Intelligent control method and system based on operation state of power equipment

Abstract

The invention discloses an intelligent control method and system based on the operation state of power equipment, and relates to the technical field of power equipment safety, and the control method comprises the steps of data collection, feature extraction, model establishment, intelligent control, model optimization and maintenance and storage. The method has the advantages that the power equipment is divided into high, medium and low risk levels, and the data acquisition frequency is dynamically adjusted according to the fault probability and the deviation degree of the key operation parameters, so that targeted data monitoring can be realized, the data acquisition efficiency is improved, the equipment state change can be responded in time, and the reliability of the power equipment is improved. The method effectively reduces the risk of fault occurrence, analyzes the feature data through employing the K-means clustering algorithm, can accurately divide the data into a normal data set, an abnormal data set and a fault data set, improves the intelligent degree of data processing, and also provides a high-quality data basis for subsequent model training.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment safety, and in particular to an intelligent control method and system based on the operating status of power equipment. Background Art

[0002] Power equipment is equipment used for transformation, transmission, and distribution in power systems, including generators, distribution equipment, transformers, mutual inductors, etc. In the power industry, the operating status monitoring and management of power equipment is an important part of ensuring the stable operation of the power system and the reliability of power supply. Traditionally, data collection and monitoring of power equipment often use a fixed frequency, resulting in the omission of key data and the inability to respond to changes in equipment status in a timely manner, increasing the risk of equipment failure. To this end, we propose an intelligent control method and system based on the operating status of power equipment. Summary of the invention

[0003] The object of the present invention is to provide an intelligent control method and system based on the operating status of power equipment.

[0004] In order to solve the problems raised in the above background technology, the present invention provides the following technical solutions: an intelligent control method based on the operating state of power equipment, comprising the following steps: Step 1: Data collection: Collect data from power equipment through sensors and monitoring equipment, classify power equipment according to risk level, set initial data collection frequency according to category, and adjust data collection frequency according to changes in equipment operating status; Step 2: feature extraction: extracting temperature features, electrical parameter features, and vibration feature data of the power equipment from the original data, and dividing the data into a normal data set, an abnormal data set, and a fault data set; Step 3: Model establishment: Use the extracted feature data to establish a power equipment status assessment model, and import the real-time feature data into the model for evaluation; Step 4: Intelligent control, adjusting equipment operating parameters according to the results of the power equipment status assessment model; Step 5: Model optimization: regularly collect feedback data from model evaluation and analyze the accuracy of the model’s judgment of equipment status; Step 6: Maintenance and storage: Regularly maintain power equipment and sensors, and store various collected data.

[0005] As a further solution of the present invention: In the step 1, the specific method of classifying the power equipment is to divide the power equipment into three risk levels: high-risk equipment, medium-risk equipment and low-risk equipment. The high-risk equipment collects equipment parameters in real time, the initial data collection frequency of the medium-risk equipment is 5 minutes, and the initial data collection frequency of the low-risk equipment is 30 minutes. The data collection frequency of the equipment is dynamically adjusted according to the operating status of the subsequent equipment using a formula. The specific formula is as follows: , in, represents the adjusted frequency, represents the initial data collection frequency, and represents the adjustment coefficient, P represents the failure probability, D represents the deviation of the key operating parameters of the equipment from the normal range. The value of P is obtained by collecting the number of failures of the same type of power equipment in all power systems in the region in the past year and dividing the number of failures by the total number of units. If D>10%, the data collection frequency of the equipment is adjusted to real-time collection.

[0006] As a further solution of the present invention: In the step 2, the specific method of feature extraction is to use the K-means clustering algorithm to perform cluster analysis on the feature data, divide the data set into a normal data set, an abnormal data set and a fault data set according to the clustering results, and divide each data set into a training set, a test set and a validation set according to a ratio of 7:2:1.

[0007] As a further solution of the present invention: In the step three, the specific method of model establishment is to use an artificial neural network to establish a power equipment status assessment model, use training set data to train the model, test set data to evaluate the model performance, and verification set data to further verify the model stability. The trained and verified model is deployed to the power equipment status assessment system. After the real-time feature data is imported into the model, the model immediately outputs the equipment status assessment result.

