Electrical equipment performance monitoring method and system based on plateau environment

Abstract

The invention discloses an electrical equipment performance monitoring method and system based on a plateau environment, and relates to the technical field of electrical equipment performance monitoring, and the method comprises the steps: building a self-adaptive compensation model based on plateau environment data and electrical equipment operation parameters through reinforcement learning in combination with a self-adaptive compensation algorithm, and carrying out the training; according to the trained adaptive compensation model, adjusting electrical equipment operation parameters in real time, and generating optimized electrical equipment operation parameters; the optimized electrical equipment operation parameters are combined with plateau environment data, and a time sequence data set is generated through data preprocessing; and constructing and training an electrical equipment performance prediction model through an LSTM algorithm in combination with the time sequence data set, predicting the performance change of the electrical equipment in real time according to the trained performance prediction model, and generating a maintenance suggestion through a maintenance rule engine. According to the method, reinforcement learning is combined with the adaptive compensation algorithm, so that the monitoring precision and prediction reliability of the equipment operation parameters are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment performance monitoring, and in particular to a method and system for electrical equipment performance monitoring based on a plateau environment. Background Art

[0002] In recent years, with the rapid development of the Internet of Things (IoT), big data, and artificial intelligence (AI), related technologies have made significant progress. Traditional monitoring methods primarily rely on regular inspections and fixed threshold alarms. While these methods can ensure safe equipment operation to a certain extent, they suffer from issues such as poor real-time performance and low prediction accuracy. With advances in sensor and communication technologies, real-time data acquisition and remote monitoring have become mainstream trends. In particular, the application of deep learning algorithms (such as LSTM and GRU) to time series data analysis has provided new solutions for equipment performance prediction. Furthermore, reinforcement learning (RL) has also matured in the field of industrial control. Through adaptive compensation algorithms, it can dynamically adjust equipment operating parameters, improving its adaptability and stability in complex environments. However, existing technologies primarily focus on equipment monitoring in plains or conventional environments. Research on adaptability to special environments such as plateaus is relatively limited, and there is a lack of comprehensive monitoring and optimization solutions for electrical equipment performance in plateau environments.

[0003] Existing technologies for monitoring the performance of electrical equipment in plateau environments suffer from the following shortcomings: First, the specific characteristics of the plateau environment (such as low air pressure, low oxygen levels, and strong ultraviolet rays) significantly impact equipment operating parameters. Existing monitoring systems are mostly designed based on standard environments and lack targeted modeling and compensation mechanisms for plateau environmental factors, resulting in insufficient monitoring accuracy and predictive reliability. Second, existing technologies often use single data-driven methods (such as LSTM or reinforcement learning), failing to effectively integrate environmental data with equipment operating parameters, making it difficult to achieve real-time optimization of equipment performance and predictive maintenance. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for monitoring the performance of electrical equipment based on a plateau environment to solve the problems of poor adaptability to the plateau environment and lack of dynamic adjustment capability of maintenance recommendations.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the present invention provides a method for monitoring the performance of electrical equipment based on a plateau environment, which includes: real-time collection of plateau environmental data and monitoring of electrical equipment operating parameters; based on the plateau environmental data and the operating parameters of the electrical equipment, an adaptive compensation model is constructed and trained through reinforcement learning combined with an adaptive compensation algorithm; according to the trained adaptive compensation model, the operating parameters of the electrical equipment are adjusted in real time to generate optimized operating parameters of the electrical equipment; the optimized operating parameters of the electrical equipment are combined with the plateau environmental data, and a time series data set is generated through data preprocessing; an electrical equipment performance prediction model is constructed and trained through the LSTM algorithm combined with the time series data set, and the performance changes of the electrical equipment are predicted in real time according to the trained performance prediction model, and maintenance recommendations are generated through a maintenance rule engine.

[0008] As a preferred embodiment of the method for monitoring electrical equipment performance based on a plateau environment according to the present invention, the plateau environmental data includes air pressure, humidity, precipitation, snow accumulation, dust concentration, and particulate matter concentration;

[0009] The operating parameters of the electrical equipment include temperature, voltage, current, power, insulation resistance, load rate, fault code and energy consumption.

