Risk assessment device for energy storage equipment based on artificial intelligence technology and use method

Abstract

The invention discloses a risk assessment device for energy storage equipment based on an artificial intelligence technology and a use method, and relates to the technical field of risk assessment devices for energy storage equipment.The risk assessment device comprises an energy storage assembly and a risk assessment system, the energy storage assembly comprises a cabinet body, a sealing door, a main circuit breaker, a battery cluster and a PCS, and a diagnosis assembly and a monitoring assembly are installed in an inner cavity of the cabinet body; the diagnosis assembly comprises an industrial personal computer, a server and a coordination controller, and the monitoring assembly comprises a BMS, a smoke sensor, a temperature sensor and an optical sensor. The diagnosis assembly and the monitoring assembly are arranged to perform risk assessment on the energy storage equipment, the BMS, the smoke sensor, the temperature sensor and the optical sensor are used for collecting operation data of the energy storage equipment, the coordination controller is used for integrating the collected data, and the server is used for storing the operation data. And the industrial personal computer analyzes and monitors the running state of the energy storage equipment in real time by using the risk assessment system, so that whether the energy storage equipment breaks down or not can be judged in time.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment devices for energy storage equipment, and in particular to a risk assessment device for energy storage equipment based on artificial intelligence technology and a method of using the same. Background Art

[0002] Energy storage equipment generally refers to equipment that can store electrical energy and release it when needed. Energy storage systems play an important role in various application scenarios, especially in power systems. They can help balance supply and demand, improve grid stability, reduce carbon emissions and promote the use of renewable energy. The rapid development of big data and artificial intelligence technologies has provided new opportunities for the intelligent control of energy storage systems. By utilizing big data analysis and artificial intelligence algorithms, real-time monitoring, prediction and optimization control of energy storage systems can be achieved.

[0003] By using big data and artificial intelligence technologies, the energy storage system can be monitored and analyzed in real time, and the control strategy can be automatically adjusted to improve the response speed of the energy storage system, enabling it to store and release electricity within milliseconds. However, current energy storage equipment lacks in-depth analysis and mining of energy storage system operating data during use, making it impossible to discover potential risks and faults, and thus unable to prevent faults in advance. Summary of the invention

[0004] The purpose of the present invention is to provide a risk assessment device and a method for using energy storage equipment based on artificial intelligence technology, which has the advantage of analyzing and evaluating the historical data and operating data of the energy storage equipment, so as to be able to predict faults, and solves the problem that the current energy storage equipment lacks in-depth analysis and mining of the energy storage system operating data during use, thereby failing to discover potential risks and faults, and thus failing to prevent faults in advance.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a risk assessment device for energy storage equipment based on artificial intelligence technology, comprising an energy storage component and a risk assessment system, wherein the energy storage component comprises a cabinet, a sealed door, a main circuit breaker, a battery cluster and a PCS, and a diagnostic component and a monitoring component are installed in the inner cavity of the cabinet, wherein the diagnostic component comprises an industrial computer, a server and a coordination controller, and the monitoring component comprises a BMS, a smoke sensor, a temperature sensor and an optical sensor, wherein the diagnostic component is used to analyze and evaluate the operating data of the energy storage equipment, the BMS is used to collect the operating data of the battery cluster, and the smoke sensor, the temperature sensor and the optical sensor are used to collect the operating data inside the cabinet.

[0006] The risk assessment system includes data collection, feature extraction, feature selection, model learning, model training, model evaluation and fault diagnosis. The data collection collects the operating data of the energy storage device and the battery cluster through the monitoring components. The feature extraction and feature selection are used to extract the characteristic information related to the fault in the operating data. The model learning, model training and model evaluation are used to learn the patterns and trends in the data and predict the time point when the equipment fails. The fault diagnosis uses the model to predict the possible failure of the equipment.