[0008] As a further solution of the present invention: In step 4, the specific method of intelligent control is: if the equipment is in a normal state, the performance of the equipment is evaluated every week, and the normal operating status parameter range of the equipment is updated; if the equipment is in an abnormal state, the location, connected equipment and operating parameter information of the abnormal state equipment are retrieved, the abnormal degree value of the equipment is calculated, and the corresponding abnormal handling plan is executed according to the abnormal type of the equipment; if the equipment is in a fault state, emergency protection measures are initiated.

[0009] As a further solution of the present invention: the calculation formula of the abnormality degree value of the device is: , Among them, E represents the abnormality value, Indicates the current parameter detection value. It indicates the average value of the parameter under normal conditions. It represents the standard deviation of the parameter under normal operating conditions. If there are multiple parameter abnormalities in the same device, the parameter value with the highest degree of abnormality is taken as the degree of abnormality of the device. If 1 < E ≤ 2, it is determined that the device is in a mild abnormal state, the abnormal information is recorded, and a warning is issued. If 2 < E ≤ 3, it is determined that the device is in a moderate abnormal state, the status parameter data of the abnormal device is retrieved and sent to the maintenance personnel via SMS. If E > 3, it is determined that the device is in a severe abnormal state, the abnormal device is shut down, and the maintenance team is notified for emergency repair. When the device is in an abnormal state, the time T when the device switches from a mild abnormal state to a moderate abnormal state is recorded. If T is less than 3s, it is determined that the device has a fault tendency, and the fault emergency treatment plan is started.

[0010] As a further solution of the present invention: In the fifth step, the specific way of model optimization is to collect the device status data and the model prediction results every week, analyze the prediction accuracy of the model and the cases of prediction errors, and according to the analysis results, put forward optimization suggestions. After retaining the version before optimization, the optimized model is tested. If the optimization effect is not significant, it is reverted to the previous version. If the optimization effect is significant, the new version is retained.

[0011] As a further solution of the present invention: In the sixth step, the specific way of maintenance and storage is to arrange a professional team to carry out maintenance work such as cleaning the equipment, replacing aging components, checking electrical connections and operating parameters every month, and calibrating the sensors on the equipment. The collected power equipment data is classified and stored in the database according to different types, and a full backup of the data is performed once a week.

[0012] In addition, the present invention also provides an intelligent control system, including: A data acquisition module that collects power equipment data, sets an initial acquisition frequency according to the risk level, and dynamically adjusts according to the status; A feature extraction module that extracts data of temperature, electrical and vibration features, divides the data set by K-means clustering, and divides it into training, test and validation sets; A model establishment module that uses an artificial neural network to establish a status evaluation model, deploys it after training, testing and validation, and evaluates the device status in real time; An intelligent control module that adjusts the device parameters according to the evaluation results, processes abnormalities and faults, calculates the degree of abnormality value, and starts the corresponding treatment plan; A model optimization module that regularly analyzes the model prediction results, puts forward optimization suggestions, and tests the new version; A maintenance and storage module that records the logs of the regular maintenance work of power equipment and sensors, and stores various data collected during the operation of the system.

[0013] Adopting the above technical solutions, compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention can realize targeted data monitoring by dividing power equipment into high, medium and low risk levels and dynamically adjusting the data collection frequency according to the failure probability and the deviation of key operating parameters, which not only improves the efficiency of data collection, but also can respond to changes in equipment status in a timely manner, effectively reducing the risk of failure; 2. The present invention uses the K-means clustering algorithm to analyze the feature data, and can accurately divide the data into three data sets: normal, abnormal, and fault, which improves the intelligent level of data processing and provides a high-quality data basis for subsequent model training, thereby enhancing the accuracy and reliability of the power equipment status assessment model; 3. The present invention can take targeted control measures according to the different equipment states by establishing an abnormality degree calculation formula and a corresponding processing scheme, which not only realizes real-time monitoring and evaluation of the equipment operation status, but also can quickly initiate protection measures when an abnormality occurs in the equipment, thereby ensuring the safe operation of the equipment and reducing the potential losses caused by equipment failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the control method steps in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Embodiment 1