[0010] As a preferred solution of the method for monitoring electrical equipment performance based on plateau environment described in the present invention, wherein: based on plateau environment data and electrical equipment operating parameters, an adaptive compensation model is constructed and trained by combining reinforcement learning with an adaptive compensation algorithm. The specific steps are as follows:

[0011] The state space is constructed using plateau environmental data and electrical equipment operating parameters. The adjustment range of the equipment's adjustable parameters is defined as the action space. Based on equipment performance changes, energy consumption, and failure risk, a weighted synthesis method is used to generate a reward function.

[0012] Build a reinforcement learning framework based on the defined state space, action space, and reward function;

[0013] Based on the reinforcement learning framework, the deep Q network is used as the core algorithm combined with the adaptive PID control algorithm to generate an adaptive compensation model, and the adaptive compensation model is trained through the transfer learning strategy.

[0014] As a preferred solution of the method for monitoring the performance of electrical equipment in a plateau environment according to the present invention, wherein: the operating parameters of the electrical equipment are adjusted in real time according to the trained adaptive compensation model to generate optimized operating parameters of the electrical equipment, the specific steps are as follows:

[0015] Real-time collected plateau environmental data and electrical equipment operating parameters are used to generate state vectors through timestamp alignment and feature engineering.

[0016] Based on the trained adaptive compensation model, the state vector is input and combined with the action decoding method to generate the adjustment value of the operating parameters of the electrical equipment;

[0017] According to the decoded electrical equipment operating parameter adjustment value, the electrical equipment operating parameters are adjusted in real time to generate optimized electrical equipment operating parameters.

[0018] As a preferred solution of the method for monitoring the performance of electrical equipment based on plateau environment described in the present invention, the optimized operating parameters of the electrical equipment are combined with plateau environment data to generate a time series data set through data preprocessing. The specific steps are as follows:

[0019] The optimized electrical equipment operating parameters and plateau environmental data were cleaned and normalized, and timestamps were aligned to generate a time series dataset.

[0020] Based on the time series dataset, data is divided into training set, validation set and test set through data segmentation.

[0021] As a preferred solution of the method for monitoring electrical equipment performance in a plateau environment according to the present invention, the electrical equipment performance prediction model is constructed and trained by combining a time series analysis algorithm with a time series data set. The specific steps are as follows:

[0022] The LSTM algorithm is used as the framework of the electrical equipment performance prediction model. The input layer, LSTM hidden layer, fully connected layer, and output layer are defined using a time series dataset to construct the electrical equipment performance prediction model.

[0023] Based on the constructed electrical equipment performance prediction model, the electrical equipment performance prediction model is trained using the training set, the hyperparameters of the electrical equipment performance prediction model are adjusted using the validation set, and the performance of the electrical equipment performance prediction model is evaluated using the test set to generate a trained electrical equipment performance prediction model.

[0024] As a preferred solution of the electrical equipment performance monitoring method based on plateau environment described in the present invention, wherein: the performance changes of electrical equipment are predicted in real time based on the trained electrical equipment performance prediction model, and maintenance recommendations are generated through the maintenance rule engine. The specific steps are as follows:

[0025] Deploy the trained electrical equipment performance prediction model to the monitoring unit;

[0026] Based on the plateau environmental data and electrical equipment operating parameters collected in real time by the monitoring unit, the electrical equipment performance prediction model is used to make real-time predictions on the future performance of the electrical equipment and output the performance change trend of the electrical equipment;

[0027] Use the maintenance rules engine to match performance trends of electrical equipment with maintenance rules and generate maintenance recommendations.

[0028] In the second aspect, the present invention provides an electrical equipment performance monitoring system based on a plateau environment, including a data acquisition module, a compensation model module, a parameter optimization module, a data generation module and a predictive maintenance module; the data acquisition module is used to collect plateau environmental data in real time and monitor the operating parameters of electrical equipment; the compensation model module is used to construct and train an adaptive compensation model based on plateau environmental data and electrical equipment operating parameters through reinforcement learning combined with an adaptive compensation algorithm; the parameter optimization module is used to adjust the operating parameters of electrical equipment in real time according to the trained adaptive compensation model, and generate optimized operating parameters of electrical equipment; the data generation module is used to combine the optimized operating parameters of electrical equipment with plateau environmental data, and generate a time series data set through data preprocessing; the predictive maintenance module is used to construct and train an electrical equipment performance prediction model through an LSTM algorithm combined with a time series data set, and predict the performance changes of electrical equipment in real time according to the trained performance prediction model, and generate maintenance recommendations through a maintenance rule engine.

