Spindle Power Prediction and Feed Rate Optimization based on Impeller Milling

Wei Wang

doi:10.62051/ijmee.v4n3.09

Authors

Wei Wang

DOI:

https://doi.org/10.62051/ijmee.v4n3.09

Keywords:

Prediction Algorithm, PID Control, Feed Rate Optimization, Stable Cutting

Abstract

This study is dedicated to solving the control hysteresis problem caused by the slow response speed of traditional fixed-parameter machining and control systems, and investigates a method aimed at combining deep learning and PID controllers to achieve intelligent optimization of the feed rate for ternary impeller milling. A two-stage step-by-step prediction of the spindle power is used to take advantage of the LSTM prediction optimization solving algorithm's fast response to time series data processing, and the PID controller adjusts the feed multiplication rate by proportional, integral and differential control strategies according to the model output to achieve constant power stable cutting with pre-adjustment of feed multiplication rate for CNC milling process. The experimental results show that the intelligent optimization scheme of feed multiplication for ternary impeller milling based on deep learning and PID controller has high effect. Through continuous learning and intelligent adjustment, the machining efficiency is significantly improved, while the machining quality and stability are ensured.

References

[1] Meijer, A., Krebs, E., Wiederkehr, P., & Biermann, D. (2018). Prediction of feed-rate slowdowns in precise micromilling processes. Journal of Manufacturing and Materials Processing, 2(1), 19.

[2] Dai Wei: Spindle power simulation and feed rate optimisation method in cavity milling based on historical data [D]. Huazhong University of Science and Technology, 2019.

[3] Zhao, H., Wei, Z., & Hu, Y. (2022). Offline optimisation method of feed rate for multi-axis milling machining. Mechanical Design and Manufacturing, (07), 183-186.

[4] Fu, Z., Yang, W., Zhang, Y., & others. (2016). Optimisation of milling feed rate for complex surface milling based on cutting force prediction model. Chinese Science: Technology Science, 46(07), 722-730.

[5] Zhao, P., Hu, P., & others. (2021). Multi-objective feed rate optimization of three-axis rough milling based on artificial neural network. The International Journal of Advanced Manufacturing Technology, 114, 1323-1339.

[6] Zhou, J.-M. (2021). A five-axis roughing feed rate optimisation method based on spindle power cutting load constraints and feed axis kinematics constraints . Huazhong University of Science and Technology.

[7] Yin, R., & Wang, Z. (2021). Optimization of milling cutting parameters under carbon efficiency target. In Journal of Physics: Conference Series (Vol. 1888, No. 1, p. 012008). IOP Publishing.

[8] Yin, Y. (2022). Optimisation of roughing feed rate for three-axis milling based on spindle power model of machining process [Unpublished doctoral dissertation]. Huazhong University of Science and Technology.

[9] Akiyama, Y., Iwaki, M., Komagamine, Y., Minakuchi, S., & Kanazawa, M. (2023). Effect of Spindle Speed and Feed Rate on Surface Roughness and Milling Duration in the Fabrication of Milled Complete Dentures: An In Vitro Study. Applied Sciences, 13(24), 13338.