Energy storage system for glue-coated PACK and working method thereof

By designing a PACK energy storage system for glue coating, using matching models and strategy customization methods to monitor and optimize power management in real time, the problem that traditional energy storage systems cannot match glue coating process requirements and cope with power fluctuations is improved, and storage is improved Energy battery pack stability and use efficiency.

CN118572901BActive Publication Date: 2024-11-15SUZHOU WEIDI PRECISION CONTROL SYST CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411038316.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-15
Estimated Expiration
2044-07-31

AI Technical Summary

Technical Problem

Traditional energy storage systems cannot accurately match the glue coating process requirements of different PACK specifications and cannot effectively deal with power fluctuations, resulting in low efficiency and poor stability of energy storage battery packs.

Method used

A PACK energy storage system for glue coating was designed, including collection modules, processing modules and output modules. Through matching models and strategy customization methods, the status of glue coating equipment and energy storage battery packs is monitored in real time, and reasonable discharge is formulated. Strategy and response strategies, optimize power management and respond to power fluctuations.

Benefits of technology

It improves the efficiency and stability of the energy storage battery pack, ensures the efficiency and quality of the glue coating process, and reduces production interruptions and losses caused by power problems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118572901B_ABST
    Figure CN118572901B_ABST
Patent Text Reader

Abstract

The invention discloses an energy storage system for a PACK for gluing and a working method thereof, and relates to the field of PACK gluing. It includes a collection module, a processing module, an output module and a plurality of energy storage battery packs, and the collection module stores a gluing process information data set, a PACK order information data set and a response report information data set. The processing module of the present invention utilizes a matching model, and can accurately match PACKs of different specifications to corresponding gluing processes and equipment based on a variety of data sets, and formulate a reasonable discharge strategy through a strategy customization method, monitor and analyze the operating status of the gluing equipment and the energy storage battery pack in real time to obtain abnormal information, and then formulate a reasonable response strategy in conjunction with a fluctuation response method, which helps to optimize the use efficiency of the energy storage battery pack, improve the efficiency and quality of the gluing process, ensure the stable operation of the energy storage battery pack, and reduce production interruptions and losses caused by power problems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of PACK gluing, and in particular to an energy storage system for a PACK for gluing and a working method thereof. Background Art

[0002] In the field of energy storage systems, with the complexity of the glue coating process and the high requirements for power supply stability, an energy storage system that can accurately match the glue coating process requirements is needed. Although traditional energy storage systems can provide basic power support, they are insufficient in dealing with changes in the glue coating process, power fluctuations, and optimizing the performance of energy storage battery packs.

[0003] The authorization publication number is CN114768142B and its name is a control method for an energy storage system and an energy storage system. The energy storage system includes multiple energy storage sub-units and a temperature control system. The temperature control system is connected to the multiple energy storage sub-units through air ducts. The energy storage sub-units are all provided with fire detection sensors and ventilation systems. The fire detection sensors include combustible gas sensors. The control method of the energy storage system includes: obtaining detection information generated by each fire detection sensor, and controlling the operating state of the energy storage system according to the detection information, including: closing the air duct corresponding to the energy storage sub-unit with a combustible gas concentration greater than or equal to a preset first-level alarm value, and controlling the ventilation system to extinguish the energy storage sub-unit, so as to perform targeted fire control to avoid affecting the normal operation of the energy storage sub-unit.

[0004] The above-mentioned technology focuses on detecting safety information such as the concentration of combustible gas in each energy storage sub-unit, and specifically shutting down or fire-fighting the energy storage sub-units with safety hazards, thereby avoiding affecting the entire energy storage system and ensuring the safety of the energy storage system. However, different PACK specifications require matching corresponding gluing processes and corresponding gluing equipment, and the corresponding gluing processes and corresponding gluing equipment have different power requirements. Therefore, this type of technology cannot adjust the discharge strategy of the energy storage system according to the corresponding gluing process and the corresponding gluing equipment, nor can it deal with the problems caused by power fluctuations to the energy storage system. Summary of the invention

[0005] The object of the present invention is to provide an energy storage system for glue-coated PACK and a working method thereof, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] In the first aspect, an energy storage system for glue-coated PACK is designed, including a collection module, a processing module, an output module, and a plurality of energy storage battery packs;

[0008] The collection module stores the glue coating process information data set, the PACK order information data set and the response report information data set, wherein the glue coating process information data set includes the glue coating equipment information data set, the collection module is provided with visual analysis software and obtains the PACK information data set by scanning the PACK based on the PACK order information data set, and the collection module obtains the monitoring data set by monitoring the glue coating equipment and the energy storage battery group through the power system monitoring equipment, and the monitoring data set includes the real-time monitoring data set and the historical monitoring data set;

[0009] The processing module is provided with a matching model, and the matching model obtains the corresponding gluing process and the corresponding gluing equipment based on the matching of the PACK information data set, the PACK order information data set and the gluing process information data set in the collection module. The processing module obtains an adjustment report based on the corresponding gluing process and the corresponding gluing equipment through a strategy customization method, which is used to formulate a reasonable discharge strategy. The processing module obtains an abnormal information data set based on real-time analysis of the real-time monitoring data set, and finally obtains a response report based on the abnormal information data set through a fluctuation response method, which is used to respond to power fluctuations and ensure stable operation of the energy storage battery pack;