[0029] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for monitoring electrical equipment performance based on a plateau environment as described in the first aspect of the present invention is implemented.

[0030] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for monitoring electrical equipment performance based on a plateau environment as described in the first aspect of the present invention.

[0031] The beneficial effects of the present invention are: by combining reinforcement learning with an adaptive compensation algorithm, the monitoring accuracy and prediction reliability of equipment operating parameters are improved; combined with a maintenance rule engine, maintenance recommendations can be dynamically generated based on the real-time status of the equipment and environmental changes, thereby improving the accuracy and timeliness of maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 This is a flow chart of the method for monitoring electrical equipment performance in a plateau environment in Example 1.

[0034] Figure 2 This is a schematic diagram of the electrical equipment performance monitoring system based on a plateau environment in Example 1.

[0035] Figure 3 This is a flow chart of the adaptive compensation model training of the electrical equipment performance monitoring method based on plateau environment in Example 1.

[0036] Figure 4 This is a flowchart for constructing an LSTM performance prediction model for the electrical equipment performance monitoring method based on a plateau environment in Example 1. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0039] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0040] Example 1, reference Figures 1 to 4 , which is the first embodiment of the present invention, provides a method for monitoring the performance of electrical equipment in a plateau environment, comprising the following steps:

[0041] S1. Real-time collection of plateau environmental data and monitoring of electrical equipment operating parameters;

[0042] The plateau environmental data include air pressure, humidity, precipitation, snow accumulation, dust concentration and particulate matter concentration;

[0043] The operating parameters of the electrical equipment include temperature, voltage, current, power, insulation resistance, load rate, fault code and energy consumption.

[0044] It should be noted that the collection of plateau environmental data and electrical equipment operating parameters is achieved by deploying a variety of high-precision sensors (such as air pressure sensors, temperature sensors, UV sensors, acceleration sensors, etc.). Sensors are installed around the equipment and at key locations inside to collect real-time environmental data such as air pressure, temperature, humidity, UV intensity, wind speed, precipitation, snow accumulation, air oxygen content, dust and particulate matter concentration, as well as equipment operating parameters such as temperature, voltage, current, power, frequency, vibration, noise, insulation resistance, load rate, operating time, fault codes, energy consumption and operating mode; data is transmitted to the data acquisition module via I2C, SPI, analog signals or digital signals, and a comprehensive data set is finally generated for equipment performance monitoring and optimization. This process ensures the comprehensiveness, real-timeness and accuracy of the data, providing reliable data support for performance monitoring and predictive maintenance of electrical equipment in plateau environments.

[0045] S2. Based on plateau environmental data and electrical equipment operating parameters, an adaptive compensation model is constructed and trained through reinforcement learning combined with an adaptive compensation algorithm.

[0046] The state space is constructed using plateau environmental data and electrical equipment operating parameters. The adjustment range of the equipment's adjustable parameters is defined as the action space. Based on equipment performance changes, energy consumption, and failure risk, a weighted synthesis method is used to generate a reward function.

[0047] Furthermore, by integrating plateau environmental data and electrical equipment operating parameters, a multidimensional state vector is constructed as a state space to describe the comprehensive operating status of the equipment. Adjustable equipment parameters are identified and their adjustment ranges are defined, and action vectors are constructed as an action space to describe the adjustment strategy for equipment operating parameters. A weighted synthesis method is used to generate a reward function based on equipment performance changes, energy consumption, and failure risk. Performance changes, energy consumption, and failure risk are quantified using squared error, exponential functions, and logarithmic functions, respectively, and then comprehensively evaluated using weight coefficients. This process enables accurate description of equipment operating status, flexible definition of parameter adjustments, and quantitative evaluation of optimization objectives, providing a complete framework for equipment performance optimization based on reinforcement learning.

