Research On Cab Noise Source Identification Of Construction Machinery Based On Wavelet Time-frequency Analysis

Noise has become a major pollution which seriously affects people’s production and living nowadays. Therefore, more and more attention is paid to the study of noise control. The noise source identification technology is the premise of achieving effective control and in-depth study on noise control, and plays a very important role in the overall work of noise control. About noise source identification research, many valued achievements have been obtained in world, which we can learn from. With the improving of manufacturing level and environmental protection requirements, the implementation of new national standards on noise limits has brought a new pressure to the entire domestic construction machinery industry. Construction machinery cab noise is a complex superposition of a variety of noise sources. Therefore, the location and characteristics of the main noise sources should be effective distinguished, noise control measures will be further made, so that the purpose of reducing cab noise can be achieved.This paper takes a certain bulldozer cab as an example, to indentify the noise sources of the cab noise. First, multiple working cases and multiple point tests of vibration and noise sources are carried out on the bulldozer. Through the analysis of structure-borne noise and air-borne noise in the bulldozer cab, the main noise sources and vibration sources affecting the cab noise are determined. It can provide theoretical and experimental basis for the bulldozer low noise technological innovation. Through the systematic analysis of the noise outside the cab, higher noise levels appeared at the vicinity of the engine and cooling fan.Wavelet transform is a good research tool for signal local analysis in both time domain and frequency domain, and can effectively extract feature information from signal. Wavelet packet analysis provides a more sophisticated method than wavelet analysis. So in this paper wavelet packet transform is employed to analyze the vibration signals from multiple points. The method is to obtain the eigenvector of signal energy by using the solution of three grades of the wavelet packets, and the several signals correlativity with the correlation coefficients of the signal energy distribution is obtained. The magnitude of the correlation coefficients reflects the degree of correlation between noise signals and vibration signals, embodies the influence level of the vibration on cab noise. Taking this basic research tool, the main noise sources are gradually identified by the correlation coefficients between noise and vibration signals. On the basis of all above analysis results, some improvement measures as sound insulation and vibration isolation are adopted. By analyzing the test results of each different stage and the final improvement stage, the following conclusions are drawn: the cab noise is mainly generated by the engine vibration, which enters into the cab through the vibration of the mechanical structure and the cab floor covering; the method of using wavelet analysis tools to indentify the cab noise sources is efficient and effective.