Automatic machine fault diagnosis based on wavelet transform and probabilistic neural networks

Abstract EN

Machinery parts always put imprint on the product during the production processing.

Industry in the developing country the discovery of defects depends on the human experience and spectrum analysis (SA).

Fast Fourier Transform (FFT) is basis for SA and uses to extract the frequency features which help the experts to identifying the causes of defected parts in machine.

In this paper presents a new technique to automatic fault diagnosis.

The proposed technique is constituted of two stages architecture: the first stage is analysis the product signals to extract the features by using wavelet transform (WT).

The second stage is devoted to the classification of defect from the features by using probabilistic neural network (PNN).

Naïve Bayesian algorithm and Bayesian net algorithm is taken for classification and compared.

The novelty of the proposed method resides in the ability not only with higher precision, but also with dimensionality reduction and higher speed than method of Fourier transform and mathematical statistics.