[0010] The output module sends instructions to the BMS to adjust the discharge strategy of the energy storage battery group based on the adjustment report of the processing module, and sends instructions to the EMS to adjust the response strategy of the energy storage battery group based on the response report of the processing module;

[0011] The energy storage battery pack is used to receive instructions from the BMS and EMS to provide stable and reliable power support for the corresponding glue coating process and the corresponding glue coating equipment;

[0012] The strategy customization method includes:

[0013] S1: The processing module determines the required parameters of the corresponding glue coating process based on the glue coating process information data set, then determines the performance parameters of the corresponding glue coating equipment based on the glue coating equipment information data set, and finally obtains the number of PACKs to be processed based on the PACK information data set;

[0014] S2: Based on the required parameter calculation and the performance parameter calculation, the power demand of each PACK and the number of PACKs processed per unit time are obtained. The total processing time is calculated based on the number of PACKs to be processed and the number of PACKs processed per unit time. The total power demand is then calculated based on the number of PACKs to be processed and the power demand of each PACK.

[0015] S3: Based on the total power demand and total processing time demand, combined with the real-time monitoring data set and required parameters, the corresponding discharge rate is calculated; based on the total processing time and the discharge rate, the corresponding discharge time is calculated; based on the real-time monitoring data set and the total power demand, the discharge sequence is adjusted and optimized, and finally a regulation report is summarized.

[0016] Furthermore, the method for adjusting and optimizing comprises:

[0017] A1: based on the realization status information set, the current status of each energy storage battery group is obtained, and multiple energy storage battery groups with the best status are selected to obtain an energy storage battery set;

[0018] A2: Based on the total power demand and the discharge capacity of each battery group in the energy storage battery set, the energy storage battery group that meets the total power demand and has sufficient discharge capacity is matched, that is, the optimal energy storage battery set is obtained;

[0019] A3: A polling algorithm is used to obtain the discharge order based on the best energy storage battery set.

[0020] Furthermore, the method for creating the matching model includes:

[0021] B1: Preprocess and integrate the PACK information dataset, PACK order information dataset, and glue coating process information dataset to obtain standard data, and divide the standard data into training set, validation set, and test set;

[0022] B2: Extract features based on the training set to construct feature vectors, and select the random forest algorithm based on the training set to train and obtain the initial model;

[0023] B3: Train and optimize the initial model to obtain a matching model, and update the matching model regularly.

[0024] Furthermore, the training, optimization and updating methods of the matching model include:

[0025] C1: Model evaluation: evaluate the trained initial model based on the validation set to obtain the evaluation results, which are used to analyze the performance of the model and identify existing problems and improvements.

[0026] C2: Model optimization, optimize the model based on the evaluation results, and use the grid search method to tune the parameters;

[0027] C3: Model verification, verifying the optimized model based on the test set to ensure the generalization ability of the model on unknown data;

[0028] C4: Model update: regularly update the matching model and use incremental learning methods to update the model in real time.

[0029] Furthermore, the fluctuation coping method includes:

[0030] D1: Based on the abnormal information data set, a statistical analysis algorithm is used to analyze and extract power fluctuation characteristics to obtain a fluctuation event information data set, wherein the fluctuation event information data set includes a power fluctuation characteristic information set and a fluctuation event type;

[0031] D2: Based on the fluctuation event information dataset and the response report information dataset, the corresponding response report is obtained by matching the cosine similarity algorithm.

[0032] Furthermore, the fluctuation coping method also includes:

[0033] E1: If the matching fails, the response report information dataset is retrieved using retrieval technology to obtain the response report with the most similar power fluctuation characteristics and fluctuation event types as the fluctuation event information dataset, and the most similar response report is compared with the fluctuation event information dataset to obtain the difference and similarity dataset;

[0034] E2: Determine the same report information in the most similar response reports based on the same data set in the similarity and difference data set, extract the key difference points through the 3sigma algorithm based on the difference data set in the similarity and difference data set, match it with the response report information data set again based on the key difference points to obtain the abnormal report information, and combine the same report information and the abnormal report information to obtain a customized response report.

[0035] Furthermore, the processing module sets corresponding warning thresholds for the gluing equipment and the energy storage battery assembly based on the gluing equipment information data set, multiple energy storage battery groups and the monitoring data set, and the warning threshold setting method includes:

[0036] Y1: Based on the historical monitoring data set, the power fluctuation data of the glue coating equipment and the energy storage battery group under normal operation is obtained by using the statistical analysis method, so as to obtain the historical fluctuation data set. Then, based on the historical fluctuation data set, the initial warning threshold value is set for each power parameter of the glue coating equipment and the energy storage battery group by using the statistical distribution method, so as to obtain the initial warning threshold value set;

[0037] Y2: Tests are performed based on the initial warning threshold set to obtain test results, and the final warning threshold information set is optimized based on the test results, so that the abnormal information data set can be obtained in a timely manner.

[0038] Furthermore, the processing module is connected to a control terminal, and the collection module, the processing module, the output module, the BMS and the EMS are connected via a dedicated local area network, and the method for establishing the dedicated local area network includes:

[0039] T1: Information acquisition, obtaining basic networking information of the collection module, processing module, output module, BMS and EMS, the basic networking information including device IP;

[0040] T2: Equipment connection, the control terminal automatically allocates the communication channel to match the IP of the collection module, the IP of the processing module, the IP of the output module, the IP of the BMS, and the IP of the EMS;

[0041] T3: Configure the network, control the terminal to configure the corresponding network services and security measures for the communication channel, and create a private LAN.

