Research On Noise Source Recognition Of Tractor Cab Based On Bias Analysis

Reducing the noise of the tractor cab can not only protect the driver's physical health and their mental health,but also have a very important significance in improving the technical level of tractor products.The premise of the tractor cabin noise reduction is to identify,separate and quantify the noise sources in the cabin.The quantified results will provide corresponding scientific basis and corrective direction for the noise control of the cab in the future.For the analysis of noise sources,a combination of multiple methods can be used to study,this method can improve the accuracy of noise source identification.Based on the above research ideas,this article conducts related research work.This thesis is based on the establishment of a multi-input and single-output sound source system model,while applying partial coherence theory and hierarchical diagnosis theory based on partial coherence theory to study the noise source of the tractor cab.Firstly,by analyzing the relationship between input and output,a multi-input single-output sound source under partial coherence theory is established.The conditional power spectrum function of partial coherence theory and the algorithm of partial coherence function are derived on the basis of constant coherence theory,and simulation examples are calculated on this basis.The results of theoretical analysis and simulation calculations show that the multi-input single-output model established with conditional power spectrum can effectively remove the coherence between the inputs and improve the accuracy of the main noise source analysis of the tractor cab.Secondly,applying the hierarchical diagnostic theory model,the conditional power spectrum and partial coherence function are used to construct the judgment matrix of the hierarchy analysis,and the judgment matrix constructed is compared with the judgment matrix constructed based on the constant coherence analysis.Using hierarchical diagnostic methods to study simulation examples,the results show that the use of partial coherent hierarchical diagnostic models can not only construct a more accurate and effective judgment matrix,but also can rank the contribution of noise sources more accurately and effectively after excluding the input coherence.Finally,the whole vehicle experiment was carried out on the sample vehicle of the tractor,the measured data of the vibration and noise of the sample vehicle were collected,and the main noise sources of the tractor cab were comprehensively analyzed using partial coherence theory and AHP.The vibration and noise of the muffler,air filter,engine,and tail components are used as inputs,and the noise near the ears is used as the output.A multi-input single-output sound source model of the tractor cab is established.This model can effectively eliminate the input.According to the conditional input spectrum and partial coherence function,the judgment matrix in the hierarchical diagnosis model is constructed,and the noise contribution of each input is ranked for the output.According to the results of partial coherence and hierarchical diagnosis,the main excitations of the noise source in the cab are analyzed and judged.In this paper,the sound source model of the multi-input single-output system of the tractor cab based on partial coherence is established,which effectively eliminates the coherence of each input sound source.The establishment of a hierarchical diagnostic model based on the conditional power spectrum can more accurately match each input to the noise source.Contribution degree is quantified and sorted,and according to the obtained sorting results,specific suggestions are made for subsequent noise reduction of the tractor cab,which provides direction for rectification work.The results of simulation and experiment show that compared with a single noise source analysis method in the study of noise source separation and quantification of complex mechanical systems,the combination of multiple methods can more comprehensively and accurately reflect the characteristics of noise sources.