Security control system and method of double-defense energy storage cabinet based on intelligent numerical control
Through the integrated intelligent control and thermal management module of intelligent control technology, the temperature and gas of the energy storage cabinet are monitored in real time, and the regulation strategy is optimized by cloud platform and reinforcement learning algorithm, the problem of low automation of the energy storage cabinet security system is solved and energy storage is improved Safety and stability of the cabinet.
Patent Information
- Application Number
- CN202410903637.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-08
AI Technical Summary
The existing energy storage cabinet security control system is not very automated, it is difficult to adapt to complex working conditions, and it is impossible to independently optimize charging and discharging and thermal management strategies, which makes it difficult to control the charging and discharging balance under thermal management, and poses safety hazards.
Adopt the dual energy storage cabinet security control system based on Zhilian CNC, integrates intelligent control, thermal management and fault processing modules, and monitors temperature and abnormal gases in real time through thermal imaging and gas monitoring technology, and uses cloud platform to transmit temperature control and power-off instructions. Combining reinforcement learning and PID algorithms to train servo control models, implement adaptive cooling and charge and discharge adjustment.
It significantly improves the safety and stability of the energy storage cabinet, enhances the adaptability to complex working conditions, realizes intelligent and automated management, effectively prevents potential safety risks, and ensures the stable operation of the energy storage cabinet.
Smart Images

Figure CN118449282B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of energy storage cabinet monitoring and protection, and in particular to a double-protection energy storage cabinet security control system and method based on intelligent numerical control. Background Art
[0002] Under the current background of energy transformation, energy storage technology is becoming increasingly important as a key link in balancing supply and demand and promoting the effective use of renewable energy. As the core equipment for storing electrical energy, energy storage cabinets are widely used in power systems, electric vehicle charging stations, distributed energy systems and other fields. However, with the expansion of the scale of energy storage systems and the complexity of application scenarios, the safety of energy storage cabinets has become the focus of industry attention, especially the fire and explosion safety challenges are particularly severe. The security and control systems of traditional energy storage cabinets mostly rely on manual monitoring and basic automatic control, and there are problems such as slow response, incomplete monitoring, and low efficiency. Especially in distributed energy systems, there are many unstable factors. For example, natural gas, a flammable and explosive gas, will explode when exposed to open flames or the temperature exceeds the critical value, which also brings more unstable factors to the safe use of energy storage cabinets.
[0003] For example, a Chinese patent with authorization announcement number CN117394548B discloses a high-protection outdoor integrated liquid-cooled energy storage cabinet performance monitoring and control system, including: a database for storing performance verification parameters of the outdoor integrated liquid-cooled energy storage cabinet, a performance monitoring level division module for dividing the performance monitoring level of the outdoor integrated liquid-cooled energy storage cabinet, thereby counting the PCS level, BMS level and gateway communication level, a performance monitoring comprehensive analysis module for performance monitoring of the outdoor integrated liquid-cooled energy storage cabinet, wherein the performance monitoring comprehensive analysis module includes a PCS level monitoring unit, a BMS level monitoring unit and a gateway communication level monitoring unit, and a management execution integrated cloud platform for dynamically regulating and managing the PCS level, BMS level and gateway communication level of the outdoor integrated liquid-cooled energy storage cabinet.
[0004] The above existing technologies have the following problems: the existing methods have a low degree of automation, poor adaptability to complex working conditions, cannot independently optimize the charging and discharging strategies and thermal management strategies, and it is difficult to control the charging and discharging balance under thermal management; in order to solve the above problems, the present invention provides a double-protection energy storage cabinet security control system and method based on Zhilian CNC. Summary of the invention
[0005] In view of the shortcomings of the prior art, the present invention proposes a dual-defense energy storage cabinet security control system and method based on intelligent numerical control. The system realizes real-time monitoring and intelligent control of the charging and discharging process, internal temperature and abnormal gas of the energy storage cabinet by integrating functional modules such as intelligent control, thermal management, fault handling and cloud platform. The system monitors the performance of the battery cell in real time through the battery cell monitoring unit, and predicts its performance change, and adjusts the charging and discharging process accordingly. The thermal management module combines thermal imaging and gas monitoring technology, and transmits temperature control and power-off instructions through the cloud platform to ensure the safe operation of the energy storage cabinet. The fault handling module discovers and handles faults in time, and realizes remote control and repair through the cloud platform. The cloud platform is responsible for strategy training updates and instruction transmission protection, which comprehensively improves the safety and stability of the energy storage cabinet.
[0006] To achieve the above object, the present invention provides the following technical solutions:
[0007] The double-protection energy storage cabinet security control system based on Zhilian CNC includes: intelligent control module and thermal management module;
[0008] The thermal management module includes a thermal imaging detection unit, an abnormal gas concentration monitoring unit, and an adaptive cooling control unit;
[0009] The intelligent control module includes a battery cell monitoring unit, a battery cell performance prediction unit, a charge and discharge adjustment unit, and a dual-protection control unit;
[0010] The thermal imaging detection unit monitors the temperature distribution data of the energy storage cabinet during the charging and discharging process in real time through the thermal imager deployed in the energy storage cabinet, and feeds back the real-time monitored temperature distribution data to the dual-defense control unit; the abnormal gas concentration monitoring unit monitors the concentration of abnormal gases in the energy storage cabinet during the charging and discharging process in real time through the gas monitoring equipment deployed in the energy storage cabinet, and feeds back the real-time monitored abnormal gas concentration data to the dual-defense control unit; the adaptive cooling control unit is used to control the cooling subsystem in the energy storage cabinet to perform adaptive dual-defense cooling on the inside of the energy storage cabinet according to the temperature control instructions generated by the dual-defense control unit and the configured adaptive cooling strategy; the thermal imager is configured with a thermal image analysis algorithm;
[0011] The battery cell monitoring unit is used to monitor the charging and discharging parameters of the battery cells of the energy storage cabinet in real time using the configured integrated sensors; the battery cell performance prediction unit is used to predict the battery cell performance in real time based on the detected charging and discharging parameters and the configured prediction algorithm to obtain the predicted battery cell performance data; the charge and discharge adjustment unit is used to adjust the charging and discharging process of the energy storage cabinet in real time based on the charge and discharge parameters and the predicted battery cell performance data and the configured charge and discharge adjustment strategy; the dual-protection control unit is used to construct the dual-protection control threshold range, and generate temperature control instructions and power-off instructions based on the temperature distribution data and abnormal gas concentration data using the integrated control algorithm;
[0012] Thermal image analysis algorithms include:
[0013] A1. Collect historical thermal image data of the energy storage cabinet, mark the corresponding temperature for each region of the collected historical thermal image data, and use anchor frames to mark the sparks in the historical thermal images where sparks appear, and mark the corresponding temperature when the sparks appear, and obtain the marked historical thermal image data;
[0014] A2. Use the YOLOv5 network pre-trained with historical thermal image data to build a thermal image analysis model, and use the annotated historical thermal image data to fine-tune the built thermal image analysis model to obtain a fine-tuned thermal image analysis model;
[0015] A3. Deploy the fine-tuned thermal image analysis model to the thermal imager in the thermal imaging detection unit for real-time monitoring to obtain the temperature distribution data in the energy storage cabinet and the extreme state data of the electric spark.