[0042] In a second aspect, based on the above-mentioned energy storage system for glue-coated PACK, a working method of the energy storage system for glue-coated PACK is designed.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The processing module uses a matching model to accurately match PACKs of different specifications to corresponding glue coating processes and equipment based on multiple data sets, and formulates reasonable discharge strategies through strategy customization methods, monitors and analyzes the operating status of glue coating equipment and energy storage battery packs in real time to obtain abnormal information, and then formulates reasonable response strategies in conjunction with the fluctuation response method, which helps to optimize the use efficiency of energy storage battery packs, improve the efficiency and quality of the glue coating process, ensure the stable operation of energy storage battery packs, and reduce production interruptions and losses caused by power problems.

[0045] At the same time, the strategy customization method can accurately match power demand, optimize processing time and power management, improve the utilization efficiency of energy storage battery packs, enhance system stability and reliability, and the fluctuation response method and warning threshold setting method can quickly respond to power fluctuations, accurately match the response report from the response report information data set, and flexibly customize the response report even if the match fails, thereby improving system stability and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the principle structure of the present invention;

[0047] Figure 2 This is a schematic diagram of the principle structure of the strategy customization method of the present invention;

[0048] Figure 3 It is a schematic diagram of the principle structure of the wave response method of the present invention;

[0049] Figure 4 It is a plan view of the BMS of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] In modern battery production lines, especially for the gluing process of PACK (battery pack), a stable and reliable power support is required to ensure the gluing quality and production efficiency. The energy storage system for gluing PACK provided by the present invention is designed for this purpose. In the battery production workshop, the system needs to be placed in a clean and tidy environment to avoid dust and other impurities from damaging the equipment. At the same time, the workshop should have good ventilation conditions to ensure that the equipment will not overheat during operation. Before using this system for gluing operations, it is necessary to ensure that the water, electricity and gas supply in the workshop is stable, and check whether the power of the energy storage battery pack is sufficient. Operators need to undergo professional training and be familiar with the operating procedures and safety requirements of the equipment to ensure that relevant safety regulations are followed during operation to avoid accidents. The system collects gluing process information, PACK order information and response report information through the collection module, and monitors the operating status of the gluing equipment and the energy storage battery pack in real time. The processing module matches the corresponding gluing process and equipment based on this information, and customizes a reasonable discharge strategy. When power fluctuations are detected, the system will quickly generate a response report to guide EMS to adopt corresponding response strategies to ensure the stable operation of the energy storage battery pack and provide stable and reliable power support for the gluing process.

[0052] like Figure 1-Figure 4 As shown, the present invention provides a technical solution: an energy storage system for glue-coated PACK, comprising a collection module, a processing module, an output module and a plurality of energy storage battery packs;

[0053] The collection module stores the glue coating process information data set, the PACK order information data set and the response report information data set. The glue coating process information data set includes the glue coating equipment information data set. The collection module is provided with visual analysis software and obtains the PACK information data set by scanning the PACK based on the PACK order information data set. The collection module monitors the glue coating equipment and the energy storage battery group through the power system monitoring equipment to obtain the monitoring data set, and the monitoring data set includes the real-time monitoring data set and the historical monitoring data set.

[0054] It should be noted that the visual analysis software can select OpenCV. The glue coating process information data set includes process parameters: glue type / glue coating amount / glue ratio / glue coating times, etc., operation process: preparation work / glue coating operation / curing treatment / drying treatment, etc., temperature, humidity, pressure and other environmental parameters, glue coating equipment information data set includes basic equipment parameters: glue coating speed / glue coating width / glue coating thickness / power supply requirements / equipment size / equipment weight, control system: control method / control panel / programmable parameters, glue coating method: roller glue coating / scraper glue coating / spraying / other specific glue coating methods, glue supply system: glue bucket capacity / glue pump type and performance / pipeline layout and material / glue flow and supply stability, maintenance: maintenance and maintenance Maintenance cycle / maintenance content, other information: equipment working environment requirements / equipment supporting equipment / equipment performance indicators, digital information: equipment number / serial number / production date / manufacturer information / equipment operation status record, PACK order information data set includes basic information: order ID / user ID / order time / order status, product information: product ID / product name / product quantity / product unit price, logistics information: delivery time / logistics company / logistics order number, response report information data set includes report ID, fluctuation event type, power fluctuation feature information, response strategy, response effect evaluation, implementation steps, resource requirements, related reports, remarks, timestamp, keywords, PACK information data set includes PACK ID / size information / shape information / surface quality / glue coating area / material information / defect detection / status information, operating status of glue coating equipment, real-time status of energy storage battery pack, relevant parameters of glue coating equipment, monitoring data sets include glue coating speed, pressure, temperature, etc., battery status parameters such as voltage, current, temperature of energy storage battery pack, warning information issued when glue coating equipment or energy storage battery pack exceeds the preset range or encounters an abnormality, such as abnormal pressure of glue coating equipment, excessive temperature of energy storage battery pack, etc., time stamp of each real-time data point to ensure the real-time and accuracy of data, regularly back up all stored data to ensure that data can be quickly restored after loss or damage.

