Application of vibration singularity analysis, stochastic tool wear, and GPR-MOPSO hybrid algorithm to monitor and optimise power consumption in high-speed milling

Power consumption in manufacturing direct affects production costs and the environment. Therefore, the process of evaluating and researching power consumption in the machining process is very important. During high-speed milling, the power consumption varie`s due to tool wear and radial deviation. Therefore, a new model power consumption optimization is proposed based on cutting mode factors taking into account tool wear and radial deviation. In the existing power consumption models, studies on the effects of radial deviation and tool wear have not been thoroughly investigated. Stochastic tool wears established during high-speed milling is established in combination with the cutting force analysis model and wavelet singularity vibration point analysis. The nonlinear processes due to stochastic tool wear and cutting edge geometry were considered in the model. To optimize power consumption and establish a model for the real-time prediction of power consumption, a new GPR–MOPSO hybrid algorithm was developed based on Gaussian process regression (GPR) and multi-objective particle swarm optimizations (MOPSO). In order to verify the feasibility proposed monitoring and optimization model, experimental processes high-speed milling have been established. Results showed that the presented improvement model will reduce power consumption by 20.38% compared with manufacturer manuals chosen process parameters.