[0016] The specific working steps of the dual protection control unit include:
[0017] B1. Divide the working temperature of the energy storage cabinet into 4 working temperature ranges, and set the upper limit threshold of the four working temperature ranges to ,and , corresponding to the low temperature upper limit threshold, the medium temperature upper limit threshold, and the high temperature upper limit threshold respectively;
[0018] B2. Set the abnormal gas explosion concentration threshold to , the corresponding explosion temperature threshold is ,and ;
[0019] B3. Temperature distribution data at time t obtained in real time based on the set threshold and the thermal imaging detection unit and the abnormal gas concentration monitoring unit Abnormal gas concentration data , build a dual defense control threshold range, specifically:
[0020] when When the first execution information is triggered, the energy storage cabinet maintains the current working state, generates an initial temperature control instruction, and does not generate a power-off instruction;
[0021] when When the second execution information is triggered, the energy storage cabinet maintains the current working state, generates the first temperature control instruction, and does not generate the power-off instruction;
[0022] when When the third execution information is triggered, the energy storage cabinet maintains the current working state, generates the second temperature control instruction, and does not generate the power-off instruction;
[0023] when , the fourth execution information is triggered, at which time a third temperature control instruction is generated, and an instruction to reduce power output is generated;
[0024] when , the fifth execution information is triggered, at which time a fourth temperature control instruction is generated, and a power-off instruction is generated;
[0025] When electric sparks are detected by the thermal imaging detection unit, When the sixth execution information is triggered, the fourth temperature control instruction is generated at this time, and a power-off instruction is generated; wherein R represents an arbitrary real number; the temperature control instruction includes an initial temperature control instruction, a first temperature control instruction, a second temperature control instruction, a third temperature control instruction and a fourth temperature control instruction;
[0026] The specific working steps of the dual protection control unit also include:
[0027] B4. Use the reinforcement learning algorithm and the PID algorithm to build a servo control model, and use the thermal imaging detection unit and the abnormal gas concentration monitoring unit to collect historical temperature data, historical abnormal gas concentration data, and dual-defense control threshold intervals to train the servo control model and obtain a trained servo control model.
[0028] B5. Deploy the trained servo control model to the dual-defense control unit, and input the temperature data and abnormal gas concentration data collected in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit into the servo control model to generate corresponding temperature control instructions.
[0029] Specifically, the specific working steps of the adaptive cooling strategy configured in the adaptive cooling control unit include:
[0030] C1. The energy storage cabinet cooling subsystem is equipped with a double-pipe metal cooling device, which includes a first-pipe metal cooling device and a second-pipe metal cooling device. The first-pipe metal cooling device is cooled by water, and the second-pipe metal cooling device is cooled by an immersion coolant with a higher specific heat capacity than water. The thermal conductivity of the first-pipe metal cooling device is , the specific heat capacity of water is ;
[0031] C2. According to the temperature data and abnormal gas concentration data monitored in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit, perform the following steps:
[0032] C21, when the first execution information is triggered and the servo control model generates an initial temperature control instruction, the dual-pipe metal cooling device is not started to perform cooling operation;
[0033] C22, when the second execution information is triggered and the servo control model generates the first temperature control instruction, the double-pipe metal cooling device is started and the first water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation;
[0034] C23, when the third execution information is triggered and the servo control model generates a second temperature control instruction, the dual-pipe metal cooling device is started and a second water cooling speed is generated. , control the first pipeline metal cooling equipment to perform cooling operation;
[0035] C24, when the fourth execution information is triggered and the servo control model generates the third temperature control instruction, the dual-pipe metal cooling device is started and the third water cooling speed is generated. , control the first pipeline metal cooling equipment to perform cooling operation.
[0036] Specifically, the specific working steps of the adaptive cooling strategy configured in the adaptive cooling control unit also include:
[0037] C25. When the fifth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation;
[0038] C26. When the sixth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , control the first pipeline metal cooling device to perform cooling operation, and at the same time start the second pipeline metal cooling device to perform immersion cooling treatment on the battery cell.
[0039] Specifically, the specific steps of the prediction algorithm in the battery cell performance prediction unit include:
[0040] D1. Use the acquired historical charge and discharge parameter data to construct the performance prediction input sequence for each battery in the cell ,in, represents the discharge depth of the jth battery in the cell at time t, The output power of the jth battery in the cell at time t, The output current of the jth battery in the cell at time t, Indicates the lower limit of the state of charge corresponding to the jth battery in the cell under discharge state;
[0041] D2. Use the particle swarm optimization algorithm optimized by BP neural network to build a prediction model, and input the constructed battery performance prediction input sequence into the prediction model for training, and obtain a trained prediction model;
[0042] D3. Deploy the obtained prediction model to the battery performance prediction unit, and predict the discharge performance of each battery in the battery according to the real-time collected discharge parameter data. , It represents the predicted lower limit of the state of charge corresponding to the jth battery in the cell under the discharge state;
[0043] D4. Input the battery performance prediction into the sequence , Replace with , , and repeat the D1-D3 process to predict the charging process of each battery ,in Indicates the upper limit of the state of charge corresponding to the jth battery in the cell under charging condition, Indicates the depth of charge of the jth battery in the cell at time t under charging status, It represents the predicted upper limit of the state of charge of the jth battery in the battery cell under charging condition.
[0044] Specifically, the specific working steps of the charge and discharge regulation strategy configured in the charge and discharge regulation unit include:
[0045] E1. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding discharge balance equation is: ,in It represents the total state of charge of all batteries in the cell at time t. represents the rated state of charge of the jth battery in the battery cell; m represents the total number of batteries in the battery cell; Indicates the discharge protection factor of the jth battery in the cell; j≤m;
[0046] E2. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding charging balance equation is: , Indicates the charging protection factor of the jth battery in the cell, Indicates the total state of charge of all cells at time t under charging condition;
[0047] E3. When the energy storage cabinet is in the fourth execution information state, an instruction to reduce the discharge power output is generated. The corresponding calculation formula for the reduced discharge power value is: ,in Indicates that the battery cell is The heat generated during the cycle, Indicates that the battery is The heat generated during the cycle, Indicates the output current that the cell needs to reduce, U is the cell output port voltage, Indicates the specific value by which the discharge power of the battery cell needs to be reduced; is the total resistance of the energy storage cabinet cells;
[0048] E4. When the energy storage cabinet is in the fifth execution information state and the sixth execution information state, the bidirectional converter at the interface of the energy storage cabinet is controlled by the generated power-off instruction so that the output current and the input current become 0.