[0007] As a further preferred embodiment of the present invention, the diagnostic component and the monitoring component are both installed in the inner cavity of the cabinet, the sealed door is hinged to the cabinet by a hinge, a display is installed on the surface of the sealed door, a communication module is also installed in the inner cavity of the cabinet, and an automatic fire extinguisher is installed at the rear end of the inner cavity of the cabinet.

[0008] As a further preferred embodiment of the present invention, the output ends of the BMS, smoke sensor, temperature sensor and optical sensor are all connected to the input end of the coordination controller, the output end of the coordination controller is respectively connected to the input ends of the industrial computer and the PCS, the output end of the industrial computer is respectively connected to the input ends of the display, communication module, automatic fire extinguisher and main circuit breaker, and the output end and input end of the industrial computer are interconnected with the output end and input end of the server.

[0009] As a further optimization of the present invention, the model learning adopts both machine learning and deep learning, the machine learning adopts a random forest regression algorithm, and the deep learning adopts a convolutional neural network algorithm.

[0010] As a further optimization of the present invention, the steps of the random forest regression algorithm are as follows:

[0011] S1, randomly extract m sample points from the training sample set S to obtain a new S...Sn sub-training set;

[0012] S2. Use the sub-training set to train a CART regression tree, and the splitting rule for each node is to first randomly select k features from all features, and then select the optimal split point from these k features to divide the left and right subtrees;

[0013] S3. Through step b, many CART regression tree models can be generated;

[0014] S4. The final prediction result of each CART regression tree is the mean of the leaf nodes reached by the sample points;

[0015] S5. The final prediction result of random forest is the mean of the prediction results of all CART regression trees;

[0016] It uses the MSE formula:

[0017]

[0018] Where N: sample size;

[0019] i: each data sample;

[0020] fi: the value regressed by the model;

[0021] yi: The actual numerical label of sample point i.

[0022] Preferably, the convolutional neural network algorithm includes an input layer, a convolution layer, an activation function, a pooling layer and a fully connected layer, and its calculation formula is:

[0023] N=(W-F+2P) / S+1

[0024] Where N: output size;

[0025] W: input size;

[0026] F: convolution size;

[0027] P: the size of the padding value;

[0028] S: step size.

[0029] As a further preferred embodiment of the present invention, the server is used to store data and provide historical data for model training, the model training data includes historical fault data and normal operation data, and the model evaluation evaluates model performance through cross-validation.

[0030] As a further preference of the present invention, the method for using the same comprises the following steps:

[0031] S1. First, the operating data of the energy storage equipment is collected through the monitoring component, and the current, voltage and full power status of the battery cluster are monitored through the BMS. The smoke sensor, temperature sensor and optical sensor are used to monitor the working environment inside the cabinet. The monitoring component sends the monitoring data to the coordination controller, and the coordination controller integrates the data and sends it to the industrial computer. The industrial computer judges the operating status of the energy storage equipment through the risk assessment system;

[0032] S2. The risk assessment system first extracts data, extracts feature information related to equipment failure from the data collected by the monitoring components, selects the most important feature information related to the failure from the extracted feature information through principal component analysis, and then trains and evaluates the model. The extracted data is analyzed and judged through machine learning and deep learning, and compared with the historical failure data stored in the server. The performance of the model is evaluated through cross-validation, and the established model is used to predict possible future failures of energy storage and provide early warning information to help maintenance personnel take measures in advance.

[0033] S3. The industrial computer can determine whether the energy storage equipment has failed based on the real-time analysis of the risk assessment system. When a failure occurs, the industrial computer can send early warning information to remote maintenance personnel through the communication module and shut down the power supply of the energy storage equipment through the main circuit breaker. The coordination controller adjusts the operating parameters of the battery cluster through the PCS according to the information monitored by the BMS. The automatic fire extinguisher can extinguish the fire in time when a fire occurs inside the cabinet.