[0055] A matching model is provided in the processing module. The matching model obtains the corresponding gluing process and the corresponding gluing equipment based on the PACK information data set, the PACK order information data set and the gluing process information data set in the collection module. The processing module obtains an adjustment report based on the corresponding gluing process and the corresponding gluing equipment through a strategy customization method, which is used to formulate a reasonable discharge strategy. The processing module obtains an abnormal information data set based on real-time analysis of the real-time monitoring data set. Finally, a response report is obtained based on the abnormal information data set through a fluctuation response method, which is used to cope with power fluctuations and ensure stable operation of the energy storage battery pack.

[0056] The output module sends instructions to the BMS to adjust the discharge strategy of the energy storage battery group based on the adjustment report of the processing module, and sends instructions to the EMS to adjust the response strategy of the energy storage battery group based on the response report of the processing module.

[0057] It should be noted that BMS is a battery management system and EMS is an energy management system. The adjustment report includes data such as the current state, expected demand, grid status, and discharge strategy adjustment recommendations based on these data. The response report includes predictions and evaluations of emergencies (such as grid failures, load changes, etc.), as well as response strategy recommendations based on these predictions. Based on the recommendations in the adjustment report, the output module will generate corresponding instructions, which will tell the BMS how to adjust the discharge strategy of the energy storage battery group, such as adjusting the discharge current, discharge time, etc. Based on the recommendations in the response report, the output module will generate corresponding instructions, which will tell the EMS how to adjust the response strategy of the energy storage battery group, such as switching to backup power supply, adjusting power output, etc. The output module sends the instructions to the BMS and EMS through a proprietary LAN.

[0058] The energy storage battery pack is used to receive instructions from the BMS and EMS to provide stable and reliable power support for the corresponding glue coating process and the corresponding glue coating equipment.

[0059] like Figure 2 As shown, the strategy customization method includes: S1: the processing module determines the required parameters of the corresponding glue coating process based on the glue coating process information data set, and then determines the performance parameters of the corresponding glue coating equipment based on the glue coating equipment information data set, and finally obtains the number of PACKs to be processed based on the PACK information data set, S2: based on the required parameter calculation and the performance parameter calculation, the power demand of each PACK and the number of PACKs to be processed per unit time are obtained, the total processing time is calculated based on the number of PACKs to be processed combined with the number of PACKs to be processed per unit time, and then the total power demand is calculated based on the number of PACKs to be processed and the power demand of each PACK, S3: based on the total power demand and the total processing time demand combined with the real-time monitoring data set and the required parameters, the corresponding discharge rate is calculated, the corresponding discharge time is calculated based on the total processing time and the discharge rate, the discharge sequence is adjusted and optimized based on the real-time monitoring data set and the total power demand, and finally the adjustment report is summarized.

[0060] It should be noted that the processing module will first read the relevant information in the glue coating process information data set to determine the required parameters of the corresponding glue coating process. These required parameters may include glue coating speed, glue coating temperature, glue coating pressure, etc. These parameters are crucial to ensuring the quality of glue coating. Next, the processing module will determine the performance parameters of the corresponding glue coating equipment based on the glue coating process information data set. The performance parameters usually include the maximum output power and maximum working efficiency of the equipment. These parameters determine the maximum working capacity that the equipment can achieve under given conditions. The processing module will also obtain the current number of PACKs to be processed from the PACK information data set. This is the basic data for determining the power demand and processing time. According to the required parameters of the glue coating process and the performance parameters of the glue coating equipment, the processing module calculates the power required for each PACK in the glue coating process, and calculates the total power demand in combination with the number of PACKs to be processed. According to the performance parameters of the glue coating equipment (such as working efficiency) and the required parameters of the glue coating process, the processing module The module calculates the number of PACKs that can be processed per unit time, and calculates the total processing time in combination with the number of PACKs to be processed. The processing module calculates the appropriate discharge rate based on the total power demand and the total processing time demand, combined with the current state of the energy storage battery group in the real-time monitoring data set (such as power, temperature, etc.), and calculates the corresponding discharge time based on the total processing time and discharge rate. In order to improve the utilization efficiency and stability of the energy storage battery group, the processing module will optimize the discharge sequence based on the real-time monitoring data set (such as the power, temperature and other status information of each battery group) and the total power demand. For example, battery groups with sufficient power and moderate temperature can be used for discharge first, or they can be rotated according to the remaining life of the battery group to extend the service life of the entire energy storage system. The processing module summarizes the above calculation and optimization results into an adjustment report, which details the optimal discharge strategy for the current glue coating process and equipment, including discharge rate, discharge time, discharge sequence and other information.

[0061] The adjustment and optimization method includes: A1: obtaining the current state of each energy storage battery group based on the implementation state information set, and selecting multiple energy storage battery groups with the best state to obtain an energy storage battery set; A2: based on the total power demand and the discharge capacity of each battery group in the energy storage battery set, matching to obtain energy storage battery groups that meet the total power demand and have sufficient discharge capacity, that is, to obtain the best energy storage battery set; A3: using a polling algorithm based on the best energy storage battery set to obtain a discharge order.

