Search

CN-121980901-A - Method, device, equipment and storage medium for adaptively optimizing lamination model parameters

CN121980901ACN 121980901 ACN121980901 ACN 121980901ACN-121980901-A

Abstract

The embodiment of the invention relates to the field of new energy battery lamination processes and discloses a lamination model parameter self-adaptive optimization method, a device, equipment and a storage medium, wherein the method comprises the steps of randomly generating a plurality of particles in a preset parameter space according to a lamination model of a battery lamination machine, and respectively representing a group of lamination model parameters; the method comprises the steps of respectively executing lamination running processes according to each group of lamination model parameters, obtaining corresponding fitness values of each group of lamination model parameters according to the offset of a swinging rod of a battery lamination machine, defining the distance and the direction of each iteration of each particle, obtaining updated lamination model parameters each iteration, calculating the updated fitness values, judging whether preset convergence conditions are met according to the updated fitness values, outputting the corresponding lamination model parameters if the preset convergence conditions are met, and continuing iteration if the preset convergence conditions are not met. The method can solve the problems of long iteration time and low optimization precision caused by polling with fixed step length in the existing lamination model parameter optimization method.

Inventors

  • HE DONGMING
  • WU ZHONGMING
  • ZHAO YONGSHENG

Assignees

  • 远景能源有限公司

Dates

Publication Date
20260505
Application Date
20251205

Claims (10)

  1. 1. The method for adaptively optimizing the lamination model parameters is characterized by comprising the following steps of: Randomly generating a plurality of particles in a preset parameter space according to a lamination model of a battery lamination machine, wherein each particle represents a group of lamination model parameters; respectively executing a lamination operation process of the battery lamination machine according to each group of lamination model parameters, and obtaining a corresponding fitness value of each group of lamination model parameters according to the offset of the swing rod of the battery lamination machine in the lamination operation process; Defining the distance and the direction of each iterative movement of each particle, obtaining updated lamination model parameters according to the distance and the direction of each iterative movement, and calculating to obtain updated fitness value; Judging whether a preset convergence condition is met according to the updated fitness value, outputting the lamination model parameters corresponding to the fitness value if the preset convergence condition is met, and continuing the iteration if the preset convergence condition is not met.
  2. 2. The lamination modeling parameter adaptive optimization method of claim 1, wherein the fitness value is inversely proportional to the square of the offset of the pendulum rod.
  3. 3. The method of claim 1, wherein defining the distance and direction each of the particles moves in each iteration comprises: Acquiring an individual optimal solution of each particle and a global optimal solution in all particles according to the fitness value calculated by the current iteration; and constructing a speed vector formula according to the individual optimal solution and the global optimal solution to express the distance and the direction of each particle moving every next iteration.
  4. 4. A method of adaptively optimizing lamination model parameters as in claim 3, wherein said velocity vector formula is: ; where i is the particle number, d is the particle dimension number, k is the number of iterations, Is the weight of the inertia, which is the weight of the inertia, Is a factor of the learning of the individual, Is a group learning factor; 、 is a random number within interval 0-1, Is the velocity vector of particle i in the d-th dimension in the kth iteration, Is the position vector of particle i in the d-th dimension in the kth iteration, Is the individual optimal solution of particle i in the d-th dimension in the kth iteration, Is the population optimal solution of the d-th dimension of the population in the kth iteration.
  5. 5. The method of claim 4, wherein the inertia weight decreases with increasing iteration number using a linear decreasing strategy.
  6. 6. The method for adaptively optimizing parameters of a lamination model according to claim 1, wherein the preset convergence condition is that a maximum number of iterations is reached or a target fitness value is reached.
  7. 7. A lamination model parameter adaptive optimization device, comprising: The parameter generation module is used for randomly generating a plurality of particles in a preset parameter space according to a lamination model of the battery lamination machine, wherein each particle represents a group of lamination model parameters; The fitness obtaining module is used for respectively executing the lamination operation process of the battery lamination machine according to each group of lamination model parameters and obtaining corresponding fitness values of each group of lamination model parameters according to the offset of the swing rod of the battery lamination machine in the lamination operation process; the fitness updating module is used for defining the distance and the direction of each iterative movement of each particle, obtaining updated lamination model parameters according to the distance and the direction of each iterative movement, and calculating to obtain updated fitness values; And the convergence judging module is used for judging whether a preset convergence condition is met according to the updated fitness value, outputting the lamination model parameter corresponding to the fitness value if the preset convergence condition is met, and continuing the iteration if the preset convergence condition is not met.
  8. 8. The device for adaptively optimizing parameters of a lamination model according to claim 7, wherein the fitness updating module is specifically configured to obtain an individual optimal solution of each particle and a global optimal solution of all particles according to the fitness value calculated in a current iteration, and construct a speed vector formula according to the individual optimal solution and the global optimal solution to represent a distance and a direction of each particle moving in a next iteration.
  9. 9. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the lamination model parameter adaptive optimization method of any one of claims 1-6.
  10. 10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the lamination model parameter adaptive optimization method of any one of claims 1 to 6.

