Iot-based photovoltaic parking shed current fault detection method and related device

Abstract

The application discloses a photovoltaic parking shed current fault detection method based on an Internet of Things and related devices, and the method comprises the following steps: collecting current data and temperature data in real time through Internet of Things sensors arranged on photovoltaic panels, inverters and charging piles; sending the collected data to an edge computing node for preprocessing, wherein the preprocessing comprises filtering, standardization and resistance-temperature relationship modeling, and a dynamic resistance reference curve is generated; uploading the preprocessed data to a cloud server, combining a long short-term memory model trained based on historical data, and analyzing the matching degree of the current waveform and the dynamic resistance reference curve; in response to detecting abnormal fluctuation of the current waveform or deviation of the dynamic resistance reference curve, a diagnostic instruction containing a fault type and positioning information is generated; after the operation and maintenance center verifies the diagnostic instruction, remote repair or alarm triggering is performed, and operation records are stored in a blockchain, so that the identification accuracy of the fault type is improved and the false positive rate is reduced.

Description

Technical Field

[0001] This application relates to the field of photovoltaic equipment technology, and in particular to a method and related device for detecting current faults in photovoltaic parking sheds based on the Internet of Things. Background Technology

[0002] Photovoltaic parking sheds, by integrating photovoltaic power generation and parking functions, have significant application value in the field of new energy utilization. However, the stability of their current system directly affects power generation efficiency and equipment lifespan. Especially in complex environments, such as high humidity and large temperature differences, abnormal current may cause equipment failure or even safety accidents. Therefore, efficient fault detection methods are urgently needed.

[0003] Currently, the detection of current faults in photovoltaic parking sheds mainly relies on maintenance personnel to judge faults by inspecting equipment operating parameters and appearance on-site, or by using sensors to collect real-time data and compare it with the threshold by setting a fixed current threshold range. If the current exceeds the range, an alarm is triggered. Manual inspection is inefficient, costly, and dependent on personnel experience. Furthermore, the threshold judgment has a high false alarm rate. Summary of the Invention

[0004] The main technical problem addressed in this application is to provide a method and related device for detecting current faults in photovoltaic parking sheds based on the Internet of Things, thereby improving the accuracy of fault type identification and reducing the false alarm rate.

[0005] To address the aforementioned technical issues, this application adopts the following technical solution: A method for detecting current faults in photovoltaic parking sheds based on the Internet of Things (IoT). This method includes: real-time acquisition of current and temperature data using IoT sensors deployed on photovoltaic panels, inverters, and charging piles; sending the acquired data to an edge computing node for preprocessing, including filtering, standardization, and resistance-temperature relationship modeling to generate a dynamic resistance reference curve; uploading the preprocessed data to a cloud server and analyzing the matching degree between the current waveform and the dynamic resistance reference curve using a long short-term memory model trained on historical data; generating a diagnostic command containing fault type and location information in response to detected abnormal fluctuations or deviations in the current waveform from the dynamic resistance reference curve; and after the operation and maintenance center verifies the diagnostic command, performing remote repair or triggering an alarm, and storing the operation record on the blockchain.

[0006] The resistance-temperature relationship modeling includes fitting an exponential relationship between resistance and temperature according to the following formula: ;in, This represents the predicted resistance value at temperature T; Reference temperature The measured resistance value; k is the material coefficient, which is related to the sensitivity of photovoltaic equipment materials to temperature changes. These are compensation parameters based on the device's self-heating.

[0007] The construction of the long short-term memory model includes: setting a dual-channel mechanism before the input layer of the long short-term memory model, which includes a first channel and a second channel. The first channel inputs the historical current waveform sequence, and the second channel inputs the dynamic resistance reference curve. Data alignment between the first and second channels is achieved through timestamp synchronization. The output layer is defined as five types of fault type labels, including: physical aging, chemical corrosion, instantaneous overload, poor contact, and normal state. The cross-entropy loss algorithm is used to calculate the loss of the long short-term memory model.

[0008] The cross-entropy loss algorithm includes the following formula: Where Loss is the loss value, used to measure the difference between the Long Short-Term Memory model and the true label. The smaller the loss value, the more accurate the prediction of the Long Short-Term Memory model. N is the total number of samples participating in training. i is the i-th sample. c represents the c-th fault label, c=1,2,3,4,5; 1 is physical aging, 2 is chemical corrosion, 3 is instantaneous overload, 4 is poor contact, and 5 is normal state. Let i be the true label of the c-th class of the i-th sample. Using one-hot encoding, if sample i belongs to class c, then... ,otherwise ; The probability that the i-th sample belongs to the c-th fault label is predicted by the Long Short-Term Memory model and output through the Softmax layer. The probability of the value is [0,1], and the sum of the probabilities of all fault labels in the same sample is 1.

[0009] The verification of diagnostic instructions includes: the operation and maintenance center generates a first dynamic key containing a timestamp and embeds the diagnostic instructions; after receiving the instructions, the edge computing node encrypts the timestamp using a pre-stored hash function to generate a second dynamic key; if the first dynamic key matches the second dynamic key, the instructions are executed; otherwise, a security alarm is triggered.