[0049] The security control method of the double-defense energy storage cabinet based on Zhilian CNC includes the following specific steps:
[0050] S1. Use integrated sensors to monitor the charging and discharging parameters of the energy storage cabinet cells in real time, and use the prediction algorithm to predict the cell performance data based on the collected charging and discharging parameters;
[0051] S2. According to the acquired charging and discharging parameters and predicted battery cell performance data, the configured charging and discharging adjustment strategy is used to adjust the charging and discharging process of the energy storage cabinet in real time. At the same time, a thermal imager and a gas monitoring and analysis device are used to monitor the temperature distribution and abnormal gas concentration in real time during the charging and discharging process, and obtain temperature distribution data and abnormal gas concentration data;
[0052] S3. Generate temperature control instructions and power-off instructions based on the acquired temperature distribution data and abnormal gas concentration data using the configured dual-defense control threshold interval and control algorithm;
[0053] S4. When a temperature control instruction is generated, the cooling subsystem in the energy storage cabinet is controlled through the configured adaptive cooling strategy to perform adaptive dual-protection cooling on the inside of the energy storage cabinet. If a power-off instruction is generated, the configured charge and discharge adjustment strategy is used to perform power-off protection on the energy storage cabinet. When the temperature returns to the dual-protection control threshold range, the charge and discharge adjustment strategy is used to control the energy storage cabinet to restore the charge and discharge state.
[0054] A computer-readable storage medium stores computer instructions, which, when executed, execute a security control method for a dual-defense energy storage cabinet based on Zhilian CNC.
[0055] Compared with the prior art, the present invention has the following beneficial effects:
[0056] 1. Aiming at the deficiencies of the prior art, the present invention sets multi-level working temperature range thresholds and abnormal gas explosion point concentration thresholds through the configured dual-protection control thresholds, and constructs accurate dual-protection control threshold ranges in combination with thermal imaging and abnormal gas concentration monitoring data. The servo control model trained by the reinforcement learning algorithm and the PID algorithm can respond to and optimize the temperature control and power-off strategies in real time. This dynamic adjustment strategy not only improves the energy efficiency and safety of the energy storage cabinet, but also significantly improves the adaptability to complex working conditions. By deploying the trained servo control model, the intelligent and automated management of the energy storage cabinet is realized, which effectively prevents potential safety risks and ensures the stable operation of the energy storage cabinet.
[0057] 2. Aiming at the deficiencies of the prior art, the present invention constructs an adaptive cooling strategy based on the multi-level working temperature range set by the dual-protection control threshold, which can respond to temperature changes and abnormal gas concentrations in real time, intelligently adjust the cooling speed and cooling method, and ensure that the energy storage cabinet maintains the best working state under complex working conditions; especially in extreme cases, such as when electric sparks appear, by starting the second pipeline metal cooling device for immersion cooling treatment, it can not only effectively reduce the temperature but also quickly extinguish the electric sparks, thereby ensuring the safe and stable operation of the energy storage cabinet;
[0058] 3. In view of the deficiencies in the prior art, the present invention configures a dual-protection control threshold and an adaptive cooling strategy, which can achieve a charge-discharge balance in the energy storage cabinet under normal working conditions. When the temperature is high or an abnormal situation occurs, the charge-discharge power can be intelligently reduced or the power can be cut off, effectively preventing overheating of the battery cell and potential safety risks. The adaptive charge-discharge adjustment strategy not only improves the utilization efficiency of the energy storage cabinet, but also significantly enhances its safety and reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 This is a module structure diagram of the double-protection energy storage cabinet security control system based on Zhilian CNC in Example 1 of the present invention;
[0060] Figure 2 This is a working principle structure diagram of the double-protection energy storage cabinet security control system based on Zhilian CNC in Example 2 of the present invention;
[0061] Figure 3 This is a flow chart of the security control method of a double-protection energy storage cabinet based on Zhilian CNC in Example 3 of the present invention. DETAILED DESCRIPTION
[0062] Example 1
[0063] See also Figure 1 , an embodiment provided by the present invention: a double-protection energy storage cabinet security control system based on Zhilian CNC, comprising: a thermal management module, an intelligent control module, a fault processing module and a cloud platform;
[0064] Thermal management module, used to monitor the internal temperature and abnormal gas of the energy storage cabinet in real time, and adjust and control the operating status of the cooling subsystem; the thermal management module includes a thermal imaging detection unit, an abnormal gas concentration monitoring unit, an adaptive cooling control unit, and a waste heat recovery unit;
[0065] The thermal imaging detection unit uses a thermal imager deployed in the energy storage cabinet to monitor the temperature distribution data inside the energy storage cabinet during the charging and discharging process in real time, and feeds back the real-time monitored temperature distribution data to the dual-protection control unit; the thermal imager is equipped with a thermal image analysis algorithm;
[0066] The abnormal gas concentration monitoring unit monitors the abnormal gas concentration in the energy storage cabinet during the charging and discharging process in real time through the gas monitoring equipment deployed in the energy storage cabinet, and feeds back the abnormal gas concentration data monitored in real time to the dual-defense control unit; the gas monitoring and analysis equipment deployed in this embodiment includes: catalytic combustion sensor, infrared (IR) gas sensor, photoionization detector and semiconductor metal oxide gas sensor;
[0067] The adaptive cooling control unit is used to control the cooling subsystem in the energy storage cabinet to perform adaptive dual-defense cooling inside the energy storage cabinet according to the temperature control instruction generated by the dual-defense control unit and using the configured adaptive cooling strategy;
[0068] The waste heat recovery unit is used to recover the heat in the cooling process through the configured cooling subsystem to improve the overall energy efficiency of the system.
[0069] Intelligent control module, used for real-time monitoring of energy storage cabinet operating parameters, dynamic control and remote adjustment of temperature-controlled charging and discharging strategies; the intelligent control module includes a battery cell monitoring unit, a battery cell performance prediction unit, a charging and discharging adjustment unit, a dual-protection control unit and a remote control unit;
[0070] The battery cell monitoring unit is used to monitor the charging and discharging parameters of the battery cells of the energy storage cabinet in real time using the configured integrated sensors; the charging and discharging parameters include current, state of charge SOC and depth of discharge DOH;
[0071] A cell performance prediction unit is used to predict the cell performance in real time based on the detected charge and discharge parameters and to obtain predicted cell performance data by using a configured prediction algorithm;
[0072] The charge and discharge regulation unit is used to adjust the charge and discharge process of the energy storage cabinet in real time according to the charge and discharge parameters and the predicted battery performance data using the configured charge and discharge regulation strategy;
[0073] The dual-defense control unit is used to construct a dual-defense control threshold range, and generate temperature control instructions and power-off instructions using an integrated control algorithm based on temperature distribution data and abnormal gas concentration data.
[0074] The remote control unit is used to remotely monitor and adjust the energy storage cabinet in real time using the configured Internet of Things algorithm; this embodiment remotely controls the energy storage cabinet by configuring 5G signals and WIFI.