[0034] The present invention has the following beneficial effects:

[0035] 1. The present invention achieves the effect of risk assessment of energy storage equipment by setting up diagnostic components and monitoring components. BMS, smoke sensors, temperature sensors and optical sensors are used to collect operating data of energy storage equipment, the coordination controller is used to integrate the collected data, the server is used to store operating data, and the industrial computer uses the risk assessment system to perform real-time analysis and monitoring of the operating status of the energy storage equipment, so as to timely determine whether the energy storage equipment has failed, and provide early warning information to help maintenance personnel take measures in advance, effectively improving the safety of the operation of the energy storage equipment.

[0036] 2. The present invention achieves the effect of displaying the data collected by the monitoring component in real time by setting a display, so that the maintenance personnel can understand the operation status of the energy storage device. By setting a communication module, the energy storage device can be remotely transmitted, so that the operation and maintenance personnel can be reminded in time when a fault occurs. By setting an automatic fire extinguisher, the energy storage device can be extinguished when a fire occurs in the energy storage device, thereby improving the safety of the operation of the energy storage device. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the main structure of the present invention;

[0038] Figure 2 It is a rear view structural schematic diagram of the present invention;

[0039] Figure 3 It is a schematic diagram of the system flow of the present invention;

[0040] Figure 4 Schematic diagram of the risk assessment system flow of the present invention.

[0041] In the figure: 1. Energy storage component; 101. Cabinet; 102. Sealed door; 103. Main circuit breaker; 104. Battery cluster; 105. PCS; 2. Diagnostic component; 201. Industrial computer; 202. Server; 203. Coordination controller; 3. Monitoring component; 301. BMS; 302. Smoke sensor; 303. Temperature sensor; 304. Optical sensor; 4. Display; 5. Communication module; 6. Automatic fire extinguisher. DETAILED DESCRIPTION

[0042] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific preferred embodiments.

[0043] In the description of the present invention, it should be understood that the terms "left side", "right side", "upper part", "lower part" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. "First", "second" and the like do not indicate the importance of the components, and therefore cannot be understood as limiting the present invention. The specific dimensions used in this embodiment are only for illustrating the technical solution, and do not limit the scope of protection of the present invention.

[0044] The energy storage component 1, cabinet 101, sealed door 102, main circuit breaker 103, battery cluster 104, PCS 105, diagnostic component 2, industrial computer 201, server 202, coordination controller 203, monitoring component 3, BMS 301, smoke sensor 302, temperature sensor 303, optical sensor 304, display 4, communication module 5 and automatic fire extinguisher 6 of the present invention are all universal standard parts or parts known to those skilled in the art, and their structures and principles can be known to those skilled in the art through technical manuals or through conventional experimental methods.

[0045] Example 1

[0046] like Figure 1-4As shown, it is the first embodiment of the present invention, which provides a risk assessment device for energy storage equipment based on artificial intelligence technology and a method of use, including an energy storage component 1 and a risk assessment system. The energy storage component 1 includes a cabinet 101, a sealed door 102, a main circuit breaker 103, a battery cluster 104 and a PCS 105. The inner cavity of the cabinet 101 is installed with a diagnostic component 2 and a monitoring component 3. The diagnostic component 2 includes an industrial computer 201, a server 202 and a coordination controller 203. The monitoring component 3 includes a BMS 301, a smoke sensor 302, a temperature sensor 303 and an optical sensor 304. The diagnostic component 2 is used to analyze and evaluate the operating data of the energy storage device, the BMS 301 is used to collect the operating data of the battery cluster 104, and the smoke sensor 302, the temperature sensor 303 and the optical sensor 304 are used to collect the operating data inside the cabinet 101.

[0047] The risk assessment system includes data collection, feature extraction, feature selection, model learning, model training, model evaluation and fault diagnosis. Data collection collects the operating data of the energy storage device and the battery cluster 104 through the monitoring component 3. Feature extraction and feature selection are used to extract feature information related to faults in the operating data. Model learning, model training and model evaluation are used to learn patterns and trends in the data and predict the time point when the equipment fails. Fault diagnosis uses models to predict possible failures of the equipment.

