish a bimodal health database; Hazard modes are identified based on abnormal operating conditions of photovoltaic (PV) brackets to establish a hazard mode database; wherein, the abnormal operating conditions include loose bolts of PV brackets, loose components of PV brackets, deformation of PV brackets, and microcracks in PV brackets.

3. The photovoltaic support health monitoring system as described in claim 2, characterized in that, The analysis of the health status of the photovoltaic support based on the dual-modal monitoring data includes: Acquire real-time generated dual-modal monitoring data; The real-time generated bimodal monitoring data is compared with the data in the bimodal health database and the hazard modality database to obtain the comparison results; Based on the comparison results, a safety status analysis and hazard warning for photovoltaic support structures are conducted.

4. The photovoltaic support health monitoring system as described in claim 3, characterized in that, Based on the comparison results, a safety status analysis and hazard warning for the photovoltaic support system are conducted, including: If the deviation between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is within a preset deviation threshold, then the health status of the photovoltaic support is determined to be safe. If the real-time generated dual-modal monitoring data matches any type of hazard mode in the hazard mode database more than a preset matching degree threshold, the health status of the photovoltaic support is determined to be abnormal and a hazard warning is triggered. If the deviation value between the real-time generated dual-modal monitoring data and the data in the dual-modal health database is not within the preset deviation threshold, and the match between the real-time generated dual-modal monitoring data and any type of hazard mode in the hazard mode database does not exceed the preset matching degree threshold, then it is determined that there is a new hazard mode in the photovoltaic support and a review alarm is triggered.

5. The photovoltaic support health monitoring system as described in claim 1, characterized in that, The digital audio acquisition device includes a digital microphone and a digital audio transmission device; The digital microphone is used to collect the vibration audio of the photovoltaic support and convert it into a digital audio signal, and then transmit the digital audio signal to the digital audio transmitting device. The digital audio transmitting device is used to transmit the digital audio signal to the audio terminal processing subsystem via wireless transmission.

6. The photovoltaic support health monitoring system as described in claim 1, characterized in that, The audio terminal processing subsystem includes a digital audio receiving device and a voiceprint extraction module; The digital audio receiving device is used to receive the digital audio signal; The voiceprint extraction module is used to preprocess the digital audio signal to extract the vibration voiceprint features of the photovoltaic bracket, and transmit the vibration voiceprint features to the dual-wavelength signal processing subsystem.

7. The photovoltaic support health monitoring system as described in claim 6, characterized in that, The preprocessing of the digital audio signal to extract the vibration acoustic signature features of the photovoltaic support based on the digital audio signal includes: The digital audio signal is subjected to noise reduction and segmentation processing; The audio feature spectrum is extracted from the processed digital audio signal to obtain the vibration acoustic characteristics of the photovoltaic support.

8. The photovoltaic support health monitoring system as described in claim 1, characterized in that, The pressure-sensitive voltage sensor is installed at the connection between the main beam and the column of the photovoltaic bracket, the connection between the main beam and the end purlin, and the connection between the end photovoltaic module and the purlin.

9. The photovoltaic support health monitoring system as described in claim 1, characterized in that, The voltage terminal processing subsystem includes a voltage signal receiving device, a voltage signal amplification device, and a voltage waveform processing module; The voltage signal receiving device is used to receive voltage signals and transmit the voltage signals to the voltage signal amplifying device; The voltage signal amplification device is used to amplify the voltage signal and transmit the amplified voltage signal to the voltage waveform processing module; The voltage waveform processing module is used to extract voltage waveform features based on the amplified voltage signal and transmit the voltage waveform features to the dual-wave signal processing subsystem.

10. A method for health monitoring of photovoltaic support structures, characterized in that, include: The vibration audio of the photovoltaic support is collected and converted into a digital audio signal; The vibration acoustic signature features of the photovoltaic bracket are extracted based on the digital audio signal; Monitor the pressure change signals between various connecting components of the photovoltaic support and convert the pressure change signals into voltage signals; The voltage waveform characteristics of the photovoltaic support are extracted based on the voltage signal; The vibration acoustic signature and the voltage waveform features are fused to obtain dual-mode monitoring data; The health status of the photovoltaic support is analyzed based on the dual-modal monitoring data.

Images