Power equipment intelligent monitoring method and device, computer device and storage medium

Abstract

The application relates to an intelligent monitoring method and device for power equipment, computer equipment and a storage medium. The method comprises the following steps: obtaining power equipment operation data of a target power grid and an operation data feature extraction model; inputting the power equipment operation data into the operation data feature extraction model to obtain a plurality of high-dimensional power equipment feature data of the target power grid; adopting a self-organizing feature fusion algorithm to fuse the high-dimensional power equipment feature data to obtain high-dimensional power equipment fusion features of the target power grid; and adopting a power equipment monitoring algorithm to monitor the high-dimensional power equipment fusion features to obtain power equipment monitoring data of the target power grid. The method can effectively improve the monitoring efficiency of the power equipment within limited computing resources, greatly enhance the sensitivity and accuracy of the power equipment state monitoring, significantly improve the stability and reliability of the power grid operation, and effectively reduce the equipment failure and operation risk.

Description

Technical Field

[0001] This application relates to the field of smart grid technology, and in particular to a method, device, computer equipment, storage medium and computer program product for intelligent monitoring of power equipment. Background Technology

[0002] With the development of power grid technology, power equipment monitoring technology has emerged. This technology utilizes various sensors and intelligent technologies to monitor and collect data on the operating status of power equipment in real time, ensuring normal operation and timely detection of potential faults. This method involves using sensors such as current, voltage, temperature, and vibration to acquire equipment operating parameters, evaluating equipment status through data analysis and processing techniques, and combining predictive maintenance and fault diagnosis technologies to ensure the safe, stable, and efficient operation of power equipment.

[0003] Due to the large number of electrical devices in the power grid, the amount of monitoring data will be enormous. Traditional power equipment monitoring, which uses conventional data analysis and processing techniques, will result in a huge amount of computation, requiring computers with high computing power. Especially when dealing with real-time monitoring data and data acquisition under complex conditions, the assessment of equipment operating status and fault prediction cannot be completed in a timely manner, and may even cause the computer to malfunction, resulting in low efficiency in monitoring power equipment. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for intelligent monitoring of power equipment that can effectively improve the monitoring efficiency of power equipment, addressing the aforementioned technical problems.

[0005] Firstly, this application provides a method for intelligent monitoring of power equipment. The method includes:

[0006] The system acquires the operating data of the power equipment in the target power grid and a feature extraction model for the operating data; the feature extraction model is constructed using a biomimetic neural network.

[0007] The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0008] A self-organizing feature fusion algorithm is used to fuse the feature data of each of the high-dimensional power equipment to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0009] A power equipment monitoring algorithm is used to monitor the fusion features of the high-dimensional power equipment to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0010] Secondly, this application also provides an intelligent monitoring device for power equipment. The device includes:

[0011] The data acquisition module is used to acquire the power equipment operation data of the target power grid and the operation data feature extraction model; the operation data feature extraction model is constructed through a biomimetic neural network.

[0012] The feature extraction module is used to input the power equipment operation data into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0013] The feature fusion module is used to fuse the feature data of each of the high-dimensional power equipment using a self-organizing feature fusion algorithm to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0014] An anomaly monitoring module is used to monitor the fused features of the high-dimensional power equipment using a power equipment monitoring algorithm to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0015] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0016] The system acquires the operating data of the power equipment in the target power grid and a feature extraction model for the operating data; the feature extraction model is constructed using a biomimetic neural network.

[0017] The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0018] A self-organizing feature fusion algorithm is used to fuse the feature data of each of the high-dimensional power equipment to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0019] A power equipment monitoring algorithm is used to monitor the fusion features of the high-dimensional power equipment to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0020] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0021] The system acquires the operating data of the power equipment in the target power grid and a feature extraction model for the operating data; the feature extraction model is constructed using a biomimetic neural network.

[0022] The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0023] A self-organizing feature fusion algorithm is used to fuse the feature data of each of the high-dimensional power equipment to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0024] A power equipment monitoring algorithm is used to monitor the fusion features of the high-dimensional power equipment to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0025] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0026] The system acquires the operating data of the power equipment in the target power grid and a feature extraction model for the operating data; the feature extraction model is constructed using a biomimetic neural network.

[0027] The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0028] A self-organizing feature fusion algorithm is used to fuse the feature data of each of the high-dimensional power equipment to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0029] A power equipment monitoring algorithm is used to monitor the fusion features of the high-dimensional power equipment to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0030] The aforementioned intelligent monitoring method, device, computer equipment, storage medium, and computer program product for power equipment acquires power equipment operation data and an operation data feature extraction model of the target power grid. The operation data feature extraction model is constructed using a biomimetic neural network. The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid. A self-organizing feature fusion algorithm is used to fuse the various high-dimensional power equipment feature data to obtain high-dimensional power equipment fusion features of the target power grid. The self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain. A power equipment monitoring algorithm is used to monitor the high-dimensional power equipment fusion features to obtain power equipment monitoring data of the target power grid. The power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal. The power equipment monitoring algorithm is obtained through genetic algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms.