[0075] The fault processing module is used to perform real-time abnormal monitoring of the operation process of the energy storage cabinet, and use the monitoring results to perform power-off protection and early warning feedback on the energy storage cabinet; the fault processing module includes a fault monitoring unit, a power-off protection unit and an early warning unit;
[0076] A fault monitoring unit is used to detect abnormalities occurring during the operation of the energy storage cabinet in real time using a configured fault detection algorithm, and to feed back the fault detection results to the integrated cloud platform; the fault detection algorithms used in this embodiment include a fault tree analysis algorithm and a model matching algorithm;
[0077] The power-off protection unit controls the energy storage cabinet to perform adaptive power-off protection according to the fault detection result uploaded to the cloud platform or the power-off instruction generated by the dual-protection control unit; further, in this embodiment, the power-off protection is implemented by a bidirectional converter configured at the interface of the energy storage cabinet;
[0078] The early warning unit is used to provide real-time abnormal early warning according to the configured early warning signal and the detection result of the fault monitoring unit; the early warning signal includes SMS reminder, setting a red alarm light or setting an alarm sound.
[0079] The cloud platform is used for the construction, training, and updating of strategies within the system units, as well as the interactive transmission and encryption protection of instructions generated within the corresponding units.
[0080] Furthermore, the thermal image analysis algorithm specifically includes:
[0081] A1. Collect historical thermal image data of the energy storage cabinet, mark the corresponding temperature of the collected historical thermal image data by region, and use anchor frames to mark the sparks in the historical thermal images where sparks appear, and mark the corresponding temperature when the sparks appear, and obtain the marked historical thermal image data; In this embodiment, the corresponding temperature is marked in different regions, that is, different temperature regions in the same image are displayed in different colors, and the corresponding temperature values are marked for the regions displayed in different colors. For example, the low temperature area is marked in blue, and the temperature range is from 0°C to about 40°C; the medium temperature area is indicated by green, and the temperature is between 40°C and 80°C; the high temperature area is indicated by red, and the temperature is between 80°C and 120°C; the extremely high temperature area is indicated by white, and the high temperature exceeds 120°C;
[0082] A2. Use the YOLOv5 network pre-trained with historical thermal image data to build a thermal image analysis model, and use the annotated historical thermal image data to fine-tune the built thermal image analysis model to obtain a fine-tuned thermal image analysis model;
[0083] A3. Deploy the fine-tuned thermal image analysis model to the thermal imager in the thermal imaging detection unit for real-time monitoring to obtain the temperature distribution data in the energy storage cabinet and the extreme state data of electric sparks.
[0084] This process collects and annotates the historical thermal image data of the energy storage cabinet, uses the YOLOv5 network to build and fine-tune the thermal image analysis model, and realizes real-time monitoring of the temperature distribution and spark status in the energy storage cabinet, significantly improving the accuracy and response speed of thermal management, effectively preventing potential safety risks, and ensuring the stable operation of the energy storage cabinet.
[0085] Furthermore, the specific working steps of the dual protection control unit in this embodiment include:
[0086] B1. Divide the working temperature of the energy storage cabinet into 4 working temperature ranges, and set the upper limit threshold of the four working temperature ranges to ,and , corresponding to the low temperature upper limit threshold, the medium temperature upper limit threshold, and the high temperature upper limit threshold respectively; for example, assuming that the low temperature working temperature range is [0, 40], then the corresponding , The corresponding values are The corresponding threshold is obtained as shown in the acquisition process; the upper limit threshold of the corresponding interval in this embodiment is specifically set by those skilled in the art according to the specific working environment;
[0087] B2. Set the abnormal gas explosion concentration threshold to , the corresponding explosion temperature threshold is ,and ;
[0088] In this embodiment, the abnormal gas explosion point concentration threshold and the explosion temperature threshold are set by those skilled in the art according to the specific working environment and the corresponding specific abnormal gas attribute parameters;
[0089] B3. Temperature distribution data at time t obtained in real time based on the set threshold and the thermal imaging detection unit and the abnormal gas concentration monitoring unit Abnormal gas concentration data , build a dual defense control threshold range, specifically:
[0090] when When the first execution information is triggered, the energy storage cabinet maintains the current working state, generates an initial temperature control instruction, and does not generate a power-off instruction;
[0091] when When the second execution information is triggered, the energy storage cabinet maintains the current working state, generates the first temperature control instruction, and does not generate the power-off instruction;
[0092] when When the third execution information is triggered, the energy storage cabinet maintains the current working state, generates the second temperature control instruction, and does not generate the power-off instruction;
[0093] when When the fourth execution information is triggered, a third temperature control instruction is generated, an instruction to reduce power output is generated, and an early warning signal is issued through the early warning unit;
[0094] when When the fifth execution information is triggered, a fourth temperature control instruction is generated, a power-off instruction is generated, and an early warning signal is issued through the early warning unit;
[0095] When electric sparks are detected by the thermal imaging detection unit, When the sixth execution information is triggered, the fourth temperature control instruction is generated at this time, and the power-off instruction is generated; wherein R represents an arbitrary real number; wherein R represents an arbitrary real number; that is, in this embodiment, when an electric spark is detected, regardless of the abnormal gas concentration data Why does the power-off instruction always generate to power off the energy storage cabinet?
[0096] B4. Use reinforcement learning algorithm and PID algorithm to build a servo control model, and use the thermal imaging detection unit and the abnormal gas concentration monitoring unit to collect historical temperature data, historical abnormal gas concentration data and dual-defense control threshold range, train the servo control model, and obtain the trained servo control model.
[0097] The reinforcement learning algorithms in this embodiment include: Q-Learning algorithm, Deep Q-Network algorithm, SAC algorithm and TD3 algorithm; further, the specific steps of constructing the servo control model in this embodiment include:
[0098] B41. Construct the input state of the reinforcement learning adjustment layer in the servo control model based on temperature distribution data, abnormal gas concentration data and the set double-defense control threshold interval , where C represents the set of dual defense control threshold intervals in B3; construct the joint execution action at the current time t according to the corresponding triggered execution information in B3 ,in, Indicates the action that triggers the execution of the first execution information. Indicates the action that triggers the execution of the corresponding second execution information. Indicates the action that triggers the execution of the corresponding third execution information. Indicates the action that triggers the execution of the corresponding fourth execution information. Indicates the action that triggers the execution of the corresponding second execution information. Indicates the action that triggers the corresponding execution of the sixth execution information. Indicates a joint execution action; in this embodiment, the first execution information satisfies the dual defense control threshold interval: The corresponding execution action information of generating an initial temperature control instruction and not generating a power-off instruction; the second execution information to the sixth execution information are the same as the first execution information, that is, the execution action information corresponding to the double-protection control threshold interval that satisfies the corresponding trigger;
[0099] B42, input the constructed input state and joint input action into the reinforcement learning adjustment layer to generate the adjustment amount of the corresponding execution action;
[0100] B43. Use the command control layer in the PID algorithm servo control model to initialize the corresponding parameters in the command control layer using the error between the real-time collected temperature and the corresponding temperature range threshold and the error between the real-time collected abnormal gas concentration data and the abnormal gas explosion point concentration threshold. , and input the generated adjustment amount corresponding to the execution action into the initialized instruction control layer to generate the corresponding control instruction; represents the initial scale factor, represents the initial integration time coefficient, Represents the initial differential time coefficient.