[0017] Please see attached Figure 1 The present invention provides an intelligent control method based on the operating state of electric power equipment, comprising the following steps: Step 1: Data collection: Collect data from power equipment through sensors and monitoring equipment, classify power equipment according to risk level, set initial data collection frequency according to category, and adjust data collection frequency according to changes in equipment operating status; Step 2: feature extraction: extracting temperature features, electrical parameter features, and vibration feature data of the power equipment from the original data, and dividing the data into a normal data set, an abnormal data set, and a fault data set; Step 3: Model establishment: Use the extracted feature data to establish a power equipment status assessment model, and import the real-time feature data into the model for evaluation; Step 4: Intelligent control, adjusting equipment operating parameters according to the results of the power equipment status assessment model; Step 5: Model optimization: regularly collect feedback data from model evaluation and analyze the accuracy of the model’s judgment of equipment status; Step 6: Maintenance and storage: regularly maintain power equipment and sensors, and store all types of collected data; In step 1, the specific method of classifying power equipment is to divide power equipment into three risk levels: high-risk equipment, medium-risk equipment and low-risk equipment. High-risk equipment collects equipment parameters in real time, the initial data collection frequency of medium-risk equipment is 5 minutes, and the initial data collection frequency of low-risk equipment is 30 minutes. The data collection frequency of the equipment is dynamically adjusted according to the operating status of the subsequent equipment using a formula. The specific formula is as follows: ,

[0018] in, represents the adjusted frequency, represents the initial data collection frequency, and represents the adjustment coefficient, P represents the failure probability, D represents the deviation of the key operating parameters of the equipment from the normal range. The value of P is obtained by collecting the number of failures of the same type of power equipment in all power systems in the region in the past year and dividing the number of failures by the total number of units. If D>10%, the data collection frequency of the equipment is adjusted to real-time collection.

[0019] Specific workflow: In a city's power supply system, there are a large number of power equipment, such as transformers and switch cabinets in substations, and distribution boxes in distribution networks. We classify these equipment by risk level. For example, 100 110kV transformers are located in hub substations in the core area of ​​the city. They are classified as high-risk equipment and collect data in real time. 50 transformers located in ordinary substations are classified as medium-risk equipment with an initial collection frequency of 5 minutes. 30 transformers located in small distribution stations are low-risk equipment with an initial collection frequency of 30 minutes. In the past year, there were 10 failures of the same type of transformers in the area. For one of the medium-risk transformers, the failure probability P=0.1. During operation, its oil temperature deviated from the normal range by 15%. The data collection frequency was adjusted to real-time collection according to the formula. In this way, the operation and maintenance personnel discovered the problem of excessive oil temperature in time and avoided failures that might be caused by overheating. Compared with areas that did not adopt this method, the failure rate was reduced by about 30%, effectively improving the data collection efficiency and the ability to respond to changes in equipment status.

[0020] Furthermore, by classifying power equipment into high, medium, and low risk levels and dynamically adjusting the data collection frequency according to the failure probability and the deviation degree of key operating parameters, targeted data monitoring can be achieved, which not only improves the efficiency of data collection but also promptly responds to changes in equipment status and effectively reduces the risk of failures.