[0048] like Figure 1-4 As shown, a risk assessment model is established through model learning, model training and model evaluation. The industrial computer 201 performs real-time analysis and monitoring of the energy storage equipment according to the risk assessment model. The BMS 301, smoke sensor 302, temperature sensor 303 and optical sensor 304 are used to collect the operating parameters of the energy storage equipment and integrate them through the coordination controller 203. The coordination controller 203 sends the data to the industrial computer 201. The industrial computer 201 uses the established risk assessment model to perform real-time analysis on the data, so as to predict possible future failures of the energy storage equipment and provide early warning information to help maintenance personnel take measures in advance.

[0049] Example 2

[0050] Reference Figure 1-4 , which is the second embodiment of the present invention, and this embodiment is based on the previous embodiment.

[0051] In this embodiment, the diagnostic component 2 and the monitoring component 3 are both installed in the inner cavity of the cabinet 101, the sealed door 102 is hinged to the cabinet 101 by a hinge, a display 4 is installed on the surface of the sealed door 102, a communication module 5 is also installed in the inner cavity of the cabinet 101, and an automatic fire extinguisher 6 is installed at the rear end of the inner cavity of the cabinet 101.

[0052] The output ends of BMS301, smoke sensor 302, temperature sensor 303 and optical sensor 304 are all connected to the input end of coordination controller 203, the output end of coordination controller 203 is respectively connected to the input ends of industrial computer 201 and PCS105, the output end of industrial computer 201 is respectively connected to the input ends of display 4, communication module 5, automatic fire extinguisher 6 and main circuit breaker 103, and the output end and input end of industrial computer 201 are interconnected with the output end and input end of server 202.

[0053] like Figure 1-4 As shown, the cabinet 101 and the sealed door 102 cooperate to provide installation and protection space for the battery cluster 104, PCS 105, diagnostic component 2, monitoring component 3, display 4, communication module 5 and automatic fire extinguisher 6. The communication module 5 adopts the LoRa module to realize long-distance wireless transmission. The PCS 105 is used to control the charging and discharging process of the battery cluster 104 and perform AC-DC conversion. The server 202 is used to store historical fault data and data analyzed during normal operation.

[0054] Example 3

[0055] Reference Figure 1-4 , which is the third embodiment of the present invention, and this embodiment is based on the previous two embodiments.

[0056] In this embodiment, model learning adopts two methods: machine learning and deep learning. Machine learning adopts a random forest regression algorithm, and deep learning adopts a convolutional neural network algorithm.

[0057] The steps of random forest regression algorithm are as follows:

[0058] S1, randomly extract m sample points from the training sample set S to obtain a new sub-training set S1...Sn;

[0059] S2. Use the sub-training set to train a CART regression tree, and the splitting rule for each node is to first randomly select k features from all features, and then select the optimal split point from these k features to divide the left and right subtrees;

[0060] S3. Through step b, many CART regression tree models can be generated;

[0061] S4. The final prediction result of each CART regression tree is the mean of the leaf nodes reached by the sample points;

[0062] S5. The final prediction result of random forest is the mean of the prediction results of all CART regression trees;

[0063] It uses the MSE formula:

[0064]

[0065] Where N: sample size;

[0066] i: each data sample;

[0067] fi: the value regressed by the model;

[0068] yi: The actual numerical label of sample point i.

[0069] The convolutional neural network algorithm consists of an input layer, a convolution layer, an activation function, a pooling layer, and a fully connected layer. Its calculation formula is:

[0070] N=(W-F+2P) / S+1

[0071] Where N: output size;

[0072] W: input size;

[0073] F: convolution size;

[0074] P: the size of the padding value;

[0075] S: step size.

[0076] Server 202 is used to store data and provide historical data for model training. The model training data includes historical fault data and normal operation data. Model evaluation evaluates model performance through cross-validation.