[0031] The operational data feature extraction model built using biomimetic neural networks can meticulously capture and accurately extract multi-dimensional operational data features of power equipment, significantly improving the accuracy and comprehensiveness of feature extraction. A self-organizing feature fusion algorithm, simulating the brain's multi-sensory fusion mechanism, can efficiently integrate multi-source high-dimensional feature data to generate highly representative and robust fused features of power equipment. Power equipment monitoring algorithms built using gene algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms achieve precise monitoring and anomaly identification of fused features. This effectively improves the monitoring efficiency of power equipment within limited computing resources, while greatly enhancing the sensitivity and accuracy of power equipment condition monitoring, significantly improving the stability and reliability of power grid operation, effectively reducing equipment failures and operational risks, and comprehensively improving the intelligent management level of power equipment. Attached Figure Description

[0032] Figure 1 This is an application environment diagram of an intelligent monitoring method for power equipment in one embodiment;

[0033] Figure 2 This is a flowchart illustrating an intelligent monitoring method for power equipment in one embodiment;

[0034] Figure 3 This is a flowchart illustrating a method for obtaining high-dimensional power equipment fusion features in one embodiment;

[0035] Figure 4 This is a flowchart illustrating a method for obtaining high-dimensional power equipment fusion features in another embodiment;

[0036] Figure 5 This is a flowchart illustrating the method for obtaining high-dimensional power equipment fusion features in yet another embodiment;

[0037] Figure 6 This is a flowchart illustrating a method for obtaining power equipment monitoring data in one embodiment;

[0038] Figure 7 This is a flowchart illustrating a method for obtaining power equipment monitoring data in another embodiment;

[0039] Figure 8 This is a structural block diagram of an intelligent monitoring device for power equipment in one embodiment;

[0040] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] The intelligent monitoring method for power equipment provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Server 104 obtains the operating data of the target power grid's power equipment and an operating data feature extraction model through terminal 102. The operating data feature extraction model is constructed using a biomimetic neural network. The operating data of the power equipment is input into the operating data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid. A self-organizing feature fusion algorithm is used to fuse the various high-dimensional power equipment feature data to obtain the high-dimensional power equipment fusion features of the target power grid. The self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain. A power equipment monitoring algorithm is used to monitor the high-dimensional power equipment fusion features to obtain the power equipment monitoring data of the target power grid. The power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal. The power equipment monitoring algorithm is obtained through genetic algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0043] In one embodiment, such as Figure 2 As shown, an intelligent monitoring method for power equipment is provided, which can be applied to... Figure 1 Taking the server in the example, the following steps are included:

[0044] Step 202: Obtain the power equipment operation data and operation data feature extraction model of the target power grid.

[0045] Power equipment operation data can include various parameter information collected during the operation of power equipment, such as voltage, current, power, temperature, and load. This data reflects the real-time operating status and performance of the power equipment and can be used to monitor equipment health, diagnose faults, perform predictive maintenance, and optimize grid operating efficiency.

[0046] Among them, the operational data feature extraction model can be an algorithm or method used to extract key features from the operational data of power equipment. By processing and analyzing the raw data, feature values ​​that can represent the health status, performance changes and abnormal conditions of the equipment are extracted.

[0047] Specifically, operational data of all power equipment in the target power grid is acquired, including parameters such as voltage, current, temperature, and load. This operational data is then cleaned and normalized using data preprocessing techniques. Simultaneously, an operational data feature extraction model is constructed. This model mimics the structure and function of a biological neural network to extract features from the processed operational data. Specifically, the biomimetic neural network model extracts key features from the operational data, such as the health status and abnormal patterns of the power equipment, through the connection and weight adjustment of multiple layers of neurons.

[0048] Step 204: Input the power equipment operation data into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid.

[0049] Among them, high-dimensional power equipment feature data can be key feature data of multiple dimensions extracted from power equipment operation data.

[0050] Specifically, power equipment operation data, including parameters such as voltage, current, temperature, and load, are input into a pre-built operation data feature extraction model. This model processes the input data through a biomimetic neural network, automatically identifying and extracting key features that reflect the health status and performance changes of the equipment, and generating several high-dimensional power equipment feature data.

[0051] Step 206: Use a self-organizing feature fusion algorithm to fuse the feature data of each high-dimensional power equipment to obtain the high-dimensional power equipment fusion features of the target power grid.

[0052] Among them, the self-organizing feature fusion algorithm can be a type of algorithm that integrates feature data from different sources and dimensions by simulating the brain's multi-sensory fusion mechanism. It utilizes the self-organizing characteristics of neural networks to fuse different data by adaptively adjusting network weights and connections.

[0053] Among them, the high-dimensional power equipment fusion feature can be a high-dimensional comprehensive feature vector of power equipment obtained by integrating multiple high-dimensional feature data from different power equipment through a self-organizing feature fusion algorithm.