[0021] Embodiment 2

[0022] Based on Embodiment 1, in Step 2, the specific method of feature extraction is to use the K-means clustering algorithm to perform clustering analysis on the feature data. According to the clustering results, the data set is divided into a normal data set, an abnormal data set, and a failure data set, and each data set is divided into a training set, a test set, and a validation set according to a ratio of 7:2:1. In Step 3, the specific method of model establishment is to use an artificial neural network to establish a power equipment status evaluation model. The training set data is used to train the model, the test set data is used to evaluate the model performance, and the validation set data is used to further verify the model stability. The trained and verified model is deployed to the power equipment status evaluation system. When real-time feature data is imported into the model, the model immediately outputs the status evaluation result of the equipment. In Step 4, the specific method of intelligent control is as follows: If the equipment is in a normal state, the performance of the equipment is evaluated every week, and the normal operating state parameter range of the equipment is updated. If the equipment is in an abnormal state, the location of the abnormal state equipment, the connected equipment, and the operating parameter information are retrieved, the abnormal degree value of the equipment is calculated, and according to the abnormal type of the equipment, the corresponding abnormal handling scheme is executed. If the equipment is in a failure state, emergency protection measures are initiated. The calculation formula for the abnormal degree value of the equipment is: , where E represents the abnormal degree value, represents the current parameter detection value, represents the average value of this parameter in the normal state, represents the standard deviation of this parameter in the normal operating state. If there are multiple parameter abnormalities in the same equipment, the parameter value with the highest abnormal degree value is taken as the abnormal degree value of the equipment. If 1 < E ≤ 2, it is determined that the equipment is in a mild abnormal state, the abnormal information is recorded, and a warning is issued. If 2 < E ≤ 3, it is determined that the equipment is in a moderate abnormal state, the abnormal equipment status parameter data is retrieved and sent to the maintenance personnel via text message. If E > 3, it is determined that the equipment is in a severe abnormal state, the abnormal equipment is shut down, and the maintenance team is notified for emergency repair. When the equipment is in an abnormal state, the time T when the equipment transitions from a mild abnormal state to a moderate abnormal state is recorded. If T is less than 3s, it is determined that the equipment has a failure tendency, and a failure emergency handling scheme is initiated.

[0023] Specific workflow: Taking the power system of a large factory as an example, the operating data of its motors, distribution cabinets and other equipment needs to be analyzed. After a large amount of raw data is collected through sensors, the K-means clustering algorithm is used to process the data. For example, for the operating data of the motor, characteristic data such as temperature, current, and vibration are extracted; During the data segmentation process, the data set was successfully divided into three categories: normal, abnormal, and fault. The normal data set accounted for 70%, including data with parameter characteristics such as the temperature range of 30-70°C, the current stable near the rated current, and the vibration amplitude less than 0.05mm during normal operation of the motor. The abnormal data set accounted for 20%, such as the data when the motor current exceeded the rated value by 10%-20% and the temperature rose to about 80°C when overloaded. The fault data set accounted for 10%, such as the vibration amplitude of the motor exceeding 0.1mm due to bearing damage and the temperature rising sharply to more than 100°C. These divided data are used to train the artificial neural network model. The model's prediction accuracy for motor failures has increased from the original 70% to 90%, providing high-quality data for subsequent power equipment status assessment model training and enhancing the model's accuracy and reliability.

[0024] Furthermore, by using the K-means clustering algorithm to analyze the feature data, the data can be accurately divided into three data sets: normal, abnormal, and faulty, which improves the intelligence of data processing and provides a high-quality data foundation for subsequent model training, thereby enhancing the accuracy and reliability of the power equipment status assessment model.

[0025] Embodiment 3

[0026] Based on the second embodiment, in step 5, the specific method of model optimization is to collect device status data and model prediction results every week, analyze the prediction accuracy of the model and the cases of prediction errors, and make optimization suggestions based on the analysis results. After retaining the version before optimization, test the optimized model. If the optimization effect is not significant, roll back to the previous version. If the optimization effect is significant, retain the new version. In step six, the specific method of maintenance and storage is to arrange a professional team to carry out maintenance work every month to clean equipment, replace aging parts, check electrical connections and operating parameters, calibrate sensors on the equipment, store the collected power equipment data in the database according to different types, and perform a full back-up of the data once a week.

[0027] Specific workflow: In the power system of an industrial park, there are various types of power equipment, such as large motors in the production workshop. Take a motor with a rated power of 500kW as an example. During normal operation, the current is 700A, the temperature is 60℃, and the vibration amplitude is 0.03mm. The standard deviations of these parameters under normal conditions are 50A, 5℃, and 0.005mm respectively. At a certain moment, the motor current detection value reaches 800A. According to the abnormality degree value calculation formula , the current abnormality value E= (800-700) / 50=2, which is in a slightly abnormal state. The system records the information and issues an early warning. Subsequently, the motor temperature rises to 90°C, and the temperature abnormality value E= (90-60) / 5=6, which is judged to be a serious abnormal state. The system immediately shuts down the motor and notifies the maintenance team. The conversion time from a slight abnormality to a moderate abnormality is 2s (less than 3s), and the fault emergency handling plan is initiated. In this way, further damage to the motor is avoided. Compared with previous control methods, the loss of major failures caused by abnormalities in the equipment is reduced by about 40%, ensuring the safe operation of the equipment.

