Optimization of Surface Roughness for Titanium Alloy Based on Multi- strategy Fusion Snake Algorithm

Abstract

Titanium alloy has the characteristics of low thermal conductivity, small elastic modulus, and serious work hardening, which makes it difficult to predict the surface quality after high-speed milling. Surface quality is known to significantly impact the wear resistance, fatigue strength, and corrosion resistance of parts. To improve the service performance of titanium alloy parts, it is of great significance to optimize the milling parameters based on the improvement of surface quality. Therefore, this paper proposes a milling parameter optimization method based on the snake algorithm based on multi-strategy fusion. The surface roughness was used as the optimization goal to optimize the parameters. Firstly, the response surface method was used to establish a prediction model of titanium alloy milling surface roughness to realize the prediction of surface roughness and make it continuous. Then, the snake algorithm with multi-strategy fusion was proposed, which initialized the population based on the orthogonal matrix initialization strategy, so that the population individuals were more evenly distributed in space, increased the diversity of the population, improved the model of food quantity and temperature in the algorithm, optimized the change mechanism of food quantity and temperature in the original algorithm into a dynamic adaptive mechanism, accelerated the convergence speed, used the joint reverse strategy to select and generate individuals with higher fitness, and strengthened the ability of the algorithm to escape the local optimal solution. Experimental results on five benchmarks with multiple comparative optimization algorithms show that the MSSO algorithm has faster convergence speed and higher convergence accuracy. Finally, the multi-strategy snake algorithm was used to optimize the optimization objective equation, and the milling parameter experiment shows that the surface roughness of Ti64 is increased by 55.7 percent compared with that before optimization, and the surface roughness of the specimen optimized by the multi-strategy fusion snake algorithm is significantly reduced, the surface toolpath row spacing is reduced, and the average height of the texture is reduced. This method can reduce the optimization time and ensure better optimization results than the classical optimization algorithm.