[0010] Storing operation records on the blockchain includes: generating maintenance logs based on smart contracts, which contain timestamps, operator-specific identification codes, and equipment status, and writing the log hash values ​​into chaincode based on Hyperledger Fabric.

[0011] The method also includes periodically accessing historical data in the blockchain and assessing the health status of the photovoltaic system by comparing the attenuation rate of the measured resistance with that of the historical resistance.

[0012] To address the aforementioned technical issues, another technical solution adopted in this application is to provide an IoT-based photovoltaic parking shed current fault detection device. This IoT-based photovoltaic parking shed current fault detection device includes: a data acquisition module, a processing module, an analysis module, a response module, and an operation and maintenance and storage module. Specifically, the data acquisition module collects current and temperature data in real time using IoT sensors deployed on photovoltaic panels, inverters, and charging piles; the processing module sends the collected data to an edge computing node for preprocessing, including filtering, standardization, and resistance-temperature relationship modeling to generate a dynamic resistance reference curve; the analysis module uploads the preprocessed data to a cloud server and analyzes the matching degree between the current waveform and the dynamic resistance reference curve using a long short-term memory model trained on historical data; the response module generates a diagnostic command containing fault type and location information in response to detected abnormal fluctuations or deviations in the current waveform from the dynamic resistance reference curve; and the operation and maintenance and storage module, after verifying the diagnostic command at the operation and maintenance center, executes remote repair or triggers an alarm and stores the operation record on the blockchain.

[0013] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide an IoT-based photovoltaic parking shed current fault detection device, which includes: a memory and at least one processor, wherein the memory stores instructions; and at least one processor calls the instructions in the memory to cause the IoT-based photovoltaic parking shed current fault detection device to perform the steps of the IoT-based photovoltaic parking shed current fault detection method as described above.

[0014] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a computer-readable storage medium storing instructions, which, when processed and executed, implement the steps of the Internet of Things-based photovoltaic parking shed current fault detection method as described above.

[0015] Unlike existing technologies, the advantages of this application are as follows: By deploying multimodal IoT sensors on photovoltaic panels, inverters, and charging piles, real-time data on current, temperature, and ambient humidity are collected, covering the full-dimensional operating status of the current system and solving the blind spot problem of traditional single current monitoring; a dynamic resistance reference curve is generated based on the resistance-temperature index relationship model, improving temperature drift adaptability; the matching degree between the current waveform and the dynamic reference curve is analyzed through a cloud-based long short-term memory model, improving the accuracy of fault type identification and reducing the false alarm rate; dynamic keys are embedded in diagnostic commands to ensure the security of command transmission and prevent malicious tampering; and operation and maintenance records are written to the blockchain through smart contracts, realizing the immutable storage of operation timestamps, personnel-specific identification codes, and equipment status, improving data traceability efficiency. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating one implementation method of the photovoltaic parking shed current fault detection method based on the Internet of Things in this application.

[0017] Figure 2 This is a schematic diagram of the structural framework of one embodiment of the photovoltaic parking shed current fault detection device based on the Internet of Things in this application.

[0018] Figure 3 This is a schematic diagram of the structural framework of one embodiment of the photovoltaic parking shed current fault detection device based on the Internet of Things in this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions 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, 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.

[0020] The terms "comprising" and "having," and any variations thereof, used in the embodiments of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0021] To facilitate understanding of this embodiment, a method for detecting current faults in photovoltaic parking sheds based on the Internet of Things, as disclosed in this embodiment of the invention, will first be described in detail. For example... Figure 1 As shown, Figure 1 This is a flowchart illustrating one embodiment of the Internet of Things-based photovoltaic parking shed current fault detection method of this application. The method includes the following steps.

[0022] Step S11: Real-time current and temperature data are collected using IoT sensors deployed on photovoltaic panels, inverters, and charging piles.

[0023] Specifically, during the operation of photovoltaic parking sheds, comprehensive and real-time acquisition of operational data from relevant equipment is required for effective detection of current faults. Internet of Things (IoT) sensors, as key components for data acquisition, are deployed in important locations such as photovoltaic panels, inverters, and charging piles.

[0024] In some specific embodiments, IoT sensors with high-precision sensing technology can be selected. For example, for current data acquisition, a current sensor based on the Hall effect principle can be used, which can accurately measure the current in the circuit and has a fast response speed, capturing instantaneous changes in current in a timely manner. For temperature data acquisition, a thermistor temperature sensor can be used, which accurately measures temperature by sensing changes in resistance value with temperature. These sensors have good stability and reliability, can work continuously in complex environments, and ensure that the acquired data accurately reflects the operating status of the device.