Description

Method, device, equipment and storage medium for adaptively optimizing lamination model parameters Technical Field The embodiment of the invention relates to the field of new energy battery lamination processes, in particular to a lamination model parameter self-adaptive optimization method, device and equipment and a storage medium. Background The lamination model parameters of the lamination machine in the current market basically use visual measurement, and are judged according to the positions of the shafts, so that the judgment precision is poor. In addition, an adaptive algorithm is also provided, and the addition and subtraction iteration is performed by directly writing the lamination model parameters. The method is to directly replace parameters into a program to generate a cam curve of a movable diaphragm, and then to repeat a round of movement process of the equipment to obtain an optimal value of the deflection of the swing rod. Since the lamination machine parameters are comparatively many, for example, each parameter is polled according to the upper and lower limits of 1mm, if the polling is performed according to the accuracy of 0.2mm, each parameter needs to be polled 10 times, and 4 parameters directly influencing the curve in the lamination machine parameters are needed, therefore, if the iteration is performed according to the traditional polling mode, 10 times of 4 times of polling are caused, namely 10000 times of polling, and the action cycle of each round of movement from the laminating machine laminating shaft is 20S, if the iteration is performed according to the traditional polling mode, 55 hours are needed, and the time is extremely long. This scheme not only requires several days of time to poll to determine a more appropriate solution, but the final accuracy is only 0.2mm. Disclosure of Invention The embodiment of the invention aims to provide a method, a device, equipment and a storage medium for adaptively optimizing parameters of a lamination model, which can solve the problems of long iteration time and low optimization precision caused by polling by adopting a fixed step length in the existing method for optimizing parameters of the lamination model. In order to solve the technical problems, the embodiment of the invention provides a lamination model parameter self-adaptive optimization method, which comprises the steps of randomly generating a plurality of particles in a preset parameter space according to a lamination model of a battery lamination machine, wherein each particle represents a group of lamination model parameters, respectively executing a lamination operation process of the battery lamination machine according to each group of lamination model parameters, obtaining a corresponding fitness value of each group of lamination model parameters according to the offset of a swinging rod of the battery lamination machine in the lamination operation process, defining the distance and the direction of each particle in each iteration, obtaining updated lamination model parameters according to the distance and the direction of each iteration, calculating to obtain updated fitness values, judging whether preset convergence conditions are met according to the updated fitness values, outputting the lamination model parameters corresponding to the fitness values if the preset convergence conditions are met, and continuing the iteration if the preset convergence conditions are not met. In the embodiment of the invention, the fitness value is inversely proportional to the square of the offset of the swing rod. In the embodiment of the invention, the definition of the distance and the direction of each particle in each iteration comprises the steps of obtaining an individual optimal solution of each particle and a global optimal solution in all particles according to the fitness value calculated in the current iteration, and constructing a speed vector formula according to the individual optimal solution and the global optimal solution to express the distance and the direction of each particle in each next iteration. In an embodiment of the present invention, the velocity vector formula is expressed as: wherein i is the particle number, d is the particle dimension number, k is the number of iterations, Is the weight of the inertia, which is the weight of the inertia,Is a factor of the learning of the individual,Is a group learning factor;、 is a random number within interval 0-1, Is the velocity vector of particle i in the d-th dimension in the kth iteration,Is the position vector of particle i in the d-th dimension in the kth iteration,Is the individual optimal solution of particle i in the d-th dimension in the kth iteration,Is the population optimal solution of the d-th dimension of the population in the kth iteration. In the embodiment of the invention, the inertia weight adopts a linear decreasing strategy and decreases with the increase of the iteration times. In the embo