[0062] It should be noted that the processing module processes the data in the realization status information set and evaluates the current status of each energy storage battery pack. This usually involves a comprehensive analysis of multiple indicators such as power, temperature, and health status to determine the performance status and remaining working capacity of the battery pack. From all energy storage battery packs, select those battery packs with the best current status. These battery packs usually have higher power, suitable temperature, and better health status. These battery packs are combined into energy storage battery sets. According to the calculation in the strategy customization method, the processing module has determined the total power demand of the glue coating process. For each battery pack in the multiple energy storage battery packs, the processing module evaluates its discharge energy according to its performance parameters and current status information. The power is the maximum power output that the battery pack can provide under given conditions. According to the total power demand and the discharge capacity of each battery pack, the processing module matches and selects those energy storage battery packs that can meet the total power demand and have sufficient discharge capacity to form the best energy storage battery set. This matching process may need to consider multiple factors, such as the remaining power of the battery pack, discharge efficiency, service life, etc. The polling algorithm is a commonly used scheduling algorithm used to evenly distribute tasks or requests among multiple available resources. In this scenario, the polling algorithm is used to determine the discharge order of each battery pack in the energy storage battery set. The processing module uses the polling algorithm to determine the discharge order based on the best energy storage battery set, and selects the battery packs in order from the battery packs. Select battery groups for discharge. When the power of a battery group is lower than the preset threshold or reaches other discharge conditions, it is removed from the discharge sequence and the next battery group is selected from the remaining battery groups to continue discharging. In this way, while ensuring that the power demand is met, the energy storage battery group can be used as evenly as possible to extend its service life. The specific implementation steps of the polling algorithm are as follows: 1. Initialization: sort the best energy storage battery set (that is, the battery set that meets the total power demand and has sufficient discharge capacity) in a certain order (such as numbering order, order of joining the system, etc.), 2. Select the first battery group: select the first battery group from the sorted energy storage battery set to start discharging, 3. Discharge process: according to the discharge strategy and the current The discharge capacity of the previous battery pack is used to discharge the selected battery pack while monitoring its status (such as power, temperature, etc.). 4. Check the discharge conditions: During the discharge process, the discharge conditions of the current battery pack are continuously checked. If the power of the battery pack is lower than the preset threshold (such as 20% of the remaining power) or other discharge conditions are met (such as too high temperature), the discharge of the battery pack is stopped. 5. Select the next battery pack: Select the next battery pack from the remaining battery packs (usually the next one in the sorting order) and repeat steps 3 and 4. 6. Loop: This process will continue to loop until all battery packs meet the discharge conditions or reach the predetermined discharge time, power demand and other goals. 7.Dynamic adjustment: During the discharge process, if the collection module detects a significant change in the state of the glue coating equipment or the energy storage battery pack (such as a sudden increase in power demand, a battery pack failure, etc.), the processing module can dynamically adjust the discharge strategy and discharge sequence based on real-time monitoring data.

[0063] The method for creating the matching model includes: B1: pre-processing and integrating the PACK information data set, the PACK order information data set and the glue coating process information data set to obtain standard data, and dividing the standard data into a training set, a validation set and a test set; B2: extracting features based on the training set to construct a feature vector, and selecting a random forest algorithm based on the training set to train to obtain an initial model; B3: training and optimizing the initial model to obtain a matching model, and regularly updating the matching model.

[0064] It should be noted that the PACK information dataset, PACK order information dataset and glue coating process information dataset are collected, and whether there are missing values, outliers or duplicate records in the dataset is checked. Missing values ​​are filled (such as using mean, median, mode, etc.), or records with missing values ​​are deleted. Outliers are processed, such as standardization, normalization or threshold filtering. The three datasets are merged according to a common identifier (such as order number, product number, etc.). According to the needs, the required fields are selected for integration to form standard data. The standard data is divided into a certain proportion (such as 70% training set, 15% validation set, 15% test set) is divided into training set, validation set and test set, and features related to the matching model are extracted from the standard data. The features may include the physical properties of the PACK, order requirements, glue coating process parameters, etc. The extracted features are encoded (such as one-hot encoding, label encoding, etc.) for the convenience of model processing, and the extracted features are combined into feature vectors. Each sample corresponds to a feature vector. The random forest algorithm is selected as the basis of the model, and the random forest model is trained using the training set. The hyperparameters of the model (such as the number of trees, maximum depth, splitting criteria, etc.) are set, and the hyperparameters are tuned using methods such as cross-validation.

[0065] The training, optimization and updating methods of the matching model include: C1: model evaluation, evaluating the trained initial model based on the validation set to obtain an evaluation result, which is used to analyze the performance of the model and find out the existing problems and improvement points; C2: model optimization, optimizing the model based on the evaluation result, and using the grid search method to tune the parameters; C3: model verification, verifying the optimized model based on the test set to ensure the generalization ability of the model on unknown data; C4: model update, regularly updating the matching model, and using the incremental learning method to update the model in real time.

[0066] It should be noted that the pre-divided validation set is loaded from the dataset, and the trained initial model is used to predict the validation set to obtain the prediction results. According to the prediction results and the true labels of the validation set, the evaluation indicators such as accuracy, recall, F1 score, AUC-ROC, etc. are calculated. These indicators will be used to analyze the performance of the model, analyze the evaluation results, and find out the possible problems and improvement points of the model. For example, if it is found that the model performs poorly on a certain type of data, more feature engineering or data enhancement may be required for this type of data. Based on the evaluation results, the grid search method is selected to optimize the model, and the search range is set for the model parameters that need to be optimized. These parameters may include the number of trees in the random forest, the maximum depth, the splitting criterion, etc. An exhaustive search is performed within the specified parameter range to find the parameter combination that optimizes the model performance. The model is retrained using the optimized parameter combination, and the pre-divided test set is loaded from the dataset. The optimized model is used to predict the test set to obtain the prediction results. Test results, calculate validation indicators based on the predicted results and the true labels of the test set, which will be used to evaluate the generalization ability of the model on unknown data, analyze the validation results, and ensure that the model has good performance on unknown data. If the validation results are not ideal, it may be necessary to return to step C2 for further optimization. Over time, new data is collected. These data may include new PACK information, order information, and glue coating process information. Preprocess the newly collected data to keep it in the same format and feature space as the existing data. Use incremental learning methods to use the new data to update the model. Incremental learning allows the model to learn new data while retaining previously learned knowledge. Incremental learning is usually achieved through the following steps: load existing model parameters, fine-tune the model using new data, that is, continue training the model on new data, and update the model parameters. After updating the model, the validation set or test set can be used to evaluate and verify the model to ensure that the updated model still has good performance.