[0054] Specifically, the high-dimensional power equipment feature data obtained from the operational data feature extraction model are input into the self-organizing feature fusion algorithm. The self-organizing feature fusion algorithm simulates the multi-sensory fusion mechanism of the brain to adaptively fuse and integrate feature data from different sources and dimensions. At the same time, the self-organizing feature fusion algorithm utilizes the self-organizing characteristics of neural networks to adjust the weights and connection methods, and finally fuses the various high-dimensional feature data into a comprehensive high-dimensional feature vector to obtain the high-dimensional power equipment fusion features.

[0055] Step 208: Use power equipment monitoring algorithms to monitor the fusion characteristics of high-dimensional power equipment and obtain power equipment monitoring data of the target power grid.

[0056] Among them, the power equipment monitoring algorithm can be a comprehensive algorithm that combines genetic algorithms, adaptive evolutionary mechanisms and swarm intelligence algorithms by simulating the process of natural selection and group cooperation, and is used to monitor and analyze the high-dimensional fusion characteristics of power equipment in real time.

[0057] Among them, power equipment monitoring data can be generated by real-time analysis of the fusion characteristics of high-dimensional power equipment through power equipment monitoring algorithms, which is used to monitor various different devices in the target power grid.

[0058] Specifically, high-dimensional fused features of power equipment are input into the power equipment monitoring algorithm. This algorithm combines genetic algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms to monitor and analyze high-dimensional fused features in real time by simulating natural selection and group cooperation processes. During this process, the power equipment monitoring algorithm continuously optimizes and adjusts its own parameters to identify and extract potential abnormal patterns and features, generating detailed power equipment monitoring data.

[0059] In the aforementioned intelligent monitoring method for power equipment, the following steps are taken: First, operational data of power equipment in the target power grid and an operational data feature extraction model are acquired. The operational data feature extraction model is constructed using a biomimetic neural network. The operational data of the power equipment is input into the operational data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid. A self-organizing feature fusion algorithm is used to fuse the various high-dimensional power equipment feature data to obtain the high-dimensional power equipment fusion features of the target power grid. The self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain. A power equipment monitoring algorithm is used to monitor the high-dimensional power equipment fusion features to obtain power equipment monitoring data of the target power grid. The power equipment monitoring data is used to monitor whether any abnormalities occur in the operating status of the target power grid. The power equipment monitoring algorithm is obtained through genetic algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms.

[0060] The operational data feature extraction model built using biomimetic neural networks can meticulously capture and accurately extract multi-dimensional operational data features of power equipment, significantly improving the accuracy and comprehensiveness of feature extraction. A self-organizing feature fusion algorithm, simulating the brain's multi-sensory fusion mechanism, can efficiently integrate multi-source high-dimensional feature data to generate highly representative and robust fused features of power equipment. Power equipment monitoring algorithms built using gene algorithms, adaptive evolutionary mechanisms, and swarm intelligence algorithms achieve precise monitoring and anomaly identification of fused features. This effectively improves the monitoring efficiency of power equipment within limited computing resources, while greatly enhancing the sensitivity and accuracy of power equipment condition monitoring, significantly improving the stability and reliability of power grid operation, effectively reducing equipment failures and operational risks, and comprehensively improving the intelligent management level of power equipment.

[0061] In one embodiment, such as Figure 3 As shown, the self-organizing feature fusion algorithm includes deep neural networks, spiking neural networks, and self-organizing mechanism fusion networks. Using this algorithm, feature data from various high-dimensional power equipment are fused to obtain the high-dimensional power equipment fusion features of the target power grid, including:

[0062] Step 302: Input the feature data of each high-dimensional power equipment into a deep neural network for feature extraction to obtain local feature data of the high-dimensional equipment.

[0063] Deep neural networks are machine learning models composed of multiple layers of artificial neurons. Each layer receives and processes the output of the previous layer, gradually extracting and abstracting features from the input data. They are widely used in fields such as image recognition, natural language processing, and signal processing to achieve high-precision prediction and classification tasks.

[0064] Among them, high-dimensional device local feature data can be local high-dimensional feature data of the target power grid or local power equipment.

[0065] Specifically, the feature data of each high-dimensional power device is input into a pre-trained deep neural network. This deep neural network has a multi-layered neuron structure, which can extract and abstract features from the input data layer by layer. Through nonlinear transformations and weight adjustments at each layer, the deep neural network gradually extracts the key features from the data. Finally, the deep neural network outputs high-dimensional local feature data of the devices, which contains deep information and local patterns of the original features.

[0066] Step 304: Input the feature data of each high-dimensional power equipment into the spiking neural network to obtain time-series information fusion feature data.

[0067] Among them, spiking neural networks can be artificial neural networks that simulate the spiking mechanism of biological neurons, processing data through time encoding and pulse signals. Unlike traditional neural networks, neurons in spiking neural networks communicate through discrete pulse signals, which can effectively capture and process the temporal dynamics and temporal correlations in data.