[0025] In various application scenarios, photovoltaic parking sheds may face different usage conditions. In commercial parking lots, vehicle charging needs are frequent, and charging piles are used densely. In this case, IoT sensors need to have high-frequency data acquisition capabilities to monitor the current and temperature changes of charging piles under different charging states in a timely manner. In residential parking lots, the use of photovoltaic parking sheds is relatively dispersed, but may be affected by environmental factors such as large temperature differences between day and night. This requires sensors to be able to adapt to a wide temperature range and accurately collect data. In addition, in photovoltaic parking sheds in industrial plants, due to the complex surrounding electromagnetic environment, sensors also need to have strong anti-interference capabilities to ensure that the collected data is not affected by external interference, providing a reliable basis for subsequent fault detection.

[0026] Step S12: Send the collected data to the edge computing node for preprocessing. Preprocessing includes filtering, standardization, and resistance-temperature relationship modeling to generate a dynamic resistance reference curve.

[0027] Specifically, the collected current and temperature data may contain noise and other interference factors, and the data collected by different sensors differ in terms of dimensions and numerical ranges. Filtering is used to remove noise from the data, making it smoother and more accurate for subsequent analysis; standardization unifies data with different ranges and dimensions to a standard scale, facilitating model processing and comparison.

[0028] In some specific embodiments, the filtering can employ the Kalman filter algorithm. Kalman filtering is an algorithm that uses the state equations of a linear system to optimally estimate the system state based on the system's input and output observation data. When processing data collected from photovoltaic parking sheds, it can effectively predict and correct noise in the data, improving data reliability. For standardization, the Z-score standardization method can be used, which transforms the original data into standard data with a mean of 0 and a standard deviation of 1 by calculating the mean and standard deviation of the data.

[0029] In some application scenarios, when photovoltaic parking sheds are in high-temperature and high-humidity environments, the data collected by sensors fluctuates significantly, and noise increases. In such cases, the Kalman filter algorithm can better handle this fluctuating data and ensure filtering effectiveness. In scenarios where multiple types of sensors are used together, due to the large differences in data collected by different sensors, the Z-score normalization method can effectively unify the data scale, making subsequent analysis more accurate. For photovoltaic equipment using new materials, determining the resistance-temperature relationship model parameters through extensive experiments and generating a dynamic resistance reference curve that conforms to actual conditions helps to more accurately detect current faults.

[0030] In some embodiments, resistance-temperature modeling includes fitting an exponential relationship between resistance and temperature according to the following formula: ;in, This represents the predicted resistance value at temperature T; Reference temperature The measured resistance value; k is the material coefficient, which is related to the sensitivity of photovoltaic equipment materials to temperature changes. These are compensation parameters based on the device's self-heating.

[0031] Specifically, the material coefficient k can be determined as follows: In a laboratory environment, for a specific type of photovoltaic device, multiple sets of temperature-resistance data points are collected within a controllable temperature range, and then the optimal k value is obtained by fitting the data using a nonlinear regression algorithm. This can be obtained through equipment thermal characteristic testing. For example, under normal operating conditions, the difference between the equipment surface temperature and the ambient temperature can be measured to establish a temperature compensation model.

[0032] In some specific embodiments, the nonlinear regression algorithm can employ the least squares method to optimize the k value by minimizing the mean square error between the predicted resistance value and the actual measured value. During data acquisition, the temperature range can be set to -20℃ to 80℃, with one set of data collected every 5℃, for a total of 21 sets of data points, to ensure the applicability and accuracy of the model.

[0033] In some application scenarios, a database of material coefficients (k) can be established for different models of photovoltaic equipment. The corresponding k value can be retrieved directly from the database based on the equipment model, improving modeling efficiency. Meanwhile, It can be dynamically adjusted according to the operating power of the equipment, for example, by establishing... A mapping table between temperature and equipment power enables more accurate temperature compensation.

[0034] Step S13: Upload the preprocessed data to the cloud server, and analyze the matching degree between the current waveform and the dynamic resistance reference curve by combining the long short-term memory model trained with historical data.

[0035] In some specific embodiments, a dual-channel mechanism is set before the input layer of the Long Short-Term Memory (LSTM) model. The first channel inputs the historical current waveform sequence, and the second channel inputs the dynamic resistance reference curve. Data alignment between the two channels is achieved through timestamp synchronization. The output layer is defined with five types of fault labels: physical aging, chemical corrosion, instantaneous overload, poor contact, and normal state, using one-hot encoding. For example, [1,0,0,0,0] represents physical aging. The loss function is calculated using the following cross-entropy loss algorithm formula: Where Loss is the loss value, used to measure the difference between the Long Short-Term Memory model and the true label. The smaller the loss value, the more accurate the prediction of the Long Short-Term Memory model. N is the total number of samples participating in training. i is the i-th sample. c represents the c-th fault label, c=1,2,3,4,5; 1. Physical aging, 2. Chemical corrosion, 3. Instantaneous overload, 4. Poor contact, 5. Normal state. Let i be the true label of the c-th class of the i-th sample. Using one-hot encoding, if sample i belongs to class c, then... ,otherwise ; The probability that the i-th sample belongs to the c-th fault label is predicted by the Long Short-Term Memory model and output through the Softmax layer. The probability of the value is [0,1], and the sum of the probabilities of all fault labels in the same sample is 1.

