CN-121998392-A - Island type power battery production data management method and system

CN121998392ACN 121998392 ACN121998392 ACN 121998392ACN-121998392-A

Abstract

The invention discloses a method and a system for managing island-type power battery production data, which relate to the technical field of battery manufacturing, wherein the method comprises the steps of adopting a distributed architecture based on cloud-edge cooperation to collect the production data of each process island in real time; the method comprises the steps of carrying out quality assessment on products by using an intelligent prediction model constructed based on machine learning, checking and screening the products, monitoring the running state of each process island device, recording the assembly information and time stamp of materials by using a block chain technology, carrying out information tracing on key materials, integrating production data of each process by using a unified data platform, establishing a virtual mapping model of a production line by using a digital twin technology, adjusting the state of the corresponding production process on the virtual mapping model by using the change of real-time data, and adjusting the current production beat. By the technical scheme, the quality consistency, the operation efficiency and the intelligent level of the power battery production are greatly improved as a whole.

Inventors

Liu changye
Lin Tingfeng
PAN ZHIXING
GAN MINGHUI
LIU WENMING
Lu Jincui

Assignees

上汽通用五菱汽车股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251217

Claims (11)

1. An island type power battery production data management method, characterized by comprising the following steps: the production data of each process island are collected in real time by adopting a distributed architecture based on cloud-edge cooperation through a unified data platform; According to the production data of each process island, performing quality evaluation on products of the process island in the previous process by utilizing an intelligent prediction model constructed based on machine learning, and checking and screening the products according to evaluation results, wherein the intelligent prediction model is trained by using historical quality data; monitoring the running state of each process island device according to the collected production data; According to the collected production data, the assembly information and the time stamp of the materials are recorded through a block chain technology, and information tracing is carried out on key materials in the power battery production assembly process; The method comprises the steps of integrating production data of all processes through a unified data platform, establishing a virtual mapping model of a production line by adopting a digital twin technology, adjusting the state of the corresponding production process on the virtual mapping model through the change of real-time data, and adjusting the current production beat.
2. The island-type power battery production data management method according to claim 1, further comprising: Scheduling a production plan of the power battery, managing the production quantity of the power battery, and calling the process parameters of the power battery product from a process parameter database; The production plan, the production quantity and the process parameters are respectively issued to controllers of all process islands of a production line, and power battery products of required types are produced according to the corresponding process parameters, specifically comprising the following steps: the method comprises the steps of preparing a battery model, order quantity, order priority and a delivery time stamp of a power battery product according to production requirements, and scheduling to obtain a production schedule of the power battery product; According to the battery model to be produced by the production plan, automatically calling the technological parameters of the battery products of the corresponding model from a technological parameter database; serializing the production plan, the production quantity and the process parameters in a preset format, and asynchronously transmitting the serialized production plan, the production quantity and the process parameters to controllers of corresponding process islands through message queues; The process island controls the actuator to produce according to the received battery model and process parameters; In the production process, according to the acquired real-time equipment state and material characteristics, a neural network model which is continuously updated by an online learning mechanism is adopted to dynamically and adaptively optimize the technological parameters, wherein the process comprises the steps of adjusting the welding current and the welding speed of the welding technological parameters, adjusting the charge and discharge parameters of a battery cell, and adjusting the torque and the pressure of a wire harness assembly process, and the neural network model is trained according to historical production data.
3. The island-type power battery production data management method according to claim 1, wherein the quality evaluation of the product of the island of the previous process is performed by using an intelligent prediction model constructed based on machine learning, and the product is checked and screened according to the evaluation result, comprising: Training an intelligent prediction model based on machine learning by utilizing historical quality data; Adopting a pre-verification mode, when a product finishes processing a process island of a previous process and enters a waiting buffer zone of a next process, calling all process data of the previous process, and adopting the intelligent prediction model to perform online quality evaluation on the product of the process island of the previous process; and determining whether the product is qualified or not according to the quality qualification probability obtained by evaluation, and screening out unqualified products.
4. The island-type power battery production data management method according to claim 1, wherein the information tracing of key materials in the power battery production and assembly process is performed, and the specific process comprises the following steps: providing RFID and two-dimensional code identification for key materials of the power battery; Acquiring RFID and two-dimensional code identification information of each assembled key material by adopting a block chain technology in the production and assembly process, and generating a unique traceability tag for each battery pack; and carrying out full life cycle recording on configuration identifiers and assembly association relations of the battery cells, the box bodies and the wire harnesses of the power battery based on a blockchain technology, and tracing material sources according to blockchain recording data.
5. The island power battery production data management method of claim 4, further comprising: Based on the block chain technology, raw information of manufacturer incoming material data and battery material, a timestamp and a provider digital signature are commonly uplink, and based on the block chain information, material information is subjected to certificate storage management and management maintenance.