[0068] Among them, time-series information fusion feature data can be comprehensive feature information extracted from time series data, reflecting the dynamic patterns and trends of data changes over time.

[0069] Specifically, the characteristic data of each high-dimensional power equipment is input into a spiking neural network. This spiking neural network processes the characteristic data of each high-dimensional power equipment by simulating the pulse firing mechanism of biological neurons. For example, the spiking neural network uses the temporal information of pulse signals to capture and fuse the temporal correlation and dynamic patterns in the characteristic data of each high-dimensional power equipment. Combined with the pulse firing and time encoding of neurons in each layer of the spiking neural network, the final time-series information fused characteristic data is generated. These data can comprehensively reflect the temporal dynamics and state changes during the operation of power equipment.

[0070] Step 306: Input the high-dimensional equipment local feature data and time-series information fusion feature data into the self-organizing mechanism fusion network to obtain the high-dimensional power equipment fusion features.

[0071] Among them, the self-organizing mechanism fusion network can utilize adaptive weight adjustment and hierarchical structure to automatically coordinate and optimize the fusion process of input data, generating a neural network of fused data.

[0072] Specifically, high-dimensional equipment local feature data and time-series information fusion feature data are input into a self-organizing mechanism fusion network. This self-organizing mechanism fusion network fuses and integrates different input feature data by simulating the self-organizing mechanism of biological systems. For example, the self-organizing mechanism fusion network uses adaptive weight adjustment and hierarchical structure to deeply fuse local features and time-series features to generate more comprehensive and accurate high-dimensional power equipment fusion features.

[0073] In this embodiment, feature data of each high-dimensional power device are input into a deep neural network and a spiking neural network for feature extraction and temporal information fusion. The extracted high-dimensional local feature data and the fused temporal information feature data are then input into a self-organizing mechanism fusion network. The resulting high-dimensional power device fusion features more comprehensively reflect the operating status of the equipment. This method leverages the powerful feature extraction capabilities of deep neural networks and the temporal information processing advantages of spiking neural networks, further optimizing feature integration through a self-organizing mechanism fusion network. This provides a more accurate and comprehensive power equipment monitoring and fault diagnosis model, effectively improving the accuracy and reliability of equipment condition assessment.

[0074] In one embodiment, such as Figure 4 As shown, high-dimensional equipment local feature data and time-series information fusion feature data are input into a self-organizing mechanism fusion network to obtain high-dimensional power equipment fusion features, including:

[0075] Step 402: Combine the high-dimensional equipment local feature data and time-series information fusion feature data to obtain preliminary power equipment fusion features.

[0076] Among them, the initial power equipment fusion characteristics can be the compatibility characteristics obtained after fusion processing, but they have not been optimized and may not meet the data usage standards.

[0077] Specifically, the high-dimensional equipment local feature data and the time-series information fusion feature data are standardized to ensure that they are compared and combined on the same scale. Then, the standardized high-dimensional equipment local feature data and the time-series information fusion feature data are combined. The combination method can be through splicing or weighted averaging to fuse the local features and time-series features into a new feature vector, which serves as the initial power equipment fusion feature.

[0078] Step 404: Based on the brain multi-sensory fusion algorithm in the self-organizing mechanism fusion network, optimize the preliminary power equipment fusion features to obtain high-dimensional power equipment fusion features.

[0079] Among them, the brain multi-sensory fusion algorithm can be an algorithm that simulates the mechanism by which the human brain processes and integrates information from different senses. It can remove redundant and noisy information and enhance the expressive power of key features.

[0080] Specifically, the preliminary power equipment fusion features are input into the brain-based multi-sensory fusion algorithm within the self-organizing mechanism fusion network. This algorithm simulates the brain's mechanism for integrating and processing multi-sensory information. By adaptively adjusting network weights and structure, the algorithm deeply optimizes and refines the preliminary features, comprehensively considering local features and temporal dynamic information, removing redundancy and noise, and enhancing the expressive power of key features. The resulting high-dimensional power equipment fusion features can more accurately and comprehensively reflect the operating status and performance of the power equipment.

[0081] In this embodiment, preliminary fused features of the power equipment are obtained by combining high-dimensional local feature data and temporal information fusion feature data. These are then optimized using a multi-sensory fusion algorithm within a self-organizing mechanism fusion network, ultimately yielding high-dimensional fused features of the power equipment. This method effectively integrates local features and temporal dynamic information of the equipment, leveraging the optimization capabilities of the multi-sensory fusion algorithm to enhance the expressive power and accuracy of the feature data, thereby providing a more comprehensive and accurate description of the power equipment's state. This process improves the accuracy and reliability of power equipment monitoring and fault diagnosis, enabling more timely and accurate identification of potential problems and optimization of equipment management and maintenance strategies.