[0036] When training the Long Short-Term Memory (LSTM) model, the input data consists of historical current waveform sequences and dynamic resistance baseline curves, and the output is a matching score, ranging from 0 to 1. The label data consists of manually labeled fault types, such as normal, physical aging, and chemical corrosion. The loss function is cross-entropy loss, and the optimizer is the Adam algorithm. The learning rate can be set to 0.001, and the number of training iterations can be set to 1000.

[0037] Step S14: In response to the detection of abnormal fluctuations or deviations in the current waveform from the dynamic resistance reference curve, generate a diagnostic command containing fault type and location information.

[0038] Specifically, the quantization threshold for "deviation from the dynamic resistance reference curve" is set to 5%. In other words, if the real-time acquired resistance value deviates from the resistance value at the corresponding point on the dynamic resistance reference curve by more than 5%, that point is determined to be an anomaly. This threshold was determined through analysis and verification of historical fault data, effectively reducing the false alarm rate while ensuring the fault detection rate.

[0039] In some specific embodiments, the anomaly detection threshold can be dynamically adjusted according to the device's operating status. For example, during the device startup phase, the threshold can be appropriately relaxed to 10% to avoid false detections caused by startup current surges. During the stable operation phase of the device, the threshold is restored to 5% to ensure detection sensitivity.

[0040] In some application scenarios, different anomaly detection thresholds can be set for power equipment of varying importance. For critical equipment, the threshold can be set to 3% to increase the stringency of detection; for non-critical equipment, the threshold can be set to 7% to reduce unnecessary alarms. Furthermore, a dynamic threshold adjustment model can be established by combining historical operating data of the equipment to automatically adjust the threshold based on the equipment's health status.

[0041] In some embodiments, the verification of diagnostic instructions includes: the operation and maintenance center generates a first dynamic key containing a timestamp and embeds the diagnostic instructions; after receiving the instructions, the edge computing node encrypts the timestamp using a pre-stored hash function to generate a second dynamic key; if the first dynamic key matches the second dynamic key, the instructions are executed; otherwise, a security alarm is triggered.

[0042] Specifically, to ensure that diagnostic commands are not tampered with during transmission, the operations and maintenance center generates a first dynamic key. The addition of a timestamp gives the key timeliness, recording the exact time the command was generated. This first dynamic key, containing the timestamp, is embedded in the diagnostic command and transmitted along with it to the edge computing node. Upon receiving the command, the edge computing node extracts the timestamp and encrypts it using a pre-stored hash function. A hash function is a special algorithm that can convert data of arbitrary length into a fixed-length hash value, and different data processed by the hash function will almost never produce the same hash value. Here, the edge computing node obtains a second dynamic key by encrypting the timestamp. Then, the generated second dynamic key is compared with the received first dynamic key. If they match perfectly, it means the command has not been modified during transmission, and the edge computing node executes the diagnostic command; otherwise, if they do not match, it means the command may have been tampered with, and the edge computing node triggers a security alarm, notifying relevant personnel to check the command's security.

[0043] In some specific implementations, the operations and maintenance center can use the current system time as a timestamp, for example, accurate to milliseconds. Assuming the current time is 10:10:10:10,000 milliseconds on October 10, 2024, this is used as the timestamp. Then, the SHA-256 hash function is used to generate the first dynamic key. SHA-256 is a widely used hash function that can generate a 256-bit hash value. After receiving the instruction, the edge computing node extracts the timestamp "10:10:10:10,000 milliseconds on October 10, 2024" and also uses the SHA-256 hash function for encryption to obtain the second dynamic key. If the two keys match, for example, both are "abcdef1234567890..." (the actual hash value is a 256-bit hexadecimal string), the instruction is executed; if they do not match, such as the second dynamic key being "ghijkl7890123456...", a security alarm is triggered.

[0044] In some application scenarios, within a distributed photovoltaic parking shed network, multiple edge computing nodes are responsible for monitoring equipment in different areas. When a current fault is detected in a certain area and a diagnostic command is generated, the operation and maintenance center encrypts the command and sends it. Due to the complexity of the network environment, malicious attacks may attempt to tamper with the command. For example, a hacker might want to interfere with the fault repair process by deliberately modifying the diagnostic command. In this case, if the first dynamic key in the command received by the edge computing node does not match the second dynamic key it generated, a security alarm will be triggered. Upon receiving the alarm, the operation and maintenance personnel will immediately check the network security status, trace the command transmission path, investigate whether there is any malicious attack behavior, and resend the correct diagnostic command to ensure the safe and reliable operation of the photovoltaic parking shed fault detection and repair process.

[0045] Step S15: After verifying the diagnostic instructions, the operation and maintenance center executes remote repair or triggers an alarm and stores the operation record to the blockchain.

