The Experimental Study Of TC4 Titanium Alloy Machinability Base On Heat Treatment Process

TC4 titanium alloy has excellent performance, such as mechanical, physical and chemical properties, which is widely used in Aerospace, Aviation, Ships, Nuclear Power and other industrial areas. However, the machinability of the alloy is poor. It can produce great unfavorable influence into the machined surface integrity of parts.In order to get three different type of TC4 titanium alloy microstructure, the paper uses three different heat treatment methods. The dry machining performance test is respectively carried out under the same conditions to improve the cutting performance of titanium. The trial program of TC4 titanium alloy cutting test uses orthogonal design method which based on the three typical microstructure. The research of cutting force and surface roughness relied on cutting depth ap, feed rate f, cutting speed vc. The multiple-variable linear regression equations of the cutting force and Surface roughness are built in term of three parameters using the MINITAB software and an accurate empirical formula of TC4 cutting force was established. The heat treatment’s technical laws of cutting force of TC4 titanium alloy are further undertaken, which provide a theoretical basis for improving cutting performance and optimizing turning parameter.In addition, this paper uses BP neural network back propagation to predict the cutting model. It finds that individual samples can not be correctly identified and the convergence is slowly. It is also easy to fall into local minimum. The genetic algorithm is introduced into the BP neural network model training process to optimized. The simulation results show that: the BP neural network based on genetic algorithm can overcome these shortcomings, and can correctly identify the signs of deviation from the training samples. It is also can improve the model prediction accuracy and speed.