[0082] In one embodiment, such as Figure 5 As shown, based on the brain-like multi-sensory fusion algorithm in the self-organizing mechanism fusion network, the preliminary power equipment fusion features are optimized to obtain high-dimensional power equipment fusion features, including:

[0083] Step 502: Based on the brain-based multi-sensory fusion algorithm in the self-organizing mechanism fusion network, competitive learning is performed on the preliminary power equipment fusion features to obtain fusion feature competition information.

[0084] Among them, the fusion feature competition information can be key feature data extracted from the initial fusion features of power equipment through the competitive learning mechanism in the brain's multi-sensory fusion algorithm.

[0085] Specifically, the preliminary fusion features of power equipment are input into a brain-based multi-sensory fusion algorithm within a self-organizing mechanism fusion network for competitive learning. This algorithm simulates the competition mechanism between neurons in the brain, adaptively adjusting network weights and connection structures. Features compete within the network, with important features being strengthened and unimportant features suppressed or eliminated. Through multiple iterations and weight updates, the brain-based multi-sensory fusion algorithm extracts the most representative fusion feature competition information.

[0086] Step 504: Based on the brain-based multi-sensory fusion algorithm in the self-organizing mechanism fusion network, assist learning is performed on the preliminary power equipment fusion features to obtain fusion feature assistance information.

[0087] Among them, the fusion feature assistance information can be key feature data obtained after assisted learning through brain multi-sensory fusion algorithms in a self-organizing mechanism fusion network.

[0088] Specifically, the preliminary fusion features of the power equipment are input into a brain-based multi-sensory fusion algorithm within a self-organizing mechanism fusion network for assisted learning. This algorithm, by simulating the collaborative mechanism between neurons in the brain, adaptively adjusts the network's weights and connection structure. Features collaborate within the network, complementing and reinforcing each other to form a comprehensive feature representation. Through multiple iterations and collaborative optimization, the brain-based multi-sensory fusion algorithm extracts the fusion feature assistance information.

[0089] Step 506: Based on the competition information and assistance information of the fusion features, optimize the preliminary power equipment fusion features to obtain the high-dimensional power equipment fusion features.

[0090] Specifically, deep processing is performed based on the optimization algorithm in the self-organizing mechanism fusion network. The competition and cooperation relationships between features are comprehensively considered. The network weights and structure are adaptively adjusted. Based on the competition information and assistance information of the fusion features, the key features of the initial power equipment fusion features are optimized by means of redundancy removal, noise reduction and enhancement. The high-dimensional power equipment fusion features are generated after multiple rounds of iteration and optimization.

[0091] In this embodiment, a brain-like multi-sensory fusion algorithm in a self-organizing mechanism fusion network is used. First, competitive learning is performed on the initial fusion features of power equipment to obtain competitive information about the fusion features. Then, cooperative learning is performed to obtain cooperative information about the fusion features. Finally, this information is used to optimize the initial features, resulting in high-dimensional fusion features of power equipment. This method effectively combines competition and cooperation mechanisms, preserving key features while enhancing the synergistic effect between features during feature extraction, thereby improving the accuracy and richness of feature representation. This optimized high-dimensional fusion feature of power equipment provides a more comprehensive and accurate description of equipment operating status, helping to improve the accuracy and reliability of power equipment monitoring and fault diagnosis.

[0092] In one embodiment, such as Figure 6 As shown, a power equipment monitoring algorithm is used to monitor the fusion characteristics of high-dimensional power equipment, obtaining power equipment monitoring data for the target power grid, including:

[0093] Step 602: Based on the fusion characteristics of high-dimensional power equipment, determine the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm.

[0094] Among them, the monitoring strategy information can be the strategies and methods used to monitor the operating status of power equipment in real time.

[0095] Among them, diagnostic strategy information can be strategies and methods used for real-time diagnosis of abnormal states of power equipment.

[0096] Specifically, by utilizing the fusion features of high-dimensional power equipment, the operating status and performance indicators of power equipment are analyzed and identified. Furthermore, monitoring strategies for power equipment monitoring algorithms are developed to determine key parameters and anomaly thresholds requiring focused monitoring. Simultaneously, fault modes and anomaly characteristics from the fusion features of high-dimensional power equipment are used to formulate diagnostic strategies, including fault type identification criteria and diagnostic procedures.

[0097] Step 604: Using the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm, the fusion characteristics of high-dimensional power equipment are monitored to obtain power equipment status assessment data and power equipment fault risk data.

[0098] Among them, power equipment condition assessment data can be assessment data used to evaluate the overall performance and operating efficiency of the equipment.

[0099] Among them, power equipment failure risk data can be used to assess the probability and potential severity of equipment failure.

[0100] Specifically, based on the monitoring strategy information of the power equipment monitoring algorithm, the fused characteristics of high-dimensional power equipment are monitored in real time to capture changes in key performance indicators and compare them with preset anomaly thresholds. Simultaneously, based on diagnostic strategy information, potential or already detected abnormal features in the fused characteristics of high-dimensional power equipment are analyzed to identify potential fault types and risk levels. Through these two processes, power equipment status assessment data is generated, reflecting the overall operating status of the equipment, and power equipment fault risk data is obtained to assess the probability and severity of fault occurrence.

