Machine fault prediction method based on MFCC feature extraction

Abstract

The invention discloses a machine fault prediction method based on MFCC feature extraction, and belongs to machine fault prediction methods. The machine fault prediction method comprises the steps that the feature of a current acoustical signal of running of a machine is obtained through an acoustical sensor installed on the machine, the acoustical signal is preprocessed, and then Mel conversion is carried out on the preprocessed acoustical signal to obtain an MFCC feature vector of the acoustical signal; according to the obtained MFCC feature vector, prediction is carried out on the health condition of the machine, the specific clustering process is that a SVM conducts clustering on the MFCC feature extracted when the machine runs and stored sample data obtained when the machine runs normally, the clustering result is analyzed through a vote method, and then the machine fault is predicted. The machine fault prediction method based on the MFCC feature extraction has the advantages that the acoustical feature of the machine is extracted and converted into the Mel domain, then clustering analysis is carried out on the feature vector through the SVM, the health condition of the machine can be rapidly, accurately and easily predicted, operation is easy, prediction precision is high, the prediction speed is high, the anti-noise performance is good, and nonlinear, random and time-varying signals can be accurately predicted.

Description

technical field [0001] The invention relates to a machine failure prediction method, in particular to a machine failure prediction method based on MFCC feature extraction. Background technique [0002] Machine failure prediction refers to predicting the health status of the machine and whether there will be gradual faults in the future based on the current or historical running state of the machine. The accuracy of the prediction is an important guarantee for the daily maintenance, normal operation, and safe production of the machine. It is related to economic benefits and maintenance costs. [0003] Different from fault diagnosis, fault prediction can make judgments in advance on machines that may have gradual faults, which provides a prerequisite for planned repair and maintenance. The sound signal is very sensitive to most of the machines. For a machine in normal operation, the sound characteristics of each start, run, and end of the machine do not change much. But when...

AI Technical Summary

Problems solved by technology

Among them, curve fitting is only predicted by regression analysis of historical data. Although the principle is the simplest, the error is relatively large; the Gaussian mixture model quantifies the features through the probability density function, and counts the frequency of each quantified feature, which can realize machine failure. However, its stability and accuracy are greatly affected by the learning mechanism; the neural network has the advantages of super anti-noise performance, no need to establish a mathematical model with certain rules, and strong nonlinear mapping ability. Good application prospects, but there are also problems such as slow convergence speed and easy to fall into local optimal solutions; gray model and hidden Markov model have the advantages of high prediction accuracy and strong adaptability in short-term prediction, but their accuracy depends on The number of characteristic parameters, when the characteristic parameters are small, the prediction effect is average; SVM is a new type of machine learning method, which has the only global optimal solution and excellent machine learning ability, and can well solve small samples, nonlinear , high-dimensional and other issues, can not meet the randomness, time-varying, nonlinear, etc. of machine failure characteristics

Images