[0077] like Figure 1-4 As shown, the risk assessment system first extracts data, extracts feature information related to equipment failure from the data collected by the monitoring component 3, selects the most important feature information related to the failure from the extracted feature information through principal component analysis, and then trains and evaluates the model, analyzes and judges the extracted data through machine learning and deep learning, compares it with the historical failure data stored in the server 202, evaluates the performance of the model through cross-validation, and uses the established model to predict possible future failures of the energy storage equipment, and provides early warning information to help maintenance personnel take measures in advance.

[0078] A method for using a risk assessment device for energy storage equipment based on artificial intelligence technology, firstly, the operating data of the energy storage equipment is collected through the monitoring component 3, the current, voltage and full power state of the battery cluster 104 are monitored through the BMS 301, the smoke sensor 302, the temperature sensor 303 and the optical sensor 304 are used to monitor the working environment inside the cabinet 101, the monitoring component 3 sends the monitoring data to the coordination controller 203, the coordination controller 203 integrates the data and sends it to the industrial computer 201, and the industrial computer 201 judges the operating state of the energy storage equipment through the risk assessment system;

[0079] The risk assessment system first extracts data, extracts feature information related to equipment failure from the data collected by the monitoring component 3, selects the most important feature information related to the failure from the extracted feature information through the principal component analysis method, and then trains and evaluates the model, analyzes and judges the extracted data through machine learning and deep learning, compares the historical failure data stored in the server 202, evaluates the performance of the model through cross-validation, and uses the established model to predict possible future failures of the energy storage equipment, and provides early warning information to help maintenance personnel take measures in advance;

[0080] The industrial computer 201 can determine whether the energy storage device fails based on the real-time analysis of the risk system. When a failure occurs, the industrial computer 201 can send an early warning message to the remote maintenance personnel through the communication module 5 and shut down the power supply of the energy storage device through the main circuit breaker 103. The coordination controller 203 adjusts the operating parameters of the battery cluster 104 through the PCS 105 according to the information monitored by the BMS 301. The automatic fire extinguisher 6 can extinguish the fire in time when a fire occurs inside the cabinet 101.

[0081] The standard parts used in this application document can all be purchased from the market and can be customized according to the description in the specification and drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. The machinery, parts and equipment all adopt conventional models in the prior art. The control method is automatic control through a controller. The control circuit of the controller can be implemented through simple programming by technicians in this field, which is common knowledge in the field. This application is mainly used to protect mechanical devices, so this application no longer explains the control method and circuit connection in detail.

[0082] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A risk assessment device for energy storage equipment based on artificial intelligence technology, comprising an energy storage component (1) and a risk assessment system, characterized in that: The energy storage component (1) comprises a cabinet (101), a sealed door (102), a main circuit breaker (103), a battery cluster (104) and a PCS (105); a diagnostic component (2) and a monitoring component (3) are installed in the inner cavity of the cabinet (101); the diagnostic component (2) comprises an industrial computer (201), a server (202) and a coordination controller (203); the monitoring component (3) comprises a BMS (301), a smoke sensor (302), a temperature sensor (303) and an optical sensor (304); the diagnostic component (2) is used to analyze and evaluate the operating data of the energy storage device; the BMS (301) is used to collect the operating data of the battery cluster (104); the smoke sensor (302), the temperature sensor (303) and the optical sensor (304) are used to collect the operating data inside the cabinet (101); The risk assessment system includes data collection, feature extraction, feature selection, model learning, model training, model evaluation and fault diagnosis. The data collection collects the operating data of the energy storage device and the battery cluster (104) through the monitoring component (3). The feature extraction and feature selection are used to extract feature information related to faults in the operating data. The model learning, model training and model evaluation are used to learn the patterns and trends in the data and predict the time point when the device fails. The fault diagnosis uses the model to predict possible faults of the device.

2. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 1, characterized in that: The diagnostic component (2) and the monitoring component (3) are both installed in the inner cavity of the cabinet (101); the sealed door (102) is hinged to the cabinet (101) via a hinge; a display (4) is installed on the surface of the sealed door (102); a communication module (5) is also installed in the inner cavity of the cabinet (101); and an automatic fire extinguisher (6) is installed at the rear end of the inner cavity of the cabinet (101).

3. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 2 is characterized in that: The output ends of the BMS (301), smoke sensor (302), temperature sensor (303) and optical sensor (304) are all connected to the input end of the coordination controller (203); the output end of the coordination controller (203) is respectively connected to the input ends of the industrial computer (201) and the PCS (105); the output end of the industrial computer (201) is respectively connected to the input ends of the display (4), the communication module (5), the automatic fire extinguisher (6) and the main circuit breaker (103); the output end and input end of the industrial computer (201) are both interconnected with the output end and input end of the server (202).

4. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 1, characterized in that: The model learning of the risk assessment system adopts both machine learning and deep learning. The machine learning adopts a random forest regression algorithm, and the deep learning adopts a convolutional neural network algorithm.

5. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 4 is characterized in that: The steps of the random forest regression algorithm are as follows: S1, randomly extract m sample points from the training sample set S to obtain a new sub-training set S1...Sn; S2. Use the sub-training set to train a CART regression tree, and the splitting rule for each node is to first randomly select k features from all features, and then select the optimal split point from these k features to divide the left and right subtrees; S3, through S2, many CART regression tree models can be generated; S4. The final prediction result of each CART regression tree is the mean of the leaf nodes reached by the sample points; S5. The final prediction result of random forest is the mean of the prediction results of all CART regression trees; It uses the MSE formula: Where N: sample size; i: each data sample; fi: the value regressed by the model; yi: The actual numerical label of sample point i.

6. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 4 is characterized in that: The convolutional neural network algorithm includes an input layer, a convolution layer, an activation function, a pooling layer and a fully connected layer, and its calculation formula is: N=(W-F+2P) / S+1 Where N: output size; W: input size; F: convolution size; P: the size of the padding value; S: step size.

7. The risk assessment device for energy storage equipment based on artificial intelligence technology according to claim 1 is characterized in that: The server (202) is used to store data and provide historical data for model training, wherein the model training data includes historical fault data and normal operation data, and the model evaluation evaluates model performance through cross-validation.

8. The method for using the risk assessment device for energy storage equipment according to any one of claims 1 to 7, characterized in that: The steps include: S1. First, the operation data of the energy storage device is collected through the monitoring component (3), and the current, voltage and full power state of the battery cluster (104) are monitored through the BMS (301). The smoke sensor (302), the temperature sensor (303) and the optical sensor (304) are used to monitor the working environment inside the cabinet (101). The monitoring component (3) sends the monitoring data to the coordination controller (203), and the coordination controller (203) integrates the data and sends it to the industrial computer (201). The industrial computer (201) determines the operation state of the energy storage device through the risk assessment system; S2, the risk assessment system first extracts data, extracts feature information related to equipment failure from the data collected by the monitoring component (3), selects the most important feature information related to the failure from the extracted feature information through principal component analysis, and then trains and evaluates the model, analyzes and judges the extracted data through machine learning and deep learning, compares the extracted data with historical failure data stored in the server (202), evaluates the performance of the model through cross-validation, and uses the established model to predict possible future failures of the energy storage device, and provides early warning information to help maintenance personnel take measures in advance; S3. The industrial computer (201) can determine whether a fault occurs in the energy storage device based on the real-time analysis of the risk assessment system. When a fault occurs, the industrial computer (201) can send an early warning message to a remote maintenance personnel through the communication module (5) and shut down the power supply of the energy storage device through the main circuit breaker (103). The coordination controller (203) adjusts the operating parameters of the battery cluster (104) through the PCS (105) based on the information monitored by the BMS (301). The automatic fire extinguisher (6) can promptly extinguish a fire when a fire occurs inside the cabinet (101).

Images