[0101] Step 606: Obtain power equipment monitoring data for the target power grid based on power equipment condition assessment data and power equipment fault risk data.

[0102] Specifically, by integrating and analyzing power equipment condition assessment data and power equipment fault risk data, and by comparing and summarizing the real-time operating status and fault risk level of the equipment, such as the comprehensive evaluation of various performance indicators, the quantitative analysis of fault warnings and risks, and the judgment of the overall health status of the equipment, the generated power equipment monitoring data can comprehensively reflect the operating status and potential risks of each power equipment in the target power grid.

[0103] In this embodiment, by fusing high-dimensional power equipment features, the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm are determined. These strategy information are then used to monitor the high-dimensional features, obtaining power equipment status assessment data and fault risk data, ultimately generating power equipment monitoring data for the target power grid. This method achieves end-to-end optimization from feature extraction to monitoring and diagnosis, providing comprehensive and accurate equipment status assessment and fault early warning information. Its beneficial effects include the ability to accurately monitor the operating status of power equipment, promptly identify potential faults, optimize equipment management and maintenance strategies, and significantly improve the safety and reliability of power grid operation.

[0104] In one embodiment, such as Figure 7 As shown, based on the fusion characteristics of high-dimensional power equipment, the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm are determined, including:

[0105] Step 702: Based on the fusion characteristics of high-dimensional power equipment, construct the initial population equipment state model for the power equipment monitoring algorithm.

[0106] The initial population device state model can be a population model constructed from the fusion features of high-dimensional power equipment and covering different aspects of the equipment's operating state.

[0107] Specifically, high-dimensional power equipment fusion features are used as input data to initialize multiple equipment state models, each representing a different equipment operating state. These models are then integrated into an initial population of equipment state models using predefined rules and parameters.

[0108] Step 704: Using genetic algorithms and swarm intelligence algorithms, the initial population device state model is iteratively optimized to obtain an optimized device state model.

[0109] Among them, gene algorithms can be search and optimization algorithms that simulate the natural evolutionary process, iteratively optimizing individual populations through operations such as selection, crossover, and mutation.

[0110] Among them, swarm intelligence algorithms can be optimization algorithms based on the collaboration and information sharing of multiple individuals, simulating the process of group behavior in nature (such as ant colonies foraging and bird flocks flying), and gradually finding the optimal solution to the problem through the interaction and self-organization mechanism between individuals.

[0111] Among them, the optimized equipment state model can be an improved equipment operating state model through a series of optimization processes (such as iterative algorithms, genetic algorithms, or swarm intelligence algorithms).

[0112] Specifically, a genetic algorithm is used to iteratively optimize the initial population of equipment state models. This involves initializing multiple individuals within the initial population of equipment state models using a genetic algorithm, with each individual representing a different equipment state model. These models initially cover the possible equipment operating state space. By evaluating the fitness of these individuals, selection, crossover, and mutation operations are performed on high-performing individuals to generate new population individuals. Through multiple iterations, the original population of equipment state models is gradually optimized, resulting in new population equipment state models. Next, these new population equipment state models are combined with swarm intelligence algorithms, such as particle swarm optimization, to further enhance the model's global search capability and convergence speed through information sharing and collaborative search among individuals. Through multiple iterations and optimizations, models with higher fitness are gradually selected, ultimately yielding the optimized equipment state model.

[0113] Step 706: Adopt an adaptive evolutionary mechanism to set the reinforcement learning strategy information for the power equipment monitoring algorithm.

[0114] Among them, the adaptive evolution mechanism can be an optimization method based on evolutionary principles and adaptive adjustment, which continuously optimizes the parameters and structure of the system by simulating genetic variation, crossover and selection in the natural selection process.

[0115] Among them, reinforcement learning policy information can be the policy that the algorithm learns and optimizes autonomously by defining states, actions and reward functions during the reinforcement learning process.

[0116] Specifically, an adaptive evolutionary mechanism is applied to initialize the parameters and model of the power equipment monitoring algorithm, and a preliminary evaluation is performed based on equipment operating data. Then, the state, action, and reward function for reinforcement learning are defined, and the algorithm parameters are continuously adjusted through experimentation and feedback. Through multiple rounds of iteration and optimization, the algorithm adaptively adjusts its learning strategy, enabling it to select the optimal monitoring and diagnostic actions under different operating conditions, and ultimately setting the optimal reinforcement learning strategy information.

[0117] Step 708: Based on reinforcement learning strategy information, input the fused features of high-dimensional power equipment into the optimized equipment state model to obtain monitoring strategy information and diagnostic strategy information.