[0101] B5. Deploy the trained servo control model to the dual-defense control unit, and input the temperature distribution data and abnormal gas concentration data collected in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit into the servo control model to generate corresponding temperature control instructions.
[0102] This process sets multi-level working temperature range thresholds and abnormal gas explosion point concentration thresholds, combines thermal imaging and abnormal gas concentration monitoring data, and constructs an accurate dual-protection control threshold range. The servo control model trained by the reinforcement learning algorithm and the PID algorithm can respond to and optimize the temperature control and power-off strategies in real time. This dynamic adjustment strategy not only improves the energy efficiency and safety of the energy storage cabinet, but also significantly improves its adaptability to complex working conditions. By deploying the trained servo control model, the intelligent and automated management of the energy storage cabinet is realized, which effectively prevents potential safety risks and ensures the stable operation of the energy storage cabinet.
[0103] Furthermore, the specific working steps of the adaptive cooling strategy configured in the adaptive cooling control unit of this embodiment include:
[0104] C1. The energy storage cabinet cooling subsystem is equipped with a double-pipe metal cooling device, which includes a first-pipe metal cooling device and a second-pipe metal cooling device. The first-pipe metal cooling device is cooled by water, and the second-pipe metal cooling device is cooled by an immersion coolant with a higher specific heat capacity than water. The thermal conductivity of the first-pipe metal cooling device is , the specific heat capacity of water is ;
[0105] C2. According to the temperature data and abnormal gas concentration data monitored in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit, perform the following steps:
[0106] C21, when the first execution information is triggered and the servo control model generates an initial temperature control instruction, the dual-pipe metal cooling device is not started to perform cooling operation;
[0107] C22, when the second execution information is triggered and the servo control model generates the first temperature control instruction, the double-pipe metal cooling device is started and the first water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation;
[0108] Furthermore, in this embodiment, the specific steps of generating the first water delivery cooling speed to control the dual-pipe metal cooling device to perform cooling operation include:
[0109] C221, Assume that the cross-sectional area of the first pipeline metal cooling device is , set the low temperature upper threshold As the maximum temperature threshold for normal operation of the energy storage cabinet here, the current temperature of the water sent into the pipeline is and , based on the real-time monitoring of the temperature distribution data inside the energy storage cabinet and , using the formula Calculate the heat value that needs to be cooled and removed , where M is the mass of cooling water required, is the specific heat capacity of water, Temperature distribution data for real-time monitoring and The temperature difference between
[0110] C222, according to the caloric value obtained , calculate the first water supply cooling speed corresponding to the first temperature control instruction according to the thermodynamic formula, feed back the obtained first water supply cooling speed to the cooling subsystem, generate the corresponding water supply control instruction by using the integrated PID algorithm, and control the first pipeline metal cooling device to perform water supply cooling according to the first water supply cooling speed; wherein the thermodynamic calculation formula for the first water supply cooling speed is: , is the thermal conductivity of the first pipe metal, is the density of water, Real-time monitoring of temperature data and The temperature difference between Indicates the first water cooling speed corresponding to the first temperature control instruction, and is a constant value. Indicates that the temperature monitored in real time drops to the maximum temperature threshold for normal operation of the energy storage cabinet. The cycle time of this embodiment is set by professionals in this field according to the minimum degree of damage to the battery caused by temperature;
[0111] C23, when the third execution information is triggered and the servo control model generates a second temperature control instruction, the dual-pipe metal cooling device is started and a second water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation;
[0112] In this embodiment, the second water delivery cooling speed is generated The specific steps for controlling the dual-pipe metal cooling equipment to perform cooling operations are the same as those in steps C221 to C222;
[0113] C24, when the fourth execution information is triggered and the servo control model generates the third temperature control instruction, the dual-pipe metal cooling device is started and the third water cooling speed is generated. , control the first pipeline metal cooling device to perform cooling operation; in this embodiment, generate a third water cooling speed The specific steps of controlling the dual-pipe metal cooling equipment to perform cooling operations include:
[0114] C241. Since the output power is reduced during the temperature control process corresponding to the fourth execution information, and since the output voltage of the battery cell is constant, that is, the output current is reduced, assuming that the heat loss rate inside the energy storage cabinet before the output current is reduced is 1, then the heat loss rate at the current moment is , where Q represents the internal battery of the energy storage cabinet before the output current is reduced The heat generated during the cycle, Indicates the internal cells of the energy storage cabinet after reducing the output current The heat generated during the cycle, , where I represents the current output by the cell before the current is reduced. is the total resistance of the energy storage cabinet battery cell; the thermodynamic calculation formula for the first water cooling rate is:
[0115]
[0116] in, Indicates the heat value that needs to be cooled and removed corresponding to the third temperature control instruction. Its calculation method is the same as same, The temperature distribution data corresponding to the third temperature control instruction is monitored in real time and The temperature difference between the two, the heat loss rate in this embodiment represents the rate at which the heat in the energy storage cabinet decreases per unit time;
[0117] C242, according to the third water cooling speed obtained Feedback is given to the cooling subsystem, and the corresponding water delivery control instruction is generated by using the integrated PID algorithm to control the first pipeline metal cooling device to perform water delivery cooling at the first water delivery cooling speed;
[0118] C25. When the fifth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , control the first pipeline metal cooling device to perform cooling operation; the fourth water cooling speed of this embodiment The specific steps of the calculation are:
[0119] C251, at the time corresponding to the triggering of the fifth execution information, the battery cell issues a large power-off instruction, and the heat loss rate at the current moment is 2, then the corresponding fourth water supply cooling speed The calculation formula is , Indicates the heat value that needs to be cooled and removed corresponding to the fourth temperature control instruction. Its calculation method is the same as same, The temperature distribution data corresponding to the fourth temperature control instruction is monitored in real time and The temperature difference between
[0120] C252, according to the fourth water cooling speed obtained Feedback is given to the cooling subsystem, and a corresponding water delivery control instruction is generated by using an integrated PID algorithm to control the first pipeline metal cooling device to perform water delivery cooling at a fourth water delivery cooling speed;
[0121] C26. When the sixth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , control the first pipeline metal cooling device to perform cooling operation, and at the same time start the second pipeline metal cooling device to perform immersion cooling treatment on the battery cell.
[0122] In this embodiment, the cooling operation of the first pipeline metal cooling device is the same as that in C251 to C252. The second pipeline metal cooling device performs immersion cooling treatment on the battery cell, generates immersion control instructions through PID to cool the energy storage cabinet, and extinguishes the generated electric sparks. When the temperature of the energy storage cabinet drops to When the PID algorithm is used to generate control instructions, the second pipe metal cooling device in the cooling subsystem is controlled to extract the immersed coolant.