6. The island-type power battery production data management method according to claim 1, wherein the island-type power battery production data management method is characterized in that the island-type power battery production data management method integrates production data of all processes through a unified data platform, a virtual mapping model of a production line is established by adopting a digital twin technology, and states of corresponding production processes on the virtual mapping model are adjusted through real-time data changes, and specifically comprises the following steps: Comprehensively analyzing the acquired production data, and establishing a virtual mapping model of the production line by adopting a digital twin technology; Calculating the deviation of the production period, the waiting time and the theoretical beats of each procedure based on a time sequence analysis method according to the equipment operation signals and the production beat data of each process island in the production data, and visually displaying the real-time position of the product in each process island according to the virtual mapping model so as to realize the beat analysis of the production line; According to the result of the beat analysis of the production line, the equipment utilization rate of each station is calculated, the key station which most affects the production efficiency is determined to be the production bottleneck station, and dynamic identification is carried out in the virtual mapping model.
7. The island-type power battery production data management method of claim 6, further comprising: Performing pattern recognition by using a machine learning algorithm according to machine vision detection data, sensor measurement data and online test data of each process island in the production data, and mapping the quality fluctuation trend of a product and the defect distribution rate of each process island in real time based on the virtual mapping model according to a recognition result to realize quality analysis; And identifying similar defect modes by adopting a clustering algorithm according to the quality analysis result, determining a defect generation procedure according to the similar defect identification result, positioning defect stations on the virtual mapping model, and connecting adjacent defect stations to construct a defect tracing path.
8. The island-type power battery production data management method according to claim 6, wherein the island-type power battery production data management method is characterized in that the island-type power battery production data management method integrates production data of each process through a unified data platform, a virtual mapping model of a production line is established by adopting a digital twin technology, states of corresponding production processes on the virtual mapping model are adjusted through real-time data changes, and actual production beats are adjusted, and specifically comprises the following steps: the method comprises the steps of acquiring real-time data of corresponding process islands through edge nodes deployed on each process island and uploading the real-time data to a cloud end, wherein the real-time data comprise the number of products, the equipment utilization rate, the product qualification rate and the product qualification rate to be flowed into a downstream in a buffer area; Generating a control instruction according to the real-time data through the calculation capability of the cloud, wherein the control instruction is generated according to the real-time data, and the control instruction is generated for shortening the interval of issuing a production task and shortening the production beat when the number of products in the corresponding downstream process island is lower than the lower limit of the products in the corresponding downstream process island, the equipment utilization rate of the corresponding downstream process island is lower than the target utilization rate, the product qualification rate of the first upstream process island in a preset time window is greater than or equal to a first quality threshold value, the product qualification rate to be flowed into the buffer zone of the first upstream process island is greater than a first preset qualification rate, and the control instruction for shortening the interval of issuing the production task and prolonging the production beat is generated when the number of products in the corresponding downstream process island exceeds the upper limit of the products in the corresponding downstream process island, and/or the product qualification rate of the first upstream process island in the time window is lower than a second quality threshold value and the product qualification rate to be flowed into the corresponding downstream process island is less than or equal to the first preset qualification rate; and adjusting the state of the corresponding production process on the virtual mapping model according to the control instruction.
9. The island-type power battery production data management method according to claim 1, wherein the real-time acquisition of production data of each process island specifically comprises: And collecting voltage and internal resistance data of an OCV battery cell detection process, welding current and temperature data of a laser welding process, torque and pressure data of a PACK front loading process, voltage and current curve data of a charge and discharge test process and assembly completion data of a PACK rear loading process in real time, and storing and inquiring by utilizing a time sequence database.
10. The island-type power battery production data management method according to claim 1, further comprising: Collecting the number of products in each process island, the length of a queue to be processed, the running time of equipment, the discharging speed and the material consumption speed, and calculating the real-time capacity and load rate of each process island in real time; when the number of products in the first process island is recognized to exceed a first threshold value and the productivity load rate of a second process island at the downstream is lower than a target range, issuing a production task matched with the backlog product model of the first process island to the second process island; When the length of a queue to be processed of a third process island in each process island exceeds the queue guard length and the operating time of equipment exceeds the upper limit of intervention, judging that the third process island is in an overload state; And when the duration that the discharging speed of the first process island is higher than the material consumption speed of the second process island is longer than the duration, generating a control instruction for reducing the discharging beat of the first process island or reducing the transmission speed of the material circulation system.
11. An island-type power battery production data management system, characterized by applying the island-type power battery production data management method according to any one of claims 1 to 10, comprising: the production data acquisition module is used for acquiring production data of each process island in real time by adopting a distributed architecture based on cloud edge cooperation through a unified data platform; The data evaluation screening module is used for evaluating the quality of the products of the process island in the previous working procedure by utilizing an intelligent prediction model constructed based on machine learning according to the production data of each process island, and checking and screening the products according to the evaluation result, wherein the intelligent prediction model is trained by using historical quality data; The operation state monitoring module is used for monitoring the operation state of each process island device according to the collected production data; The data tracing recording module is used for recording assembly information and time stamps of materials through a block chain technology according to the collected production data and tracing information of key materials in the power battery production assembly process; The data mapping adjustment module is used for integrating production data of all processes through a unified data platform, establishing a virtual mapping model of the production line by adopting a digital twin technology, adjusting the state of the corresponding production process on the virtual mapping model through the change of real-time data, and adjusting the current production beat.