[0067] like Figure 3 As shown, the fluctuation response method includes: D1: based on the abnormal information data set, a statistical analysis algorithm is used to analyze and extract the power fluctuation characteristics to obtain a fluctuation event information data set, and the fluctuation event information data set includes a power fluctuation characteristic information set and a fluctuation event type; D2: based on the fluctuation event information data set and the response report information data set, a cosine similarity algorithm is used to match to obtain a corresponding response report.

[0068] It should be noted that the abnormal information data set should be cleaned, noise and irrelevant data should be removed, and the data should be standardized or normalized for subsequent analysis. Statistical analysis algorithms (such as mean, standard deviation, skewness, kurtosis, etc.) should be used to calculate the statistical characteristics of power fluctuations and extract the time series characteristics of power fluctuations, such as periodicity, trend, seasonality, etc. The time series can be segmented using technologies such as sliding windows to extract the characteristics of each time period and associate the extracted power fluctuation feature information set with the fluctuation event type. The fluctuation event type can be defined according to needs, such as voltage fluctuation, frequency fluctuation, power fluctuation, etc. A fluctuation event information data set is constructed, including the power fluctuation feature information set and the fluctuation event type, to prepare for the response to the alarm. The report information dataset should contain historical response reports and their corresponding fluctuation event information. The response reports are processed by text, such as word segmentation, stop word removal, stem extraction, etc. For the fluctuation event information dataset and the response report information dataset, feature vectors are constructed. For the fluctuation event information, the power fluctuation feature information set can be used as the feature vector. For the response report, the bag-of-words model can be used to convert the text into a feature vector. The cosine similarity algorithm is used to calculate the similarity of each record in the fluctuation event information dataset and the response report information dataset. According to the cosine similarity calculation result, the historical response report that is most similar to the current fluctuation event information is found, and the corresponding response report is returned as a reference. The cosine similarity algorithm formula is:

[0069] , where A and B are two eigenvectors, n is the dimension of the eigenvector, i=1, 2, 3, ..., n.

[0070] like Figure 3 As shown, the fluctuation response method also includes: E1: if the matching fails, based on the response report information data set, a retrieval technology is used to retrieve the response report with the power fluctuation characteristics and fluctuation event type that are most similar to the fluctuation event information data set, and the most similar response report is compared with the fluctuation event information data set to obtain a difference and similarity data set, E2: based on the same data set in the difference and similarity data set, the same report information in the most similar response report is determined, based on the difference data set in the difference and similarity data set, the key difference points are extracted through the 3sigma algorithm, based on the key difference points, it is matched again with the response report information data set to obtain abnormal report information, and the same report information and abnormal report information are combined to obtain a customized response report.

[0071] It should be noted that when matching power fluctuation characteristics and fluctuation event types with the existing response report information dataset, if the match fails (that is, no completely matching response report is found), then enter this step and use text retrieval to retrieve the response report information dataset that is most similar to the current power fluctuation characteristics and fluctuation event type. The retrieved most similar response report is compared with the current fluctuation event information dataset to identify the similarities and differences between the two and form a difference and similarity dataset. Based on the same dataset in the difference and similarity dataset, the same report information that matches the current fluctuation event information in the most similar response report is determined. These same report information may include some common response strategies, steps or resource allocation suggestions. For the difference dataset in the difference and similarity dataset, the 3sigma algorithm (or similar statistical methods) is used to identify key difference points. The 3sigma algorithm is usually used to identify points in the dataset that deviate far from the average value. These points may represent the key differences between the current fluctuation event and previous events. Based on the extracted key difference points, match it with the response report information dataset again to try to find abnormal report information related to these difference points. These abnormal report information may contain specific Special response strategies or precautions for determining the difference points, integrate the same report information determined in step E1 and the abnormal report information obtained in step E2 to form a customized response report, which contains both general response strategies and special suggestions for special circumstances of the current volatility event. The 3sigma algorithm is a statistical principle or rule based on the normal distribution. In statistics, the 3sigma principle or 3sigma rule is used to define the range of most values ​​in a data set. Specifically, for a normal distribution, approximately 68.27% of the data values ​​fall within the mean value (μ). Within one standard deviation (σ), approximately 95.45% of the data values ​​fall within two standard deviations of the mean, and approximately 99.73% of the data values ​​fall within three standard deviations of the mean. The specific steps in the fluctuation response method are: calculate the mean (μ) and standard deviation (σ) of the relevant features in the fluctuation event information dataset and the response report information dataset, identify those data points or features that differ from the mean by more than 3 standard deviations (μ±3σ), which may be key difference points. Based on these key difference points, use the cosine similarity algorithm to find matching response reports or information in the response report information dataset.