[0046] Specifically, when the operations and maintenance center receives the diagnostic command verification results from the edge computing node and confirms that the command has not been tampered with and is genuine and valid, it will take corresponding measures based on the fault type and severity in the diagnostic command. If the fault type is one that can be resolved remotely, such as software parameter errors or remote restarts of some devices, the operations and maintenance center will use professional remote control software or systems to send repair commands to the faulty device to achieve remote repair. If the fault is more serious and cannot be resolved remotely, such as hardware damage, the operations and maintenance center will immediately trigger an alarm and notify relevant technical personnel to handle the situation on-site via SMS, email, or a professional operations and maintenance management system. Simultaneously, to ensure the traceability of operations and the security of data, the operations and maintenance center will store this operation record on the blockchain. The operation record includes detailed content of the diagnostic command, verification results, executed operations (specific steps of remote repair or alarm information), operation time, and operator information. The application of blockchain technology ensures that these operation records are encrypted and stored in a distributed ledger, making them tamper-proof and ensuring data integrity and trustworthiness.

[0047] In some specific embodiments, suppose an inverter in a photovoltaic parking shed experiences abnormal current fluctuations. Diagnostic commands indicate that this is caused by incorrect control software parameters in the inverter. After verifying the commands, the maintenance center connects to the inverter via remote control software and, following a pre-defined repair procedure, sends parameter adjustment commands to the inverter, correcting the erroneous parameters to the correct values, thus completing the remote repair. After the repair is complete, the maintenance center compiles the operation record, including the fault occurrence time, diagnostic command content, specific repair commands and execution time, and the operator's employee number. Based on a smart contract, a maintenance log containing this information is generated, and the log hash value is written to the chaincode based on Hyperledger Fabric. Hyperledger Fabric is a common blockchain framework that provides enterprise-level blockchain solutions, ensuring secure data storage and efficient management.

[0048] In some application scenarios, such as photovoltaic parking sheds in urban public parking lots, the high traffic volume and frequent use of photovoltaic equipment lead to a relatively high probability of malfunctions. When a malfunction occurs and cannot be repaired remotely, the operation and maintenance center triggers an alarm to notify technicians. For example, if a charging pile experiences a serious short circuit, after the diagnostic command is verified, the operation and maintenance center sends an alarm SMS and email to the technicians responsible for that area, informing them of the fault location and type. Upon receiving the notification, the technicians go to the site with specialized tools for repair. Throughout the process, the operation and maintenance center stores the operation records on the blockchain. This not only facilitates subsequent retrieval of the fault handling process but also provides data support for equipment maintenance and performance optimization. For instance, by analyzing the operation records in the blockchain, if similar malfunctions are frequently found in charging piles in a certain area, a comprehensive inspection and upgrade of the equipment in that area can be carried out, improving the overall reliability and stability of the photovoltaic parking sheds.

[0049] In some embodiments, storing operation records to the blockchain includes: generating a maintenance log based on a smart contract that includes a timestamp, an operator-specific identification code, and equipment status, and writing the log hash value into chaincode based on Hyperledger Fabric.

[0050] Specifically, a smart contract is a computer program that automatically executes contract terms and is deployed on a blockchain. When operation records need to be stored, the system invokes the smart contract. Timestamps record the precise moment an operation occurs, accurate to the second or even millisecond, ensuring the traceability of the operation sequence. A unique operator identification code, such as an employee ID or identity verification code, is used to identify the operator. Equipment status covers detailed information such as operating parameters and fault states of the equipment before and after the operation. This information is integrated to form a maintenance log. To ensure data security and integrity, the log content is not directly stored on the blockchain; instead, a hash value is calculated. A hash value is a fixed-length string obtained by processing the log content using a specific hash algorithm; it is unique, meaning different log content will generate different hash values. The blockchain platform built on Hyperledger provides chaincode to manage and store data. The calculated log hash value is written to the chaincode, achieving secure storage of operation records on the blockchain.

[0051] In some specific embodiments, suppose a maintenance worker at a photovoltaic parking shed performs maintenance on a photovoltaic panel exhibiting abnormal current. After the operation is completed, the system obtains the current precise time as a timestamp, such as "2024-10-15 14:30:25.123". The maintenance worker's unique identification code is "YW00123". The equipment status record shows the specific data of the abnormal current before the photovoltaic panel's maintenance and the operating parameters after the maintenance restored normal operation. This information is combined into a maintenance log according to the format specified by the smart contract: "[Timestamp: 2024-10-15 14:30:25.123, Operator's unique identification code: YW00123, Equipment status: Abnormal current before maintenance, specific value is...; Current restored to normal after maintenance, operating parameters are...]". Then, the SHA-256 hash algorithm is used to calculate the hash value of this log, resulting in a string similar to the following:

[0052] "56a7f95d3c2c45f88f76d959c8d9a2b37d969358d8676a58f7c9a6f8a56d7e13".

[0053] Finally, the hash value is written into the chaincode based on the Hyperledger Fabric via a smart contract, completing the storage of the operation record.