[0048] Build a reinforcement learning framework based on the defined state space, action space, and reward function;

[0049] Based on the reinforcement learning framework, the deep Q network is used as the core algorithm combined with the adaptive PID control algorithm to generate an adaptive compensation model, and the adaptive compensation model is trained through the transfer learning strategy.

[0050] It should be noted that the transfer learning strategy refers to the transfer of model parameters and knowledge trained in one environment or task to another related but different environment or task to accelerate model training and improve its performance in the new environment. In the performance monitoring of electrical equipment in plateau environments, the specific application of the transfer learning strategy includes the following steps: first, the adaptive compensation model is trained in a standard environment or a specific plateau environment to learn the relationship between equipment operating parameters and environmental factors; then, the trained model parameters are transferred to the new plateau environment and fine-tuned using a small amount of data from the new environment to adapt to the particularities of the new environment (such as different air pressure, temperature, humidity, etc.); finally, through continuous monitoring and feedback, the model parameters are further optimized to ensure its adaptability and generalization ability in the new environment. The core advantage of the transfer learning strategy is that it can fully utilize existing knowledge, reduce dependence on new environment data, shorten model training time, and improve the stability and reliability of the model in different environments.

[0051] Furthermore, the team integrated environmental data and equipment parameters to construct a state space, identified adjustable parameters to define the action space, and designed a reward function to evaluate the effectiveness of actions. Based on a reinforcement learning framework, the team used a deep Q-network as the core algorithm, trained an adaptive compensation model through an experience replay mechanism, and combined it with an adaptive PID control algorithm to generate a precise device operating parameter adjustment strategy, generating an adaptive compensation model. The adaptive compensation model was fine-tuned through a transfer learning strategy to improve its adaptability and generalization capabilities in new environments. Finally, the performance of the adaptive compensation model was evaluated and deployed to actual equipment to achieve dynamic optimization of equipment performance. This process, through the synergy of the deep Q-network and the adaptive PID control algorithm, as well as the application of a transfer learning strategy, provides an efficient and accurate solution for performance monitoring and optimization of electrical equipment in plateau environments.

[0052] S3. Adjust the operating parameters of the electrical equipment in real time according to the trained adaptive compensation model to generate optimized operating parameters of the electrical equipment;

[0053] Real-time collected plateau environmental data and electrical equipment operating parameters are used to generate state vectors through timestamp alignment and feature engineering.

[0054] The specific process is as follows: first, environmental data and equipment operating parameters are collected and timestamped for each piece of data; second, the environmental data and equipment operating parameters are aligned in chronological order based on the timestamps; then, features are directly extracted from the collected data, and dynamic and composite features are constructed based on the time series data; finally, the aligned data is combined with the extracted and constructed features to generate a multidimensional state vector at each time point, which describes the comprehensive operating state of the equipment at that moment. This process, through direct feature extraction and dynamic and composite feature construction, ensures the comprehensiveness and effectiveness of the state vector, providing high-quality input data for equipment performance monitoring and optimization.

[0055] Based on the trained adaptive compensation model, the state vector is input and combined with the action decoding method to generate the adjustment value of the operating parameters of the electrical equipment. The expression is:

[0056]

[0057] Where Δp is the adjustment value of the operating parameters of the electrical equipment, o is the number of decoders, l is the index variable of the number of l-th decoders, y is the number of cluster centers, j is the index variable of the j-th cluster center, β is the learning rate parameter, c j is the jth cluster center, c v The vth cluster center, s is the state vector, W l is the weight matrix of the l-th action decoder, b l is the bias term of the l-th action decoder, and a is the action vector.

[0058] The specific process is as follows: First, the current state vector, including plateau environmental data and equipment operating parameters, is input; second, the similarity weights between the current state and multiple cluster centers are calculated, and the contributions of different cluster centers are dynamically allocated; then, an action vector is generated and a nonlinear transformation is performed to ensure that the adjustment value is within a reasonable range; then, a weighted summation is performed to generate the final adjustment value, taking into account the similarity between the current state and multiple cluster centers; finally, the adjustment value is applied to the equipment operating parameters to optimize equipment performance in real time. This method can accurately adapt to the complexity of the plateau environment, improve the operating efficiency and reliability of equipment, and reduce energy consumption and the risk of failure.