[0072] In the processing module, corresponding warning thresholds are set for the gluing equipment and the energy storage battery assembly based on the gluing equipment information data set, multiple energy storage battery groups and the monitoring data set, and the warning threshold setting method includes: Y1: Based on the historical monitoring data set, a statistical analysis method is used to obtain the power fluctuation data of the gluing equipment and the energy storage battery group under normal operation, so as to obtain the historical fluctuation data set, and then based on the historical fluctuation data set, a statistical distribution method is used to set the initial warning threshold for each power parameter of the gluing equipment and the energy storage battery group, so as to obtain the initial warning threshold set, Y2: Based on the initial warning threshold set, a test is performed to obtain a test result, and based on the test result, the final warning threshold information set is optimized to obtain the abnormal information data set in time.

[0073] It should be noted that the power fluctuation data of the glue coating equipment and energy storage battery pack under normal operating conditions are collected from the historical monitoring data set. These data should include various power parameters, such as voltage, current, power, etc. The collected historical data are cleaned to remove noise, outliers and incomplete data. The processed historical data are analyzed using statistical analysis methods (such as mean, standard deviation, median, etc.), and the statistical characteristics of each power parameter, such as mean value, standard deviation, etc., are calculated. According to the statistical distribution method (such as normal distribution, skewed distribution, etc.), the initial warning threshold is set based on the statistical characteristics of each power parameter, and the threshold is set by adding or subtracting one or more standard deviations from the mean value. For example, the lower threshold can be set to the mean value minus 2 standard deviations, and the upper threshold can be set to the mean value plus 2 standard deviations. The initial warning thresholds of all power parameters are combined. The initial warning threshold value set is used to test the new monitoring data to check whether there are too many false alarms or missed alarms. The test results are analyzed to identify the causes of false alarms and missed alarms. False alarms may be caused by the threshold being set too sensitively, while missed alarms may be caused by the threshold being set not sensitive enough. The warning threshold is adjusted according to the test results. If there are too many false alarms, the threshold can be appropriately relaxed; if there are too many missed alarms, the threshold can be tightened. The iterative optimization method can be used to gradually adjust the threshold until satisfactory performance is achieved. The optimized warning thresholds are combined into a set, namely the final warning threshold information set. The final warning threshold information set is used to perform real-time analysis on the new monitoring data. When a power parameter value in the monitoring data exceeds or falls below the corresponding warning threshold, it is marked as abnormal and added to the abnormal information data set.

[0074] The processing module is connected to a control terminal, and the collection module, processing module, output module, BMS and EMS are connected through a proprietary local area network. The method for establishing the proprietary local area network includes: T1: information acquisition, acquiring basic networking information of the collection module, processing module, output module, BMS and EMS, and the basic networking information includes device IP, T2: device connection, the control terminal automatically allocates communication channels to match the IP of the collection module, the IP of the processing module, the IP of the output module, the IP of the BMS and the IP of the EMS, T3: configuring the network, the control terminal configures corresponding network services and security measures for the communication channel to create a proprietary local area network.

[0075] It should be noted that network services include DNS services: providing domain name to IP address resolution services to facilitate communication between devices in the LAN through domain names, gateway services: as a bridge between the LAN and the external network, responsible for forwarding communication data and file sharing between devices in the LAN and the external network, security measures include firewalls: monitoring and controlling data flows in and out of the LAN to prevent unauthorized access and attacks, access control lists: defining which IP addresses or users have the right to access specific resources or services in the LAN, and encrypted communications: protecting transmitted data through encryption technology to prevent data from being stolen or tampered with during transmission.

[0076] The technical solution also requires a method for using the energy storage system for protecting the glue-coated PACK.

[0077] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. An energy storage system for glue-coated PACK, comprising a collection module, a processing module, an output module and a plurality of energy storage battery packs, Features: The collection module stores the glue coating process information data set, the PACK order information data set and the response report information data set, wherein the glue coating process information data set includes the glue coating equipment information data set, the collection module is provided with visual analysis software and obtains the PACK information data set by scanning the PACK based on the PACK order information data set, and the collection module obtains the monitoring data set by monitoring the glue coating equipment and the energy storage battery group through the power system monitoring equipment, and the monitoring data set includes the real-time monitoring data set and the historical monitoring data set; The processing module is provided with a matching model, and the matching model obtains the corresponding gluing process and the corresponding gluing equipment based on the matching of the PACK information data set, the PACK order information data set and the gluing process information data set in the collection module. The processing module obtains an adjustment report based on the corresponding gluing process and the corresponding gluing equipment through a strategy customization method, which is used to formulate a reasonable discharge strategy. The processing module obtains an abnormal information data set based on real-time analysis of the real-time monitoring data set, and finally obtains a response report based on the abnormal information data set through a fluctuation response method, which is used to respond to power fluctuations and ensure stable operation of the energy storage battery pack; The output module sends instructions to the BMS to adjust the discharge strategy of the energy storage battery group based on the adjustment report of the processing module, and sends instructions to the EMS to adjust the response strategy of the energy storage battery group based on the response report of the processing module; The energy storage battery pack is used to receive instructions from the BMS and EMS to provide stable and reliable power support for the corresponding glue coating process and the corresponding glue coating equipment; The strategy customization method includes: S1: The processing module determines the required parameters of the corresponding glue coating process based on the glue coating process information data set, then determines the performance parameters of the corresponding glue coating equipment based on the glue coating equipment information data set, and finally obtains the number of PACKs to be processed based on the PACK information data set; S2: Based on the required parameter calculation and the performance parameter calculation, the power demand of each PACK and the number of PACKs processed per unit time are obtained. The total processing time is calculated based on the number of PACKs to be processed and the number of PACKs processed per unit time. The total power demand is then calculated based on the number of PACKs to be processed and the power demand of each PACK. S3: Based on the total power demand and total processing time demand, combined with the real-time monitoring data set and required parameters, the corresponding discharge rate is calculated; based on the total processing time and the discharge rate, the corresponding discharge time is calculated; based on the real-time monitoring data set and the total power demand, the discharge sequence is adjusted and optimized, and finally a regulation report is summarized.

