Optimization of Milling Parameters and Prediction of Milling Force Based on Grey Method and BP Neural Network
In order to optimize the milling process parameters and predict the milling force, a five-factor and four-level orthogonal experiment is meticulously designed. The gray relational analysis method is employed to optimize and investigate five process parameters, while considering milling force and processing time as the constraining factors. The influence of five process parameters on milling force is summarized, and the fundamental principles for selecting process parameters with the aim of ensuring the processing efficiency are obtained. The experimental results demonstrate that there is a marginal increase in the correlation value by 0.03, accompanied by a reduction in milling force of 7.706 N. The construction of a robust milling force prediction model utilizing the BP neural network is ultimately achieved. Remarkably, the maximum discrepancy between the predicted and actual test values stands at an impressively low 0.0038, while concurrently exhibiting a commendable alignment between the predicted value curve and its corresponding test value curve.
Authors
- Xiaorong Wei (ORCID: https://orcid.org/0009-0006-9893-9308)
- Fei Li
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-21
- DOI
- https://doi.org/10.5281/zenodo.24422834
- Primary Topic
- Advanced machining processes and optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00