Research About Sound Source Localization For Condition Monitoring Of Mechatronic Systems

With the continuous development of modern science, more and more complex mechatronic equipment have entered our daily life,which brought a great convenience, but also posed a serious challenge:the more complex mechatronic equipment, the higher probability of failure. The traditional monitoring means of the mechatronic systems based on internal sensing information have the disadvantage of less information gathering, low efficiency, low intelligence, they can not meet the ever-changing mechatronic device. Therefore, a monitoring system based on integration of external information is imperative.As an important part of the audiovisual information,sound information is critical, the paper focuses on the key technology of sound source location in mechatronic systems monitoring, which is based on the above considerations.In this paper, the abnormal sound signal of mechatronic systems is the study object. As the complex noise across a wider frequency band, we propose a signal processing method combined the traditional Butterworth filter and Wavelet denoising technology, the results show that the method can give full play to the advantages of them, the purpose of signal noise reduction is achieved.In order to achieve positioning abnormal sound signal of mechatronic systems precisely, the paper treat the principle of time delay as research method. The method is divided into two parts: one part is time estimates, the other part is positioning.This paper estimates the time successfully with the generalized correlation theory as the theoretical basis. For the positioning part, we carry out a theoretical model with a specific array of microphones,simulation results show that the positioning error of the model is in 1 mm.According to the real-time requirement of mechatronic system monitoring,we put forward an improved endpoint detection method based on signal short- time energy,with the existing basis of the endpoint detection theory.This method overcomes the shortcomings of complexity and bad real-time performance. Experimental result shows that the new method can judge the starting point of the pending signal in the sound more accurately and more quickly.In the end, we build a test platform of sound source localization system include hardware and software, positioning the abnormal sounds in the typical mechatronic system(drilling and milling). Experimental results show that within the microphone array measuring range, the system can realize abnormal sound source localization within ±5 cm.So we can achieve the purpose of monitoring the mechatronic system status based on the sound as external sensing tool.