[0054] In some application scenarios, the operation and maintenance management of large-scale photovoltaic power plants involves numerous photovoltaic devices and frequent operations. For example, during the regular inspection and maintenance of a group of inverters, each operation generates a corresponding operation log. Maintenance logs generated based on smart contracts can record detailed information such as the time of each inspection, the operator, and changes in the inverter's performance parameters before and after the inspection. These operation records are securely stored after the log hash value is written to the chaincode of the Hyperledger Fabric. When it is necessary to query the maintenance history of a specific inverter later, the corresponding hash value can be quickly retrieved through the blockchain, and the operation record can be traced back based on the hash value. This is of great significance for the full lifecycle management of equipment, fault tracing, and responsibility determination. If a serious fault occurs in an inverter, by viewing the operation records on the blockchain, the previous maintenance situation can be clearly understood, determining whether there were any improper operations or untimely maintenance issues, providing strong evidence for resolving the fault and improving the operation and maintenance process.

[0055] In some embodiments, the method further includes: periodically retrieving historical data from the blockchain and assessing the health status of the photovoltaic system by comparing the decay rate of the measured resistance with that of the historical resistance.

[0056] Specifically, historical data is periodically retrieved from the blockchain, containing resistance measurements of the photovoltaic system at different points in time. The measured resistance, on the other hand, is the actual resistance value calculated at the current moment using data collected by sensors deployed on the photovoltaic panels, inverters, and charging piles. The degradation rate is typically calculated using the formula: Degradation Rate = (Historical Resistance - Measured Resistance) / Historical Resistance × 100%. Analyzing this degradation rate helps determine the health status of the photovoltaic system. If the degradation rate is within a reasonable range, it indicates that the photovoltaic system is operating relatively stably; if the degradation rate exceeds the normal range, it may indicate potential problems with the photovoltaic system, such as equipment aging or accelerated corrosion.

[0057] In some specific embodiments, historical data from the blockchain is periodically retrieved on a weekly basis. Assuming a photovoltaic panel had a historical resistance of 10Ω a week ago and a measured resistance of 10.3Ω this week, the attenuation rate is calculated using the above formula: attenuation rate = (10-10.3) / 10×100% = -3%. By analyzing historical data and attenuation rates from a large number of similar photovoltaic panels, the normal range for attenuation rate for this type of photovoltaic panel is determined to be between -5% and 5%. Since the calculated attenuation rate of -3% is within the normal range, it can be preliminarily determined that the photovoltaic panel is currently in a healthy operating state.

[0058] In some application scenarios, large-scale photovoltaic (PV) power plants employ a vast number of PV devices. Regularly assessing the health of the PV system allows for the timely detection of potential faults, enabling proactive maintenance and preventing large-scale failures. For example, analysis might reveal that the degradation rates of multiple PV panels in a certain area are close to or exceed the upper limit of the normal range. This could indicate common problems in that area, such as abnormal sunlight conditions or increased environmental corrosion. Maintenance personnel can then use these assessment results to conduct focused inspections of that area, promptly implementing protective measures or replacing aging equipment to ensure the stable operation of the entire PV system, improve power generation efficiency, and reduce maintenance costs.

[0059] In the above solution, multimodal IoT sensors are deployed on photovoltaic panels, inverters, and charging piles to collect current, temperature, and ambient humidity data in real time, covering the full-dimensional operating status of the current system and solving the blind spot problem of traditional single current monitoring. A dynamic resistance reference curve is generated based on the resistance-temperature index relationship model to improve temperature drift adaptability. The matching degree between the current waveform and the dynamic reference curve is analyzed by a cloud-based long short-term memory model to improve the accuracy of fault type identification and reduce the false alarm rate. Dynamic keys are embedded in diagnostic commands to ensure the security of command transmission and prevent malicious tampering. Operation and maintenance records are written to the blockchain through smart contracts to realize operation timestamps, personnel-specific identification codes, and immutable evidence of equipment status, thereby improving data traceability efficiency.

[0060] Please see Figure 2 , Figure 2 This is a schematic diagram of the structural framework of one embodiment of the photovoltaic parking shed current fault detection device based on the Internet of Things, as described in this application. Figure 2 As shown, the IoT-based photovoltaic parking shed current fault detection device 20 includes: a data acquisition module 21, a processing module 22, an analysis module 23, a response module 24, and an operation and maintenance and storage module 25. Specifically, the data acquisition module 21 collects current and temperature data in real time using IoT sensors deployed on photovoltaic panels, inverters, and charging piles; the processing module 22 sends the collected data to an edge computing node for preprocessing, including filtering, standardization, and resistance-temperature relationship modeling to generate a dynamic resistance reference curve; the analysis module 23 uploads the preprocessed data to a cloud server and analyzes the matching degree between the current waveform and the dynamic resistance reference curve using a long short-term memory model trained on historical data; the response module 24 generates diagnostic instructions containing fault type and location information in response to detected abnormal fluctuations or deviations in the current waveform from the dynamic resistance reference curve; and the operation and maintenance and storage module 25 verifies the diagnostic instructions at the operation and maintenance center, executes remote repairs or triggers alarms, and stores the operation records on the blockchain.