[0028] Furthermore, by establishing a calculation formula for the degree of abnormality and a corresponding processing plan, targeted control measures can be taken according to the different equipment status, which not only realizes real-time monitoring and evaluation of the equipment operation status, but also can quickly initiate protection measures when the equipment has an abnormality, ensuring the safe operation of the equipment and reducing the potential losses caused by equipment failure.

[0029] Embodiment 4

[0030] The present invention also provides an intelligent control system, comprising: The data collection module collects power equipment data, sets the initial collection frequency according to the risk level, and dynamically adjusts it according to the status; Feature extraction module, which extracts data of temperature, electrical and vibration characteristics, and divides the data set into training, test and validation sets using K-means clustering; Model building module, which uses artificial neural network to build status assessment model, train, test, verify and deploy, and evaluate equipment status in real time; The intelligent control module adjusts equipment parameters according to the evaluation results, handles abnormalities and faults, calculates the abnormality level, and initiates the corresponding treatment plan; Model optimization module, which regularly analyzes model prediction results, makes optimization suggestions, and tests new versions; The maintenance and storage module records the logs of regular maintenance work of power equipment and sensors, as well as various types of data collected during the operation of the storage system.

[0031] Specific workflow: In the power system of a certain industrial park, the intelligent control system plays an important role. The data acquisition module lists the high-voltage transformers in the park as high-risk equipment to collect data in real time. The initial collection frequency of medium-risk equipment such as ordinary motors is set to once every 5 minutes. During operation, it is dynamically adjusted according to the fault probability and parameter deviation. For example, if a motor is overloaded, the collection frequency is accelerated; The feature extraction module extracts features from the motor data and uses K-means clustering to divide the data. The normal data set includes normal operating current, temperature and other range data, which are divided into training, test and validation sets. The model building module uses artificial neural networks to build a motor state assessment model, which is evaluated on the test set after training and evaluates the motor state in real time after deployment; The intelligent control module takes action based on the evaluation results. If the motor is normal, the performance is evaluated regularly. If it is abnormal, the abnormality level is calculated. For example, if the motor current abnormality value reaches 2.2, it is a moderate abnormality. The maintenance personnel are notified and the parameters are adjusted. Protection is activated in case of failure. The model optimization module analyzes the model prediction results and optimizes the model every week. The maintenance and storage module records the monthly maintenance logs of power equipment and sensors, stores the collected data and backs it up regularly, ensuring the stable operation of the park's power system.

[0032] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An intelligent control method based on the operating status of power equipment, characterized in that: The following steps are involved: Step 1: Data collection: Collect data from power equipment through sensors and monitoring equipment, classify power equipment according to risk level, set initial data collection frequency according to category, and adjust data collection frequency according to changes in equipment operating status; Step 2: feature extraction: extracting temperature features, electrical parameter features, and vibration feature data of the power equipment from the original data, and dividing the data into a normal data set, an abnormal data set, and a fault data set; Step 3: Model establishment: Use the extracted feature data to establish a power equipment status assessment model, and import the real-time feature data into the model for evaluation; Step 4: Intelligent control, adjusting equipment operating parameters according to the results of the power equipment status assessment model; Step 5: Model optimization: regularly collect feedback data from model evaluation and analyze the accuracy of the model’s judgment of equipment status; Step 6: Maintenance and storage: Regularly maintain power equipment and sensors, and store various collected data.

2. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In the step 1, the specific method of classifying power equipment is to divide the power equipment into three risk levels: high-risk equipment, medium-risk equipment and low-risk equipment. High-risk equipment collects equipment parameters in real time, the initial data collection frequency of medium-risk equipment is 5 minutes, and the initial data collection frequency of low-risk equipment is 30 minutes. The data collection frequency of the equipment is dynamically adjusted according to the operating status of the subsequent equipment using a formula. The specific formula is as follows: ， in, represents the adjusted frequency, represents the initial data collection frequency, and represents the adjustment coefficient, P represents the failure probability, D represents the deviation of the key operating parameters of the equipment from the normal range. The value of P is obtained by collecting the number of failures of the same type of power equipment in all power systems in the region in the past year and dividing the number of failures by the total number of units. If D>10%, the data collection frequency of the equipment is adjusted to real-time collection.

3. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In the step 2, the specific method of feature extraction is to use the K-means clustering algorithm to perform cluster analysis on the feature data, divide the data set into a normal data set, an abnormal data set and a fault data set according to the clustering results, and divide each data set into a training set, a test set and a validation set according to a ratio of 7:2:

1.

4. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In the step three, the specific method of model establishment is to use an artificial neural network to establish a power equipment status assessment model, use training set data to train the model, test set data to evaluate the model performance, and verification set data to further verify the model stability. The trained and verified model is deployed to the power equipment status assessment system. When the real-time feature data is imported into the model, the model immediately outputs the equipment status assessment results.

5. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In step 4, the specific method of intelligent control is: if the equipment is in a normal state, the performance of the equipment is evaluated every week, and the normal operating status parameter range of the equipment is updated. If the equipment is in an abnormal state, the location, connected equipment and operating parameter information of the abnormal state equipment are retrieved, and the abnormal degree value of the equipment is calculated. According to the abnormal type of the equipment, the corresponding abnormal handling plan is executed. If the equipment is in a fault state, emergency protection measures are initiated.

6. The intelligent control method based on the operating status of the power equipment according to claim 5 is characterized in that: The calculation formula of the abnormality degree value of the device is: ， Among them, E represents the abnormal degree value, represents the current parameter detection value, represents the average value of the parameter under normal conditions, represents the standard deviation of the parameter under normal operating conditions. If there are multiple parameter abnormalities in the same device, the parameter value with the highest abnormal degree value is taken as the abnormal degree value of the device. If 1 < E ≤ 2, it is judged that the device is in a slight abnormal state, the abnormal information is recorded, and a warning is issued. If 2 < E ≤ 3, it is judged that the device is in a moderate abnormal state, the parameter data of the abnormal device state is retrieved and sent to the maintenance personnel by text message. If E > 3, it is judged that the device is in a serious abnormal state, the abnormal device is shut down, and the maintenance team is notified for emergency repair. When the device is in an abnormal state, the time T when the device switches from a slight abnormal state to a moderate abnormal state is recorded. If T is less than 3s, it is judged that the device has a fault tendency, and the fault emergency treatment plan is started.

7. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In step five, the specific method of model optimization is to collect device status data and model prediction results every week, analyze the prediction accuracy of the model and cases of prediction errors, and make optimization suggestions based on the analysis results. After retaining the version before optimization, test the optimized model. If the optimization effect is not significant, roll back to the previous version. If the optimization effect is significant, retain the new version.

8. The intelligent control method based on the operating status of the power equipment according to claim 1 is characterized in that: In step six, the specific method of maintenance and storage is to arrange a professional team every month to carry out maintenance work such as cleaning equipment, replacing aging parts, checking electrical connections and operating parameters, calibrating sensors on the equipment, and storing the collected power equipment data in the database according to different types, and performing a full back-up of the data once a week.

9. The intelligent control system for the intelligent control method based on the operating status of the power equipment according to any one of claims 1 to 8, characterized in that: The system comprises: The data collection module collects power equipment data, sets the initial collection frequency according to the risk level, and dynamically adjusts it according to the status; Feature extraction module, which extracts data of temperature, electrical and vibration characteristics, and divides the data set into training, test and validation sets using K-means clustering; Model building module, which uses artificial neural network to build status assessment model, train, test, verify and deploy, and evaluate equipment status in real time; The intelligent control module adjusts equipment parameters according to the evaluation results, handles abnormalities and faults, calculates the abnormality level, and initiates the corresponding treatment plan; Model optimization module, which regularly analyzes model prediction results, makes optimization suggestions, and tests new versions; The maintenance and storage module records the logs of regular maintenance work of power equipment and sensors, as well as various types of data collected during the operation of the storage system.

Images