[0123] This process significantly improves the thermal management capability of the energy storage cabinet through the configuration of dual-pipe metal cooling equipment and refined cooling strategies. The adaptive cooling strategy can respond to temperature changes and abnormal gas concentrations in real time, and intelligently adjust the cooling speed and cooling method to ensure that the energy storage cabinet maintains the best working state under complex working conditions; especially in extreme cases, such as when electric sparks appear, by starting the second pipe metal cooling equipment for immersion cooling treatment, it can not only effectively reduce the temperature but also quickly extinguish the electric sparks, ensuring the safe and stable operation of the energy storage cabinet. This adaptive and intelligent cooling strategy not only improves the energy efficiency of the energy storage cabinet, but also enhances its safety and reliability.
[0124] Furthermore, the specific steps of the prediction algorithm in the battery cell performance prediction unit of this embodiment include:
[0125] D1. Use the acquired historical charge and discharge parameter data to construct the performance prediction input sequence for each battery in the cell ,in, It represents the discharge depth of the jth battery in the cell at time t, that is, the amount of electricity discharged at time t. The output power of the jth battery in the cell at time t, The output current of the jth battery in the cell at time t, It indicates the lower limit of the state of charge corresponding to the jth battery in the cell under discharge state, that is, the amount of power that needs to be retained to extend the battery life. The target state predicted for this process;
[0126] D2. Use the particle swarm optimization algorithm optimized by BP neural network to build a prediction model, and input the constructed battery performance prediction input sequence into the prediction model for training, and obtain the trained prediction model
[0127] D3. Deploy the obtained prediction model to the battery performance prediction unit, and predict the lower limit of the state of charge corresponding to each battery at time t during the discharge process based on the discharge parameter data of each battery in the battery cell collected in real time. , It represents the predicted lower limit of the state of charge corresponding to the jth battery in the cell under the discharge state;
[0128] D4. Input the battery performance prediction into the sequence , Replace with , , and repeat the D1-D3 process to predict the charging process of each battery ,in Indicates the upper limit of the state of charge corresponding to the jth battery in the cell under charging condition, Indicates the depth of charge of the jth battery in the cell at time t under charging status, It represents the predicted upper limit of the state of charge of the jth battery in the battery cell under charging condition.
[0129] The neural network nodes of the BP algorithm in this embodiment are calculated by the following formula: , where p represents the number of hidden layer nodes, Indicates that the number of input layer nodes is 4, Indicates that the number of output layer nodes is 5, is the offset, which is set to 4 here; therefore, in this embodiment ,In addition, the population size of the genetic algorithm is set to 300, the number of genetic iterations is 200, the mutation probability is 0.05, and the crossover probability is 0.1.
[0130] Furthermore, the specific working steps of the charge and discharge regulation strategy configured in the charge and discharge regulation unit of this embodiment include:
[0131] E1. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding discharge balance equation is: ,in It represents the total state of charge of all batteries in the cell at time t. represents the rated state of charge of the jth battery in the battery cell; m represents the total number of batteries in the battery cell; Indicates the discharge protection factor of the jth battery in the cell; j≤m;
[0132] E2. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding charging balance equation is: , Indicates the charging protection factor of the jth battery in the cell, Indicates the total state of charge of all cells at time t under charging condition;
[0133] E3. When the energy storage cabinet is in the fourth execution information state, an instruction to reduce the discharge power output is generated. The corresponding calculation formula for the reduced discharge power value is: ,in Indicates that the battery cell is The heat generated during the cycle is calculated using Joule's law and the output current. Since the current reduction time is very short, the heat generated during the current reduction period can be ignored. Indicates that the battery is The heat generated during the cycle, It indicates the output current that needs to be reduced at time t. U is the output port voltage of the battery cell, which is a constant value. Indicates the specific value of the discharge power that the battery cell needs to reduce; in this embodiment, the reduction of the output current is achieved by controlling the bidirectional converter at the energy storage cabinet interface through the instruction generated by PID; since charging and discharging are a symmetrical process, the calculation process of the charging power reduced during the charging process is the same as that of the E3 process;
[0134] E4. When the energy storage cabinet is in the fifth execution information state and the sixth execution information state, the bidirectional converter at the interface of the energy storage cabinet is controlled by the generated power-off instruction so that the output current and the input current become 0, thereby powering off the battery for protection.
[0135] The charge and discharge regulation strategy configured in this process realizes the intelligent management of the energy storage cabinet under different working conditions. Under normal working conditions, the stable operation of the battery cell is ensured through refined discharge and charge balance equations. When the temperature is high or abnormal conditions occur, the discharge power can be intelligently reduced or the power can be cut off, effectively preventing the battery cell from overheating and potential safety risks. This adaptive charge and discharge regulation strategy not only improves the utilization efficiency of the energy storage cabinet, but also significantly enhances its safety and reliability.
[0136] Example 2
[0137] See also Figure 2 Another embodiment provided by the present invention is based on the work flow of the corresponding unit of the double-protection energy storage cabinet security control system of Zhilian CNC, which specifically includes:
[0138] First, the integrated sensor configured in the battery monitoring unit in the intelligent control module monitors the charging and discharging parameters of the battery cells of the energy storage cabinet in real time, and inputs the monitored parameters into the battery performance prediction unit to use the prediction algorithm to predict the battery performance in real time. The charging and discharging adjustment unit uses the configured charging and discharging adjustment strategy based on the acquired charging and discharging parameters and predicted battery performance to adjust the charging and discharging process of the energy storage cabinet in real time.
[0139] Second, the temperature distribution data and abnormal gas concentration data in the energy storage cabinet are monitored simultaneously through the thermal imaging detection unit and the abnormal gas concentration monitoring unit in the thermal management module, and the monitoring results are fed back to the dual-defense control unit. The corresponding temperature control instructions or power-off instructions are generated using the configured dual-defense control threshold interval and control algorithm, and the generated instructions are fed back to the cloud platform for transmission;
[0140] Third, if a temperature control instruction is generated, the generated temperature control instruction is transmitted to the adaptive cooling control unit through the cloud platform, and the configured adaptive cooling strategy is used to control the cooling subsystem in the energy storage cabinet to perform adaptive dual-defense cooling on the inside of the energy storage cabinet, and the waste heat recovery unit is used to recycle and reuse the heat in the cooling process; if a power-off instruction is generated, the generated power-off instruction is transmitted to the charge and discharge regulation unit through the cloud platform, and the charge and discharge regulation strategy generated by the power-off instruction is used to control the power-off protection unit to perform adaptive power-off protection on the energy storage cabinet, and at the same time, the power-off result is transmitted to the adaptive cooling control unit through the power-off protection unit to perform dual-defense cooling protection after power off. When the energy storage cabinet is in a fault-free state, after power off, the internal temperature of the energy storage cabinet monitored by the thermal imaging detection unit returns to the dual-defense control threshold range, and a power-on instruction is sent to the charge and discharge regulation unit through the cloud platform to restore the charge and discharge state of the energy storage cabinet;
[0141] Fourth, the fault monitoring unit in the fault handling module uses the configured fault detection algorithm to perform real-time detection of faults in the first to third processes, and feeds back the detection results to the power-off protection unit and the early warning unit, which in turn performs power-off protection on the equipment and issues signal early warning for the detected faults. The early warning signal is transmitted to the remote control unit through the cloud platform, and the energy storage cabinet is remotely controlled and repaired using the configured Internet of Things algorithm.
