Optimization of SLHMS based on dynamic balance of mechanical and electronic spindle

Mechanical spindle is an important component in the machining process, and it often produce an oscillation in the process of high-speed operation, making its spindle deviate from its geometric center, so that it would seriously affect the processing accuracy in the industrial process, thus affecting the stability of the entire system. At present, some dynamic balancing methods are difficult to establish model and have low control precision. Aiming at the above problem, this paper applies the harmony search algorithm to establish the model of the amplitude of the mechanical spindle vibration to search the optimal solution, ultimately search for the vibration deviation of the A and B balance blocks in the mechanical spindle. The harmony search algorithm can significantly improve the search speed because of its simple structure, less adjustable parameters and easy implementation, but the convergence effect is not ideal. Based on the harmony search, the improved self-learning particle swarm (SLPSO) algorithm is used to improve its convergence rate, and to avoid falling into the local optimal. The experimental results verify the feasibility and compared with the actual average, vibration accuracy respectively improves 7.316%, 6.741% and 3.476%.