Description

Island type power battery production data management method and system Technical Field The invention relates to the technical field of battery manufacturing, in particular to an island type power battery production data management method and an island type power battery production data management system. Background With the rapid development of new energy automobile industry, the power battery production faces the dual challenges of improving the manufacturing flexibility and the production efficiency. Conventional power cell production lines often employ linear modes of operation, and their associated Manufacturing Execution Systems (MES) often have significant limitations. Most of the existing MES systems can only realize partial data acquisition on a single production line, production planning is required to rely on manual maintenance and adjustment, and each procedure data are mutually isolated to form an information island, so that deep data analysis and flow optimization are difficult to support. Particularly in the island type production process, the data splitting problem is particularly remarkable, the module section and the PACK section production line operate independently, so that data interaction between systems is hindered, quality check of the front working procedure and the rear working procedure is difficult to realize, and data of different product systems cannot be effectively identified and related. Furthermore, conventional systems lack the ability to monitor the overall process of production in real time. Key information such as equipment running state, material circulation information, quality detection data and the like cannot be effectively integrated, a manager cannot accurately grasp the actual beat of a production line, and fine analysis and optimization of the production process cannot be performed. The situation directly restricts the improvement of the production flexibility, so that the production line is difficult to quickly respond to the demands of product change, process adjustment and the like. When quality abnormality occurs, a great deal of time is often required to be consumed for manual investigation, and accurate whole-course tracking cannot be realized because of the scattered traceability information of key materials. With the development of power battery production to multiple varieties and small batches, the traditional MES system can not meet the requirements of modern manufacturing industry on data driving and intelligent decision making. The island production process requires the flexibility of independent operation of each process island, and meanwhile, the overall synergistic high efficiency needs to be realized, which puts higher demands on the data integration and analysis capability of the MES system. The prior art can not realize the problems of data penetration, intelligent decision, credible tracing and real-time optimization of the whole production process of the island-type power battery. Disclosure of Invention In view of the above problems, the present invention provides a method for managing island-type power battery production data, comprising: the production data of each process island are collected in real time by adopting a distributed architecture based on cloud-edge cooperation through a unified data platform; According to the production data of each process island, performing quality evaluation on products of the process island in the previous process by utilizing an intelligent prediction model constructed based on machine learning, and checking and screening the products according to evaluation results, wherein the intelligent prediction model is trained by using historical quality data; monitoring the running state of each process island device according to the collected production data; According to the collected production data, the assembly information and the time stamp of the materials are recorded through a block chain technology, and information tracing is carried out on key materials in the power battery production assembly process; The method comprises the steps of integrating production data of all processes through a unified data platform, establishing a virtual mapping model of a production line by adopting a digital twin technology, adjusting the state of the corresponding production process on the virtual mapping model through the change of real-time data, and adjusting the current production beat. In the above technical solution, preferably, the island type power battery production data management method further includes: Scheduling a production plan of the power battery, managing the production quantity of the power battery, and calling the process parameters of the power battery product from a process parameter database; The production plan, the production quantity and the process parameters are respectively issued to controllers of all process islands of a production line, and power battery products of required types a