[0142] Fifth, the strategies deployed in the first to fourth steps above are updated with real-time training and instruction transmission protection through the cloud platform.
[0143] Example 3
[0144] See also Figure 3 Another embodiment provided by the present invention is a security control method for a dual-defense energy storage cabinet based on Zhilian CNC, and the specific steps include:
[0145] S1. Use integrated sensors to monitor the charging and discharging parameters of the energy storage cabinet cells in real time, and use the prediction algorithm to predict the cell performance data based on the collected charging and discharging parameters;
[0146] S2. According to the acquired charging and discharging parameters and predicted battery cell performance data, the configured charging and discharging adjustment strategy is used to adjust the charging and discharging process of the energy storage cabinet in real time. At the same time, a thermal imager and a gas monitoring and analysis device are used to monitor the temperature distribution and abnormal gas concentration in real time during the charging and discharging process, and obtain temperature distribution data and abnormal gas concentration data;
[0147] S3. Generate temperature control instructions and power-off instructions based on the acquired temperature distribution data and abnormal gas concentration data using the configured dual-defense control threshold interval and control algorithm;
[0148] S4. When a temperature control instruction is generated, the cooling subsystem in the energy storage cabinet is controlled to perform adaptive dual-protection cooling on the inside of the energy storage cabinet through the configured adaptive cooling strategy. If a power-off instruction is generated, the configured charge and discharge adjustment strategy is used to perform power-off protection on the energy storage cabinet. When the temperature returns to the dual-protection control threshold range, the charge and discharge adjustment strategy is used to control the energy storage cabinet to restore the charge and discharge state.
[0149] S5. Perform abnormal detection on the process from S1 to S4 through the configured fault detection algorithm, and feed back the detection results to the charge and discharge adjustment strategy through the cloud platform to perform power-off protection on the energy storage cabinet. At the same time, the abnormal detection results are remotely fed back through the Internet of Things algorithm configured in the cloud platform, and the self-correction adjustment settings of the energy storage cabinet are performed based on the feedback results.
[0150] Example 4
[0151] The present embodiment discloses a computer-readable storage medium having computer instructions stored thereon. When the computer instructions are executed, the security control method of the double-defense energy storage cabinet based on Zhilian CNC is executed.
[0152] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are within the protection of the present invention.
Claims
1. Based on the double-protection energy storage cabinet security control system of Zhilian CNC, It is characterized in that include: Intelligent control module and thermal management module; The thermal management module includes a thermal imaging detection unit, an abnormal gas concentration monitoring unit and an adaptive cooling control unit; The intelligent control module includes a battery cell monitoring unit, a battery cell performance prediction unit, a charge and discharge adjustment unit and a dual protection control unit; The thermal imaging detection unit monitors the temperature distribution data of the energy storage cabinet during the charging and discharging process in real time through a thermal imager deployed in the energy storage cabinet, and feeds back the real-time monitored temperature distribution data to the dual-defense control unit; the abnormal gas concentration monitoring unit monitors the concentration of abnormal gases in the energy storage cabinet during the charging and discharging process in real time through a gas monitoring device deployed in the energy storage cabinet, and feeds back the real-time monitored abnormal gas concentration data to the dual-defense control unit; the adaptive cooling control unit is used to control the cooling subsystem in the energy storage cabinet to perform adaptive dual-defense cooling on the inside of the energy storage cabinet according to the temperature control instruction generated by the dual-defense control unit and the configured adaptive cooling strategy; the thermal imager is configured with a thermal image analysis algorithm; The battery cell monitoring unit is used to monitor the charging and discharging parameters of the battery cells of the energy storage cabinet in real time by using the configured integrated sensors; the battery cell performance prediction unit is used to predict the battery cell performance in real time based on the detected charging and discharging parameters and the configured prediction algorithm to obtain the predicted battery cell performance data; the charge and discharge adjustment unit is used to adjust the charging and discharging process of the energy storage cabinet in real time based on the charge and discharge parameters and the predicted battery cell performance data and the configured charge and discharge adjustment strategy; the dual-protection control unit is used to construct a dual-protection control threshold interval, and generate a temperature control instruction and a power-off instruction based on the temperature distribution data and the abnormal gas concentration data by using the integrated control algorithm; The thermal image analysis algorithm specifically includes: A1. Collect historical thermal image data of the energy storage cabinet, mark the corresponding temperature for each region of the collected historical thermal image data, and use anchor frames to mark the sparks in the historical thermal images where sparks appear, and mark the corresponding temperature when the sparks appear, and obtain the marked historical thermal image data; A2. Use the YOLOv5 network pre-trained with historical thermal image data to build a thermal image analysis model, and use the annotated historical thermal image data to fine-tune the built thermal image analysis model to obtain a fine-tuned thermal image analysis model; A3. Deploy the fine-tuned thermal image analysis model to the thermal imager in the thermal imaging detection unit for real-time monitoring to obtain the temperature distribution data in the energy storage cabinet and the extreme state data of the electric spark; The specific working steps of the dual protection control unit include: B1. Divide the working temperature of the energy storage cabinet into 4 working temperature ranges, and set the upper limit threshold of the four working temperature ranges to ,and , corresponding to the low temperature upper limit threshold, the medium temperature upper limit threshold, and the high temperature upper limit threshold respectively; B2. Set the abnormal gas explosion concentration threshold to , the corresponding explosion temperature threshold is ,and ; B3. Temperature distribution data at time t obtained in real time based on the set threshold and the thermal imaging detection unit and the abnormal gas concentration monitoring unit Abnormal gas concentration data , build a dual defense control threshold range, specifically: when When the first execution information is triggered, the energy storage cabinet maintains the current working state, generates an initial temperature control instruction, and does not generate a power-off instruction; when When the second execution information is triggered, the energy storage cabinet maintains the current working state, generates the first temperature control instruction, and does not generate the power-off instruction; when When the third execution information is triggered, the energy storage cabinet maintains the current working state, generates the second temperature control instruction, and does not generate the power-off instruction; when , the fourth execution information is triggered, at which time a third temperature control instruction is generated, and an instruction to reduce power output is generated; when , the fifth execution information is triggered, at which time a fourth temperature control instruction is generated, and a power-off instruction is generated; When electric sparks are detected by the thermal imaging detection unit, When the sixth execution information is triggered, a fourth temperature control instruction is generated, and a power-off instruction is generated; wherein R represents an arbitrary real number; the temperature control instruction includes an initial temperature control instruction, a first temperature control instruction, a second temperature control instruction, a third temperature control instruction and a fourth temperature control instruction; B4. Use the reinforcement learning algorithm and the PID algorithm to build a servo control model, and use the thermal imaging detection unit and the abnormal gas concentration monitoring unit to collect historical temperature distribution data, historical abnormal gas concentration data, and dual-defense control threshold intervals, train the servo control model, and obtain a trained servo control model; B5. Deploy the trained servo control model to the dual-defense control unit, and input the temperature distribution data and abnormal gas concentration data collected in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit into the servo control model to generate corresponding temperature control instructions.
