Research On Adaptive Stochastic Resonance Method For Wear Signal Detection Of End Mill

The improvement of the level of automation in the manufacturing industry has put forward new requirements for the self-diagnostic ability of CNC machine tools,and the end mill is one of the important tools in cutting processing.The wear will not only affect the dimensional accuracy and the quality of the work piece,but also cause the vibration shock generated by the machine tool to affect the machining accuracy of the machine tool,which will indirectly affect the actual machining efficiency and production cost.Therefore,the wear signal detection of end mill has become more and more important,and it has become one of the hot topics for scholars at home and abroad.The wear signal detection of end mill was studied in this paper.The acceleration signals during end mill cutting are collected and then extracted eigenvalues in time and frequency domains.An adaptive stochastic resonance model based on genetic algorithm(GA)and particle swarm optimization(PSO)was established.It is used to the wear signal detection of end mill.The main research contents of the paper are as follows:Firstly,the development process of end mill wear signal detection methods and the development and application of adaptive stochastic resonance are analyzed theoretically.Then,different adaptive stochastic resonance algorithms are introduced.Aiming at the limitations,the highly dependency of the system parameters and noise intensity,of the traditional stochastic resonance method,the parameters of the bi-stable system are optimized to improve the signal-to-noise ratio to achieve the best stochastic resonance.Secondly,the fitness function used in this paper is the output signal-to-noise ratio,contributing to the optimization of GA and PSO model.Then,the characteristic values are extracted of the acceleration signal during the end mill cutting,and the power spectrum and frequency spectrum characteristics of the end mill during different periods are analyzed.GA and PSO are used to implement the adaptive Stochastic resonance method by optimizing the parameters of the system.And the simulated experiments on the output signal and the output signal-to-noise ratio are put to verify.By the experiment of the measured fault signal of the end mill,both algorithms can detect the spindle rotation frequency and the cutter tooth passing frequency,indicating that thealgorithms are effective in end-mill fault signal diagnosis.Finally,The two algorithms are compared for their respective advantages and disadvantages.The result shows that the adaptive stochastic resonance method,based on GA or PSO proposed in this paper,can effectively detect the wear signal of end mill and open up a wider application prospect for the realization of end mill wear signal detection.