[0061] In some embodiments, the processing module 22 performs resistance-temperature relationship modeling by fitting an exponential relationship between resistance and temperature according to the following formula: ;in, This represents the predicted resistance value at temperature T; Reference temperature The measured resistance value; k is the material coefficient, which is related to the sensitivity of photovoltaic equipment materials to temperature changes. These are compensation parameters based on the device's self-heating.

[0062] In some embodiments, constructing a long short-term memory model includes: setting a dual-channel mechanism before the input layer of the long short-term memory model, the dual-channel mechanism including a first channel and a second channel, the first channel inputting a historical current waveform sequence, and the second channel inputting a dynamic resistance reference curve, and aligning the data of the first channel and the second channel through timestamp synchronization; the output layer is defined as five types of fault type labels, including: physical aging, chemical corrosion, instantaneous overload, poor contact, and normal state; and the loss of the long short-term memory model is calculated using a cross-entropy loss algorithm.

[0063] In some embodiments, the cross-entropy loss algorithm used by the processing module 22 includes the following formula: Where Loss is the loss value, used to measure the difference between the Long Short-Term Memory model and the true label. The smaller the loss value, the more accurate the prediction of the Long Short-Term Memory model. N is the total number of samples participating in training. i is the i-th sample. c represents the c-th fault label, c=1,2,3,4,5; 1 is physical aging, 2 is chemical corrosion, 3 is instantaneous overload, 4 is poor contact, and 5 is normal state. Let i be the true label of the c-th class of the i-th sample. Using one-hot encoding, if sample i belongs to class c, then... ,otherwise ; The probability that the i-th sample belongs to the c-th fault label is predicted by the Long Short-Term Memory model and output through the Softmax layer. The probability of the value is [0,1], and the sum of the probabilities of all fault labels in the same sample is 1.

[0064] In some embodiments, the verification of diagnostic instructions by the operation and maintenance and storage module 25 includes: the operation and maintenance center generating a first dynamic key containing a timestamp and embedding a diagnostic instruction; after receiving the instruction, the edge computing node encrypts the timestamp using a pre-stored hash function to generate a second dynamic key; if the first dynamic key matches the second dynamic key, the instruction is executed; otherwise, a security alarm is triggered.

[0065] In some embodiments, the operation and storage module 25 stores operation records to the blockchain by: generating maintenance logs containing timestamps, operator-specific identification codes, and equipment status based on smart contracts, and writing the log hash value into chaincode based on Hyperledger Fabric.

[0066] In some embodiments, the system further includes an evaluation module for periodically retrieving historical data from the blockchain and evaluating the health status of the photovoltaic system by comparing the attenuation rate of the measured resistance with that of the historical resistance.

[0067] In the above solution, multimodal IoT sensors are deployed on photovoltaic panels, inverters, and charging piles to collect current, temperature, and ambient humidity data in real time, covering the full-dimensional operating status of the current system and solving the blind spot problem of traditional single current monitoring. A dynamic resistance reference curve is generated based on the resistance-temperature index relationship model to improve temperature drift adaptability. The matching degree between the current waveform and the dynamic reference curve is analyzed by a cloud-based long short-term memory model to improve the accuracy of fault type identification and reduce the false alarm rate. Dynamic keys are embedded in diagnostic commands to ensure the security of command transmission and prevent malicious tampering. Operation and maintenance records are written to the blockchain through smart contracts to realize operation timestamps, personnel-specific identification codes, and immutable evidence of equipment status, thereby improving data traceability efficiency.

[0068] Figure 3This is a schematic diagram of the structural framework of one embodiment of the IoT-based photovoltaic parking shed current fault detection device of this application. The IoT-based photovoltaic parking shed current fault detection device 30 can vary significantly due to different configurations or performance. It may include one or more processors 31 and memory 32. The processor 31 may be configured to communicate with the memory 32, executing a series of instructions stored in the memory on the IoT-based photovoltaic parking shed current fault detection device to implement the steps of the aforementioned IoT-based photovoltaic parking shed current fault detection method. Those skilled in the art will understand that... Figure 3 The illustrated structure of the IoT-based photovoltaic parking shed current fault detection device does not constitute a limitation on the IoT-based photovoltaic parking shed current fault detection device provided by the present invention. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0069] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the Internet of Things-based photovoltaic parking shed current fault detection method.

[0070] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0071] If the integrated unit is implemented as a software functional unit 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 the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0072] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic carport current fault detection method based on Internet of Things, characterized in that, The method comprises: Real-time acquisition of current data and temperature data through Internet of Things sensors deployed on photovoltaic panels, inverters and charging piles; Sending the collected data to an edge computing node for preprocessing, using the preprocessed data for resistance-temperature relationship modeling to generate a dynamic resistance reference curve; the preprocessing includes filtering and standardization; The resistance-temperature relationship modeling comprises: The exponential relationship between resistance and temperature is fitted according to the following formula: wherein, represents the predicted resistance value at temperature T; is the reference temperature is the resistance value measured at the lower temperature; k is a material coefficient associated with the degree of sensitivity of the photovoltaic device material to temperature variations, is a compensation parameter based on the self-heating of the device; Uploading the preprocessed data to a cloud server, combining a long short-term memory model trained on historical data to analyze the matching degree of the current waveform and the dynamic resistance reference curve; In response to detecting abnormal fluctuations or deviation of the current waveform from the dynamic resistance reference curve, generating a diagnostic instruction containing fault type and positioning information; After the operation and maintenance center verifies the diagnostic instruction, executing remote repair or triggering an alarm, and storing operation records to a blockchain.

