A modal parameter identification method of machine tools based on particle swarm optimization

The dynamic characteristics of the computerized numerical control (CNC) machine tool directly affect its machining quality, and it is necessary to carry out the working mode analysis of the CNC machine tool. The traditional manual analysis and identification process is complicated and inefficient. In the industrial environment of big data, how to use the working modal analysis method to quickly and accurately obtain the dynamic characteristic parameters of the machine tool processing from these data has become the research difficulty at the current stage. This paper proposes an optimized particle swarm optimization algorithm to solve this problem. Based on the working modal analysis theory, the semi-self-power spectrum of the output signal can replace the frequency response function for modal parameter identification. The optimized semi-self-power spectrum signal is used as the objective function of the algorithm, and the ability of the algorithm to preprocess the data is optimized, so that the improved algorithm can automatically analyze the structural mode of the machine tool during processing. Comparing the experimental results, it is found that the natural frequency identification error of the cantilever beam is less than 1%, and the natural frequency identification error of the CNC milling machine is not more than 7%. The results show that the particle swarm optimization algorithm based on modal analysis theory can be applied to the automatic analysis of modal parameters under machine tool operating conditions, and it is efficient and accurate.