[0118] Specifically, based on the established reinforcement learning strategy, the fused features of high-dimensional power equipment are input into the optimized equipment state model. This optimized equipment state model utilizes the reinforcement learning strategy, setting reward and penalty mechanisms, to analyze and process the input fused features of high-dimensional power equipment. It then dynamically adjusts monitoring and diagnostic parameters based on the analysis and processing results, enabling the optimized power equipment monitoring model to autonomously learn and improve monitoring strategies under different states. Through multiple iterations, the optimized power equipment monitoring model gradually improves its adaptability and accuracy, ultimately outputting monitoring and diagnostic strategy information.

[0119] In this embodiment, an initial population equipment state model is constructed based on the fusion characteristics of high-dimensional power equipment. This model is then iteratively optimized using genetic algorithms and swarm intelligence algorithms to obtain an optimized equipment state model. Next, an adaptive evolutionary mechanism is used to set reinforcement learning strategy information. Based on this strategy information, the fusion characteristics of the high-dimensional power equipment are input into the optimized equipment state model, ultimately yielding monitoring and diagnostic strategy information. The beneficial effect of this method is that by integrating multiple advanced algorithms and optimization strategies, adaptive optimization and intelligent learning of the power equipment state model are achieved, thereby providing more accurate and dynamic monitoring and diagnostic capabilities, significantly improving the operating efficiency and reliability of power equipment.

[0120] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0121] Based on the same inventive concept, this application also provides an intelligent monitoring device for power equipment to implement the aforementioned intelligent monitoring method for power equipment. The solution provided by this device is similar to the implementation scheme described in the above method; therefore, the specific limitations of one or more embodiments of the intelligent monitoring device for power equipment provided below can be found in the limitations of the intelligent monitoring method for power equipment described above, and will not be repeated here.

[0122] In one embodiment, such as Figure 8 As shown, an intelligent monitoring device for power equipment is provided, comprising: a data acquisition module 802, a feature extraction module 804, a feature fusion module 806, and an anomaly monitoring module 808, wherein:

[0123] The data acquisition module 802 is used to acquire the power equipment operation data of the target power grid and the operation data feature extraction model; the operation data feature extraction model is constructed through a biomimetic neural network.

[0124] The feature extraction module 804 is used to input the power equipment operation data into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid;

[0125] The feature fusion module 806 is used to fuse the feature data of various high-dimensional power equipment using a self-organizing feature fusion algorithm to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the multi-sensory fusion mechanism of the brain.

[0126] The anomaly monitoring module 808 is used to monitor the fusion characteristics of high-dimensional power equipment using a power equipment monitoring algorithm to obtain power equipment monitoring data of the target power grid. The power equipment monitoring data is used to monitor whether there are any anomalies in the operating status of the target power grid. The power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

[0127] In one embodiment, the feature fusion module 806 is further configured to input the feature data of each high-dimensional power device into a deep neural network for feature extraction to obtain local feature data of the high-dimensional device; input the feature data of each high-dimensional power device into a spiking neural network to obtain time-series information fusion feature data; and input the local feature data of the high-dimensional device and the time-series information fusion feature data into a self-organizing mechanism fusion network to obtain high-dimensional power device fusion features.

[0128] In one embodiment, the feature fusion module 806 is further configured to combine high-dimensional device local feature data and time-series information fusion feature data to obtain preliminary power device fusion features; and optimize the preliminary power device fusion features according to the brain multi-sensory fusion algorithm in the self-organizing mechanism fusion network to obtain high-dimensional power device fusion features.

[0129] In one embodiment, the feature fusion module 806 is further configured to: perform competitive learning on the preliminary power equipment fusion features according to the brain multi-sensory fusion algorithm in the self-organizing mechanism fusion network to obtain fusion feature competition information; perform assisted learning on the preliminary power equipment fusion features according to the brain multi-sensory fusion algorithm in the self-organizing mechanism fusion network to obtain fusion feature assistance information; and optimize the preliminary power equipment fusion features according to the fusion feature competition information and the fusion feature assistance information to obtain high-dimensional power equipment fusion features.

[0130] In one embodiment, the anomaly monitoring module 808 is further configured to determine the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm based on the high-dimensional power equipment fusion characteristics; use the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm to monitor the high-dimensional power equipment fusion characteristics to obtain power equipment status assessment data and power equipment fault risk data; and obtain power equipment monitoring data of the target power grid based on the power equipment status assessment data and power equipment fault risk data.

[0131] In one embodiment, the anomaly monitoring module 808 is further configured to: construct an initial population device state model for the power equipment monitoring algorithm based on the high-dimensional power equipment fusion characteristics; iteratively optimize the initial population device state model using a genetic algorithm and a swarm intelligence algorithm to obtain an optimized device state model; set the reinforcement learning strategy information for the power equipment monitoring algorithm using an adaptive evolution mechanism; and input the high-dimensional power equipment fusion characteristics into the optimized device state model based on the reinforcement learning strategy information to obtain monitoring strategy information and diagnostic strategy information.