2. An energy storage system for glue-coated PACK according to claim 1, Features: The method for adjusting and optimizing comprises: A1: based on the realization status information set, the current status of each energy storage battery group is obtained, and multiple energy storage battery groups with the best status are selected to obtain an energy storage battery set; A2: Based on the total power demand and the discharge capacity of each battery group in the energy storage battery set, the energy storage battery group that meets the total power demand and has sufficient discharge capacity is matched, that is, the optimal energy storage battery set is obtained; A3: A polling algorithm is used to obtain the discharge order based on the best energy storage battery set.

3. An energy storage system for glue-coated PACK according to claim 1, Features: The method for creating the matching model includes: B1: Preprocess and integrate the PACK information dataset, PACK order information dataset, and glue coating process information dataset to obtain standard data, and divide the standard data into training set, validation set, and test set; B2: Extract features based on the training set to construct feature vectors, and select the random forest algorithm based on the training set to train and obtain the initial model; B3: Train and optimize the initial model to obtain a matching model, and update the matching model regularly.

4. An energy storage system for glue-coated PACK according to claim 3, Features: The training, optimization and updating methods of the matching model include: C1: Model evaluation: evaluate the trained initial model based on the validation set to obtain the evaluation results, which are used to analyze the performance of the model and identify existing problems and improvements. C2: Model optimization, optimize the model based on the evaluation results, and use the grid search method to tune the parameters; C3: Model verification, verifying the optimized model based on the test set to ensure the generalization ability of the model on unknown data; C4: Model update: regularly update the matching model and use incremental learning methods to update the model in real time.

5. The energy storage system for glue-coated PACK according to claim 1, Features: The volatility response methods include: D1: Based on the abnormal information data set, a statistical analysis algorithm is used to analyze and extract power fluctuation characteristics to obtain a fluctuation event information data set, wherein the fluctuation event information data set includes a power fluctuation characteristic information set and a fluctuation event type; D2: Based on the fluctuation event information dataset and the response report information dataset, the corresponding response report is obtained by matching the cosine similarity algorithm.

6. An energy storage system for glue-coated PACK according to claim 5, Features: The volatility response method also includes: E1: If the matching fails, the response report information dataset is retrieved using retrieval technology to obtain the response report with the most similar power fluctuation characteristics and fluctuation event types as the fluctuation event information dataset, and the most similar response report is compared with the fluctuation event information dataset to obtain the difference and similarity dataset; E2: Determine the same report information in the most similar response reports based on the same data set in the similarity and difference data set, extract the key difference points through the 3sigma algorithm based on the difference data set in the similarity and difference data set, match it with the response report information data set again based on the key difference points to obtain the abnormal report information, and combine the same report information and the abnormal report information to obtain a customized response report.

7. The energy storage system for glue-coated PACK according to claim 1, Features: In the processing module, corresponding warning thresholds are set for the glue coating equipment and the energy storage battery assembly based on the glue coating equipment information data set, multiple energy storage battery groups and the monitoring data set, and the warning threshold setting method includes: Y1: Based on the historical monitoring data set, the power fluctuation data of the glue coating equipment and the energy storage battery group under normal operation is obtained by using the statistical analysis method, so as to obtain the historical fluctuation data set. Then, based on the historical fluctuation data set, the initial warning threshold value is set for each power parameter of the glue coating equipment and the energy storage battery group by using the statistical distribution method, so as to obtain the initial warning threshold value set; Y2: Test the initial warning threshold set to obtain the test results, and optimize the test results to obtain the final warning threshold information set, so as to obtain the abnormal information data set in time.

8. The energy storage system for glue-coated PACK according to claim 1, Features: The processing module is connected to a control terminal, and the collection module, the processing module, the output module, the BMS and the EMS are connected via a dedicated local area network. The method for establishing the dedicated local area network includes: T1: Information acquisition, obtaining basic networking information of the collection module, processing module, output module, BMS and EMS, the basic networking information including device IP; T2: Equipment connection, the control terminal automatically allocates the communication channel to match the IP of the collection module, the IP of the processing module, the IP of the output module, the IP of the BMS, and the IP of the EMS; T3: Configure the network, control the terminal to configure the corresponding network services and security measures for the communication channel, and create a private LAN.

9. A working method of an energy storage system for a glue-coated PACK, Features: An energy storage system for a glue-coated PACK as described in any one of claims 1 to 8.

Citation Information

Patent Citations

  • A control method for an energy storage system and an energy storage system

    CN114768142B

  • Adaptive measurement and control system for charge-discharge process of integrated storage battery

    CN102176627A

  • Energy storage device and control method thereof

    CN114053632A

  • Scheduling control system and method of power grid energy storage system

    CN117394412A