Friction noise prediction method based on Bayesian network

Abstract

The invention provides a friction noise prediction method based on a Bayesian network. The friction noise prediction method comprises the following steps of 1, obtaining friction noise and a data sample which is associated with the friction noise; 2, performing discretization processing on the data sample obtained in the step 1; 3, performing Bayesian network learning, and establishing a Bayesian network structure chart; 4, updating conditional probability, under a corresponding father node, of each node in the Bayesian network structure chart according to a data sample set; and 5, performing prediction on the frequency and intensity of the friction noise according to the conditional probability obtained in the step 4. The friction noise prediction based on the Bayesian network, by virtue of organic combination of a directed acyclic graph and a probability theory, intuitively expresses joint probability among random variables, so that the friction noise prediction can be carried out by only processing the measured data without considering complex appearance of a sound source and various occurrence mechanisms; and therefore, the friction noise prediction method has relatively high prediction accuracy and reliability and is quite simple, convenient and quick.

Description

technical field [0001] The invention belongs to the field of machinery, and in particular relates to a method for predicting friction noise. Background technique [0002] Scholars at home and abroad have done a lot of research on friction noise and put forward various theories. For the prediction method of friction noise, there are finite element-based brake squeal modeling and analysis methods, which include the frequency domain complex eigenvalue method and the time domain transient dynamics analysis method. The frequency domain complex eigenvalue method cannot clearly point out The mechanism of brake squeal, and the assumption of linearization lacks consideration of time-varying loads and material properties and other unsteady characteristics, often resulting in "over-prediction" and "under-prediction" of unstable frequencies, which limits its Prediction accuracy and reliability, while the disadvantages of the time-domain transient analysis method are that it takes too l...

AI Technical Summary

Problems solved by technology

For the prediction method of friction noise, there are finite element-based brake squeal modeling and analysis methods, which include the frequency domain complex eigenvalue method and the time domain transient dynamics analysis method. The frequency domain complex eigenvalue method cannot clearly point out The mechanism of brake squeal, and the assumption of linearization lacks consideration of unsteady characteristics such as time-varying loads and material properties, and the "over-prediction" and "under-prediction" of unstable frequencies often occur, which limits its Prediction accuracy and reliability, and the disadvantage of time domain transient analysis method is that it takes too long to calculate and takes up a lot of disk space, and the data is difficult to be directly applied to the design

Images