[0059] According to the decoded electrical equipment operating parameter adjustment value, the electrical equipment operating parameters are adjusted in real time to generate optimized electrical equipment operating parameters.

[0060] The specific process is as follows: first, the decoded adjustment values are obtained from the adaptive compensation model. These values are calculated based on the current state vector. Second, the adjustment values are range-verified to ensure that they comply with the safe operating range of the equipment, and limiting is performed if necessary. Next, the verified adjustment values are applied to the equipment's operating parameters, updating key parameters such as voltage, current, and frequency in real time. Then, after adjusting the parameters, the equipment's operating status and performance changes are monitored in real time to evaluate whether the adjustment effect has met expectations. Finally, the adjusted equipment operating status and performance data are fed back to the adaptive compensation model for iterative optimization until the equipment performance reaches optimal levels. This process ensures efficient and stable operation of the equipment in plateau environments through dynamic adjustment and real-time monitoring, while reducing energy consumption and the risk of failure.

[0061] S4. Combine the optimized electrical equipment operating parameters with the plateau environmental data and generate a time series dataset through data preprocessing;

[0062] The optimized electrical equipment operating parameters and plateau environmental data were cleaned and normalized, and timestamps were aligned to generate a time series dataset.

[0063] The specific process is as follows: First, the data is cleaned to address missing values, outliers, and noise to ensure data integrity and reliability. Second, the cleaned data is normalized to convert data of different dimensions and ranges to a unified scale to ensure comparability between equipment operating parameters and plateau environmental data. Then, through timestamp alignment, the equipment operating parameters and plateau environmental data are aligned in the temporal dimension, ensuring that the time of each record matches. Finally, the aligned data is integrated into a time series dataset and saved as a structured file for subsequent analysis and modeling. This process, through rigorous data processing steps, ensures the accuracy, consistency, and usability of the time series dataset, providing a high-quality data foundation for electrical equipment performance prediction and optimization.

[0064] Based on the time series dataset, data is divided into training set, validation set and test set through data segmentation.

[0065] The specific process is as follows: First, load the generated time series dataset to ensure data integrity and correct format. Second, determine the split ratio of the training set, validation set, and test set based on actual needs and data volume. Typically, the splits are chronologically to avoid random splits that disrupt temporal dependencies. The specific split ratios are 70% for the training set, 15% for the validation set, and 15% for the test set. Next, select the training set, validation set, and test set sequentially from the beginning of the dataset in chronological order, ensuring that the timestamps of each record are continuous. Next, during the splitting process, ensure data integrity to avoid data fragmentation or loss of temporal dependencies due to splitting. Finally, save the split training set, validation set, and test set as separate files. This process ensures the integrity and availability of time series data through rigorous data splitting steps, providing a high-quality data foundation for the training and evaluation of electrical equipment performance prediction models.

[0066] S5. Build and train an electrical equipment performance prediction model by combining the LSTM algorithm with a time series dataset. Based on the trained electrical equipment performance prediction model, predict the performance changes of electrical equipment in real time and generate maintenance recommendations through the maintenance rule engine.

[0067] The LSTM algorithm is used as the framework of the electrical equipment performance prediction model. The input layer, LSTM hidden layer, fully connected layer, and output layer are defined using a time series dataset to construct the electrical equipment performance prediction model.

[0068] The specific process is as follows: first, load the time series dataset, ensuring that the data format is correct and the time dimension is continuous; second, define the input layer, setting the shape of the input layer to (time step, feature dimension), for example, (30, 10); then, build the LSTM hidden layer, set the number of neurons (such as 64) and select the return sequence method to capture the long-term dependencies in the time series data; then, add a fully connected layer, set the number of neurons (such as 32) and select the activation function (such as ReLU) to map the output of the LSTM layer to the target output space; then, define the output layer, ensuring that the shape of the output layer is consistent with the prediction target, for example, containing 1 neuron and using a linear activation function; finally, compile the electrical equipment performance prediction model, define the loss function (such as mean squared error), optimizer (such as Adam), and evaluation metric (such as root mean squared error). This process, through strict definition steps, ensures that the electrical equipment performance prediction model can efficiently capture the characteristics of time series data and provide high-precision electrical equipment performance prediction model support for electrical equipment performance prediction.