[0132] The various modules in the aforementioned intelligent monitoring device for power equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0133] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores server data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for intelligent monitoring of power equipment.

[0134] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0135] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0136] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0137] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and executes the computer instructions, causing the computer device to perform the steps in the above method embodiments.

[0138] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0139] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0140] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for intelligent monitoring of power equipment, characterized in that, The method includes: The system acquires the operating data of the power equipment in the target power grid and a feature extraction model for the operating data; the feature extraction model is constructed using a biomimetic neural network. The power equipment operation data is input into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid; The feature data of each of the high-dimensional power equipment is input into the deep neural network in the self-organizing feature fusion algorithm for feature extraction to obtain local feature data of the high-dimensional equipment; the self-organizing feature fusion algorithm includes the deep neural network, the spiking neural network and the self-organizing mechanism fusion network. The feature data of each of the high-dimensional power devices are input into the pulse neural network to obtain time-series information fusion feature data; The high-dimensional equipment local feature data and the time-series information fusion feature data are combined to obtain preliminary power equipment fusion features; Based on the brain-like multi-sensory fusion algorithm in the self-organizing mechanism fusion network, the preliminary power equipment fusion features are subjected to competitive learning to obtain fusion feature competition information; based on the brain-like multi-sensory fusion algorithm in the self-organizing mechanism fusion network, the preliminary power equipment fusion features are subjected to assisted learning to obtain fusion feature assistance information; based on the fusion feature competition information and the fusion feature assistance information, the preliminary power equipment fusion features are optimized to obtain the high-dimensional power equipment fusion features of the target power grid; the self-organizing feature fusion algorithm is obtained by simulating the brain-like multi-sensory fusion mechanism. A power equipment monitoring algorithm is used to monitor the fusion features of the high-dimensional power equipment to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

2. The method according to claim 1, characterized in that, The method employs a power equipment monitoring algorithm to monitor the fused features of the high-dimensional power equipment, thereby obtaining power equipment monitoring data for the target power grid, including: Based on the high-dimensional power equipment fusion characteristics, the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm are determined; The monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm are used to monitor the fusion features of the high-dimensional power equipment, thereby obtaining power equipment status assessment data and power equipment fault risk data. Based on the power equipment condition assessment data and the power equipment fault risk data, the power equipment monitoring data of the target power grid is obtained.

3. The method according to claim 2, characterized in that, The step of determining the monitoring strategy information and diagnostic strategy information of the power equipment monitoring algorithm based on the high-dimensional power equipment fusion characteristics includes: Based on the high-dimensional power equipment fusion characteristics, an initial population equipment state model for the power equipment monitoring algorithm is constructed. The initial population device state model is iteratively optimized using the gene algorithm and the swarm intelligence algorithm to obtain an optimized device state model. The adaptive evolution mechanism is used to set the reinforcement learning strategy information of the power equipment monitoring algorithm; Based on the reinforcement learning strategy information, the fused features of the high-dimensional power equipment are input into the optimized equipment state model to obtain the monitoring strategy information and the diagnostic strategy information.

4. An intelligent monitoring device for power equipment, characterized in that, The device includes: The data acquisition module is used to acquire the power equipment operation data of the target power grid and the operation data feature extraction model; the operation data feature extraction model is constructed through a biomimetic neural network. The feature extraction module is used to input the power equipment operation data into the operation data feature extraction model to obtain several high-dimensional power equipment feature data of the target power grid; A feature fusion module is used to input the feature data of each of the high-dimensional power equipment into a deep neural network in a self-organizing feature fusion algorithm for feature extraction, thereby obtaining local feature data of the high-dimensional equipment. The self-organizing feature fusion algorithm includes the deep neural network, a spiking neural network, and a self-organizing mechanism fusion network. The feature data of each of the high-dimensional power equipment is input into the spiking neural network to obtain time-series information fusion feature data. The local feature data of the high-dimensional equipment and the time-series information fusion feature data are combined to obtain preliminary power equipment fusion features. According to the brain-like multi-sensory fusion algorithm in the self-organizing mechanism fusion network, the preliminary power equipment fusion features are subjected to competitive learning to obtain fusion feature competition information. According to the brain-like multi-sensory fusion algorithm in the self-organizing mechanism fusion network, the preliminary power equipment fusion features are subjected to assisted learning to obtain fusion feature assistance information. According to the fusion feature competition information and the fusion feature assistance information, the preliminary power equipment fusion features are optimized to obtain the high-dimensional power equipment fusion features of the target power grid. The self-organizing feature fusion algorithm is obtained by simulating the brain's multi-sensory fusion mechanism. An anomaly monitoring module is used to monitor the fused features of the high-dimensional power equipment using a power equipment monitoring algorithm to obtain power equipment monitoring data of the target power grid; the power equipment monitoring data is used to monitor whether the operating status of the target power grid is abnormal; the power equipment monitoring algorithm is obtained through genetic algorithm, adaptive evolution mechanism and swarm intelligence algorithm.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images