2. The double-protection energy storage cabinet security control system based on Zhilian CNC as claimed in claim 1, It is characterized in that The specific working steps of the adaptive cooling strategy configured in the adaptive cooling control unit include: C1. The energy storage cabinet cooling subsystem is equipped with a double-pipe metal cooling device, which includes a first-pipe metal cooling device and a second-pipe metal cooling device. The first-pipe metal cooling device is cooled by water, and the second-pipe metal cooling device is cooled by an immersion coolant with a higher specific heat capacity than water. The thermal conductivity of the first-pipe metal cooling device is , the specific heat capacity of water is ; C2. According to the temperature distribution data and abnormal gas concentration data monitored in real time by the thermal imaging detection unit and the abnormal gas concentration monitoring unit, perform the following steps: C21, when the first execution information is triggered and the servo control model generates an initial temperature control instruction, the dual-pipe metal cooling device is not started to perform cooling operation; C22, when the second execution information is triggered and the servo control model generates the first temperature control instruction, the double-pipe metal cooling device is started and the first water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation; C23, when the third execution information is triggered and the servo control model generates a second temperature control instruction, the dual-pipe metal cooling device is started and a second water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation; C24, when the fourth execution information is triggered and the servo control model generates the third temperature control instruction, the dual-pipe metal cooling device is started and the third water cooling speed is generated. , control the first pipeline metal cooling equipment to perform cooling operation.
3. The double-protection energy storage cabinet security control system based on Zhilian CNC as claimed in claim 2, It is characterized in that The specific working steps of the adaptive cooling strategy also include: C25. When the fifth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , controlling the first pipeline metal cooling device to perform cooling operation; C26. When the sixth execution information is triggered and the servo control model generates the fourth temperature control instruction, the dual-pipe metal cooling device is started and the fourth water cooling speed is generated. , control the first pipeline metal cooling device to perform cooling operation, and at the same time start the second pipeline metal cooling device to perform immersion cooling treatment on the battery cell.
4. The double-protection energy storage cabinet security control system based on Zhilian CNC as claimed in claim 3, It is characterized in that The specific steps of the prediction algorithm include: D1. Obtain historical charging and discharging parameters and construct a performance prediction input sequence for each battery in the cell ,in, represents the discharge depth of the jth battery in the cell at time t, The output power of the jth battery in the cell at time t, The output current of the jth battery in the cell at time t, Indicates the lower limit of the state of charge corresponding to the jth battery in the cell under discharge state; D2. Use the particle swarm optimization algorithm optimized by BP neural network to build a prediction model, and input the performance prediction input sequence of the battery into the prediction model for training to obtain a trained prediction model; D3. Deploy the prediction model to the battery performance prediction unit, and predict the discharge performance of each battery in the battery according to the real-time collected discharge parameters. , It represents the predicted lower limit of the state of charge corresponding to the jth battery in the cell under the discharge state; D4. Input the battery performance prediction into the sequence , Replace with , , and repeat the D1-D3 process to predict the charging process of each battery ,in Indicates the upper limit of the state of charge corresponding to the jth battery in the cell under charging condition, Indicates the depth of charge of the jth battery in the cell at time t under charging status, It represents the predicted upper limit of the state of charge of the jth battery in the cell under charging condition.
5. The double-protection energy storage cabinet security control system based on Zhilian CNC as claimed in claim 4, It is characterized in that The specific working steps of the charge and discharge regulation strategy configured in the charge and discharge regulation unit include: E1. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding discharge balance equation is: ,in It represents the total state of charge of all batteries remaining in the cell at time t under discharge state. represents the rated state of charge of the jth battery in the cell; m represents the total number of batteries in the cell; Indicates the discharge protection factor of the jth battery in the cell; j≤m; E2. When the energy storage cabinet is in the first to third execution information states, no power-off instruction is generated, and the corresponding charging balance equation is: , Indicates the charging protection factor of the jth battery in the cell; Indicates the total state of charge of all cells at time t under charging condition; E3. When the energy storage cabinet is in the fourth execution information state, an instruction to reduce the discharge power output is generated. The corresponding calculation formula for the reduced discharge power value is: ,in Indicates that after executing the fourth execution information state, the battery The heat generated during the cycle, Indicates that the battery is The heat generated during the cycle, Indicates the output current that the cell needs to reduce, U is the cell output port voltage, Indicates the specific value by which the discharge power of the battery cell needs to be reduced; is the total resistance of the energy storage cabinet cells; E4. When the energy storage cabinet is in the fifth execution information state and the sixth execution information state, the bidirectional converter at the interface of the energy storage cabinet is controlled by the generated power-off instruction so that the output current and the input current become 0.
6. A security control method for a dual-defense energy storage cabinet based on Zhilian CNC, which is implemented based on a security control system for a dual-defense energy storage cabinet based on Zhilian CNC as claimed in any one of claims 1 to 5. It is characterized in that include: S1. Use integrated sensors to monitor the charging and discharging parameters of the energy storage cabinet cells in real time, and use the prediction algorithm to predict the cell performance data based on the collected charging and discharging parameters; S2. According to the acquired charging and discharging parameters and predicted battery cell performance data, the configured charging and discharging adjustment strategy is used to adjust the charging and discharging process of the energy storage cabinet in real time. At the same time, a thermal imager and a gas monitoring and analysis device are used to monitor the temperature distribution and abnormal gas concentration in real time during the charging and discharging process, and obtain temperature distribution data and abnormal gas concentration data; S3. Generate temperature control instructions and power-off instructions based on the acquired temperature distribution data and abnormal gas concentration data using the configured dual-defense control threshold interval and control algorithm; S4. When a temperature control instruction is generated, the cooling subsystem in the energy storage cabinet is controlled through the configured adaptive cooling strategy to perform adaptive dual-protection cooling on the inside of the energy storage cabinet. If a power-off instruction is generated, the configured charge and discharge adjustment strategy is used to perform power-off protection on the energy storage cabinet. When the temperature returns to the dual-protection control threshold range, the charge and discharge adjustment strategy is used to control the energy storage cabinet to restore the charge and discharge state.
7. A computer-readable storage medium, It is characterized in that Computer instructions are stored thereon, and when the computer instructions are executed, the security control method of the double-protection energy storage cabinet based on Zhilian CNC as described in claim 6 is executed.