[0069] Based on the constructed electrical equipment performance prediction model, the electrical equipment performance prediction model is trained using the training set, the hyperparameters of the electrical equipment performance prediction model are adjusted using the validation set, and the performance of the electrical equipment performance prediction model is evaluated using the test set to generate a trained electrical equipment performance prediction model.

[0070] The specific process is as follows: first, load the training set to ensure that the data format is correct and the time dimension is continuous; second, use the training set to train the electrical equipment performance prediction model, set the batch size and number of training rounds, and monitor the changes in training loss; then, load the validation set to ensure that the data format is correct and the time dimension is continuous; then, use the validation set to adjust the hyperparameters of the electrical equipment performance prediction model, including the learning rate, batch size, number of neurons in the LSTM layer, number of neurons in the fully connected layer, etc., and select the optimal hyperparameter combination based on the performance on the validation set; then, load the test set to ensure that the data format is correct and the time dimension is continuous; next, use the test set to evaluate the performance of the trained electrical equipment performance prediction model, and calculate the error or accuracy between the predicted results and the actual values, such as calculating the root mean square error (RMSE); finally, save the trained electrical equipment performance prediction model as a file for subsequent deployment and use. This process ensures that the electrical equipment performance prediction model can efficiently capture the characteristics of time series data and provide high-precision prediction results through rigorous training, tuning, and evaluation steps.

[0071] Deploy the trained electrical equipment performance prediction model to the monitoring unit;

[0072] The specific process is as follows: first, load the trained electrical equipment performance prediction model to ensure that the structure and weight of the electrical equipment performance prediction model are complete; second, configure the operating environment in the monitoring unit, including installing necessary software and libraries, and checking whether the hardware resources meet the requirements; then, integrate the loaded electrical equipment performance prediction model into the monitoring unit to ensure that the electrical equipment performance prediction model can receive real-time data input and output prediction results; then, set the data acquisition module of the monitoring unit to ensure that the electrical equipment operating parameters and plateau environmental data can be obtained in real time, and preprocess the real-time data; then, input the preprocessed real-time data into the electrical equipment performance prediction model, run the electrical equipment performance prediction model and generate prediction results; next, display the prediction results output by the electrical equipment performance prediction model in real time on the interface of the monitoring unit, and store the prediction results in the database; then, set the monitoring module in the monitoring unit to monitor the operating status and performance of the electrical equipment performance prediction model in real time; finally, update the electrical equipment performance prediction model regularly according to actual operating conditions and new data to ensure prediction accuracy and adaptability. This process, through strict deployment and monitoring steps, ensures that the electrical equipment performance prediction model can run stably in the monitoring unit and provide high-precision real-time prediction results.

[0073] Based on the plateau environmental data and electrical equipment operating parameters collected in real time by the monitoring unit, the future performance of the electrical equipment is predicted in real time through the electrical equipment performance prediction model, and the performance change trend of the electrical equipment is output. The expression is:

[0074]

[0075] Where R(t+k) is the performance prediction value of the electrical equipment at the future time point t+k, σ is the Sigmoid activation function, t is the current time point, k is the length of the future time period, n is the number of time steps, i is the index variable of the time step, and X i is the input vector representing the i-th time step, m i is the weight representing the i-th time step, b1 is the bias term of linear transformation, b2 is the bias term of weighted summation, W1 is the weight matrix of linear transformation, W2 is the weight matrix of dynamic weight adjustment, h t is the hidden state of the electrical equipment performance prediction model at time step t, ∈ is the smoothing factor;

[0076] The specific process is as follows: First, the data acquisition module of the monitoring unit is used to obtain plateau environmental data and electrical equipment operating parameters in real time; second, the real-time collected data is cleaned and normalized to ensure data quality and meet the input requirements of the electrical equipment performance prediction model; then, according to the time step requirements of the electrical equipment performance prediction model, the preprocessed data is constructed as a time series input to ensure the continuity of the time dimension; then, the constructed time series is input into the electrical equipment performance prediction model, and the electrical equipment performance prediction model is run to generate the prediction results of future performance; then, the prediction results output by the electrical equipment performance prediction model are displayed in real time on the interface of the monitoring unit and compared with historical data to generate a equipment performance change trend chart; next, the prediction results are stored in the database for subsequent analysis and historical record query. This process ensures the real-time prediction of the future performance of electrical equipment and the intuitive display of the changing trends of equipment performance through rigorous data collection, preprocessing, model prediction, and result output steps.

