Parameter setting method, device, electronic device and storage medium

Abstract

The present application discloses a parameter setting method, which includes: obtaining historical processing data, which includes historical target parameters and various historical reference parameters of materials; preprocessing historical processing data to obtain sample data; Perform training to obtain a regression model; obtain preliminary processing data based on the regression model; adjust the preliminary target parameters to obtain intermediate target parameters, preliminary reference parameters and intermediate target parameters form intermediate processing data; based on historical processing data, intermediate processing data and enhanced Learning algorithm, building a reinforcement learning model; obtaining actual reference parameters; determining the actual target parameters that need to be set according to the actual reference parameters and the reinforcement learning model. The application builds a regression model to increase the amount of processed data, and trains the reinforcement learning model through the increased amount of data to improve the accuracy of the learning model. The application also provides a parameter setting device, an electronic device, and a storage medium.

Description

technical field [0001] The present application relates to the technical field of material processing, in particular to a parameter setting method, device, electronic device and storage medium. Background technique [0002] In the process of material processing, the processing result of the material is affected by various parameters in the process of material processing. For example, in the laser welding process, the welding quality is affected by factors such as temperature, material properties, and time. In the process of anodic dyeing materials, the dyeing result is affected by various factors such as dyeing time, concentration of the anode tank solution, temperature, and PH value. . At present, in order to ensure the processing results of materials, it is often necessary to manually adjust the parameters, but manual adjustment can only rely on experience, which is cumbersome and imprecise, and the workload is heavy. For example, when using the anode tank to dye the mate...

AI Technical Summary

Problems solved by technology

At present, in order to ensure the processing results of materials, it is often necessary to manually adjust the parameters, but manual adjustment can only rely on experience, which is cumbersome and imprecise, and the workload is heavy

For example, when using the anode tank to dye the material, if the color measurement result of the dyed material is unqualified, it is necessary to determine the material according to the dyeing degree of the material and the processing parameters of the anode tank, such as temperature, PH value, concentration, etc. Secondary dyeing time; the secondary dyeing time of existing materials is mostly set manually, and it needs many tests and adjustments to ensure that the set dyeing time can make the color of the dyed material qualified

Therefore, manual debugging determines the processing parameters, resulting in low adjustment efficiency, and when the processing parameters are determined based on manual experience, the reproducibility is poor and labor costs are wasted

Images