Federal learning-based automobile production mold optimization method, system, equipment and medium

Abstract

The invention provides a federal learning-based automobile production mold optimization method and system, equipment and a medium, and particularly relates to the field of automobile intelligent manufacturing. The method comprises the following steps: acquiring an automobile production mold model encrypted and trained by each factory from a plurality of participants participating in federated learning; sequentially performing decryption processing and preprocessing on the encrypted mold model to obtain training data of each participant to form a training data set; combining the participants to obtain at least one participant combination, and selecting corresponding training data sets for training according to different participant combinations to obtain corresponding fusion models; performing quality evaluation on the fusion model, and selecting a corresponding fusion model for fusion optimization according to a quality evaluation result to obtain a federated learning model; and sending the federated learning model to a requester, and performing local training by the requester according to the federated learning model to update the mold model. According to the invention, federal learning is adopted to isolate data, data leakage is avoided, and data security can be ensured.

Description

technical field [0001] The invention relates to the field of automobile intelligent manufacturing, in particular to a federated learning-based automobile production mold optimization method, system, equipment and medium. Background technique [0002] Mold is a high-efficiency process equipment. In automobile production, the quality of mold directly affects the quality of automobiles. Among them, each manufacturer uses different parameters in the establishment of the mold, and the model parameters determine the stability and accuracy of the mold. Manufacturers need to continuously optimize and update their own molds in order to have the best performance. Therefore, there is a common need among multiple manufacturers, which is to continuously optimize the mold model. However, in actual automobile production, the mold model parameters of each manufacturer cannot be publicly shared, fearing the loss of core technology, which also affects the optimization process of the model. ...

AI Technical Summary

Problems solved by technology

[0003] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a method, system, equipment and medium for optimizing automobile production molds based on federated learning, which are used to solve the problems caused by the model itself when optimizing automobile production molds in the prior art. The core parameters cannot be shared externally to form a problem of data islands

Images