[0077] Use the maintenance rule engine to match the performance change trends of electrical equipment with maintenance rules and generate maintenance recommendations

[0078] The specific process is as follows: first, load the performance trend data to ensure data integrity and temporal continuity; second, load the maintenance rule base to ensure the rules in the rule base are complete and logically clear; then, match the performance trend data with the rules in the maintenance rule base, checking each rule to see if the performance trend meets the rule conditions; then, generate corresponding maintenance recommendations based on the matched rules; then, output the generated maintenance recommendations to the monitoring unit interface and store them in the database; next, set up a monitoring module in the monitoring unit to monitor the implementation of the maintenance recommendations in real time; finally, regularly update the maintenance rule base based on actual operating conditions and new data. This process ensures the accuracy and timeliness of maintenance recommendations through rigorous matching, generation, and monitoring steps, providing reliable support for the maintenance of electrical equipment.

[0079] This embodiment also provides an electrical equipment performance monitoring system based on a plateau environment, including: a data acquisition module, a compensation model module, a parameter optimization module, a data generation module and a predictive maintenance module; the data acquisition module is used to collect plateau environmental data in real time and monitor the operating parameters of electrical equipment; the compensation model module is used to construct and train an adaptive compensation model based on plateau environmental data and electrical equipment operating parameters through reinforcement learning combined with an adaptive compensation algorithm; the parameter optimization module is used to adjust the operating parameters of electrical equipment in real time according to the trained adaptive compensation model to generate optimized operating parameters of electrical equipment; the data generation module is used to combine the optimized operating parameters of electrical equipment with plateau environmental data and generate a time series data set through data preprocessing; the predictive maintenance module is used to construct and train an electrical equipment performance prediction model through an LSTM algorithm combined with a time series data set, predict the performance changes of electrical equipment in real time according to the trained performance prediction model, and generate maintenance recommendations through a maintenance rule engine.

[0080] This embodiment also provides a computer device, which is suitable for the performance monitoring method of electrical equipment based on a plateau environment, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the performance monitoring method of electrical equipment based on a plateau environment proposed in the above embodiment.

[0081] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0082] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for monitoring the performance of electrical equipment in a plateau environment as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0083] In summary, the present invention improves the monitoring accuracy and prediction reliability of equipment operating parameters through: reinforcement learning combined with an adaptive compensation algorithm; combined with a maintenance rule engine, it can dynamically generate maintenance recommendations based on the real-time status of the equipment and environmental changes, thereby improving the accuracy and timeliness of maintenance strategies.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring the performance of electrical equipment in a plateau environment, characterized by: include, Collect plateau environmental data in real time and monitor the operating parameters of electrical equipment; Based on plateau environmental data and electrical equipment operating parameters, an adaptive compensation model is constructed and trained through reinforcement learning combined with an adaptive compensation algorithm. According to the trained adaptive compensation model, the operating parameters of electrical equipment are adjusted in real time to generate optimized operating parameters of electrical equipment; The optimized operating parameters of electrical equipment are combined with plateau environmental data to generate a time series dataset through data preprocessing. By combining the LSTM algorithm with time series datasets, we build and train an electrical equipment performance prediction model. Based on the trained performance prediction model, we can predict the performance changes of electrical equipment in real time, and generate maintenance recommendations through the maintenance rule engine.

2. The method for monitoring electrical equipment performance in a plateau environment according to claim 1, wherein: The plateau environmental data include air pressure, humidity, precipitation, snow accumulation, dust concentration and particulate matter concentration; The operating parameters of the electrical equipment include temperature, voltage, current, power, insulation resistance, load rate, fault code and energy consumption.