2. The IoT based photovoltaic carport current fault detection method as claimed in claim 1, wherein, Building the long short-term memory model comprises: Setting a double-channel mechanism in front of the input layer of the long short-term memory model, the double-channel mechanism comprising a first channel and a second channel, the first channel inputting historical current waveform sequences, and the second channel inputting the dynamic resistance reference curve, and achieving data alignment of the first channel and the second channel through timestamp synchronization; The output layer is defined as five types of fault type labels, including physical aging, chemical corrosion, instantaneous overload, poor contact and normal state; Using a cross-entropy loss algorithm to calculate the loss of the long short-term memory model.

3. The IoT based photovoltaic carport current fault detection method as claimed in claim 2, wherein, The cross-entropy loss algorithm includes the following formula: Wherein, Loss is a loss value, used to measure the difference between the long short-term memory model and the true label, the smaller the loss value, the more accurate the prediction of the long short-term memory model; N is the total number of samples participating in training; i is the ith sample; c represents the cth fault label, c=1, 2, 3, 4, 5; 1 is physical aging, 2 is chemical corrosion, 3 is instantaneous overload, 4 is poor contact, and 5 is normal state; Yic is the cth true label of the ith sample, using one-hot encoding, if the sample i belongs to the category c, then , otherwise ; Yic is the cth true label of the ith sample, using one-hot encoding, if the sample i belongs to the category c, then The probability of the value of is [0, 1], and the sum of all fault label probabilities of the same sample is 1.

4. The IoT based photovoltaic carport current fault detection method as claimed in claim 1, wherein, The verification of the diagnostic instruction comprises: The operation and maintenance center generates a first dynamic key containing a timestamp and embeds the diagnostic instruction; After the edge computing node receives the instruction, it generates a second dynamic key by encrypting the timestamp through a pre-stored hash function; If the first dynamic key matches the second dynamic key, the instruction is executed, otherwise a security alarm is triggered.

5. The IoT-based photovoltaic carport current fault detection method according to any one of claims 1-4, characterized in that, Storing operation records to a blockchain comprises: generating a maintenance log containing a timestamp, an operator's exclusive identification code and a device state based on a smart contract, and writing the log hash value to a chain code based on the Hyperledger fabric.

6. The IoT-based photovoltaic carport current fault detection method of claim 5, wherein, The method further comprises: periodically calling historical data in the blockchain, and evaluating the health status of the photovoltaic system by comparing the decay rates of the measured resistance and the historical resistance.

7. A photovoltaic carport current fault detection device based on Internet of Things, characterized in that, The Internet of Things-based photovoltaic parking shed current fault detection device comprises: An acquisition module: real-time acquisition of current data and temperature data through Internet of Things sensors deployed on photovoltaic panels, inverters and charging piles; A processing module: sending the collected data to an edge computing node for preprocessing, using the preprocessed data for resistance-temperature relationship modeling to generate a dynamic resistance reference curve; the preprocessing includes filtering and standardization; The resistance-temperature relationship modeling comprises: The exponential relationship between resistance and temperature is fitted according to the following formula: wherein, represents the predicted resistance value at temperature T; Tref is the reference temperature Rmeas is the measured resistance value; k is a material coefficient associated with the degree of sensitivity of the photovoltaic device material to temperature variations, is a compensation parameter based on the self-heating of the device; An analysis module: uploading the preprocessed data to a cloud server, combining a long short-term memory model trained on historical data to analyze the matching degree of the current waveform and the dynamic resistance reference curve; A response module: in response to detecting abnormal fluctuations or deviation of the current waveform from the dynamic resistance reference curve, generating a diagnostic instruction containing fault type and positioning information; Operation and maintenance module: the operation and maintenance center verifies the diagnosis instruction, executes remote repair or triggers alarm, and stores operation record to the block chain.

8. A photovoltaic carport current fault detection device based on Internet of Things, characterized in that, The Internet of Things-based photovoltaic parking shed current fault detection device comprises a memory and at least one processor, and the memory stores instructions; The at least one processor calls the instructions in the memory to enable the Internet of Things-based photovoltaic parking shed current fault detection device to perform the steps of the Internet of Things-based photovoltaic parking shed current fault detection method according to any one of claims 1-6.

9. A computer-readable storage medium having stored thereon instructions, the instructions comprising, The instructions are processed to perform the steps of the Internet of Things-based photovoltaic parking shed current fault detection method according to any one of claims 1-6.

Images