3. The method for monitoring electrical equipment performance in a plateau environment according to claim 2, wherein: Based on plateau environmental data and electrical equipment operating parameters, the adaptive compensation model is constructed and trained by combining reinforcement learning with an adaptive compensation algorithm. The specific steps are as follows: The state space is constructed using plateau environmental data and electrical equipment operating parameters. The adjustment range of the equipment's adjustable parameters is defined as the action space. Based on equipment performance changes, energy consumption, and failure risk, a weighted synthesis method is used to generate a reward function. Build a reinforcement learning framework based on the defined state space, action space, and reward function; Based on the reinforcement learning framework, the deep Q network is used as the core algorithm combined with the adaptive PID control algorithm to generate an adaptive compensation model, and the adaptive compensation model is trained through the transfer learning strategy.

4. The method for monitoring electrical equipment performance in a plateau environment according to claim 3, wherein: The specific steps of adjusting the operating parameters of the electrical equipment in real time based on the trained adaptive compensation model to generate optimized operating parameters of the electrical equipment are as follows: Real-time collected plateau environmental data and electrical equipment operating parameters are used to generate state vectors through timestamp alignment and feature engineering. Based on the trained adaptive compensation model, the state vector is input and combined with the action decoding method to generate the adjustment value of the operating parameters of the electrical equipment; According to the decoded electrical equipment operating parameter adjustment value, the electrical equipment operating parameters are adjusted in real time to generate optimized electrical equipment operating parameters.

5. The method for monitoring electrical equipment performance in a plateau environment according to claim 4, wherein: The optimized electrical equipment operating parameters are combined with plateau environmental data to generate a time series data set through data preprocessing. The specific steps are as follows: The optimized electrical equipment operating parameters and plateau environmental data were cleaned and normalized, and timestamps were aligned to generate a time series dataset. Based on the time series dataset, data is divided into training set, validation set and test set through data segmentation.

6. The method for monitoring electrical equipment performance in a plateau environment according to claim 5, wherein: The time series analysis algorithm is combined with the time series data set to build and train the electrical equipment performance prediction model. The specific steps are as follows: The LSTM algorithm is used as the framework of the electrical equipment performance prediction model. The input layer, LSTM hidden layer, fully connected layer, and output layer are defined using a time series dataset to construct the electrical equipment performance prediction model. Based on the constructed electrical equipment performance prediction model, the electrical equipment performance prediction model is trained using the training set, the hyperparameters of the electrical equipment performance prediction model are adjusted using the validation set, and the performance of the electrical equipment performance prediction model is evaluated using the test set to generate a trained electrical equipment performance prediction model.

7. The method for monitoring electrical equipment performance in a plateau environment according to claim 6, wherein: The performance prediction model of the electrical equipment is trained to predict the performance changes of the electrical equipment in real time, and maintenance recommendations are generated through the maintenance rule engine. The specific steps are as follows: Deploy the trained electrical equipment performance prediction model to the monitoring unit; Based on the plateau environmental data and electrical equipment operating parameters collected in real time by the monitoring unit, the electrical equipment performance prediction model is used to make real-time predictions on the future performance of the electrical equipment and output the performance change trend of the electrical equipment; Use the maintenance rules engine to match performance trends of electrical equipment with maintenance rules and generate maintenance recommendations.

8. A plateau-based electrical equipment performance monitoring system, based on the plateau-based electrical equipment performance monitoring method according to any one of claims 1 to 7, characterized in that: Including data acquisition module, compensation model module, parameter optimization module, data generation module and predictive maintenance module; Data acquisition module, used to collect plateau environmental data in real time and monitor the operating parameters of electrical equipment; The compensation model module is used to build and train an adaptive compensation model based on plateau environmental data and electrical equipment operating parameters through reinforcement learning combined with an adaptive compensation algorithm. Parameter optimization module, used to adjust the operating parameters of electrical equipment in real time based on the trained adaptive compensation model and generate optimized operating parameters of electrical equipment; The data generation module is used to combine the optimized operating parameters of electrical equipment with plateau environmental data and generate a time series data set through data preprocessing; The predictive maintenance module is used to build and train a performance prediction model for electrical equipment by combining the LSTM algorithm with a time series dataset. Based on the trained performance prediction model, it predicts performance changes of electrical equipment in real time and generates maintenance recommendations through the maintenance rule engine.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for monitoring performance of electrical equipment based on a plateau environment according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for monitoring performance of electrical equipment based on a plateau environment according to any one of claims 1 to 7 are implemented.

Images