Research Of Small Air-cooled Diesel In Fault Diagnosis System

Nowadays industry level is highly developed, using machine is applied to engineering machine,electric power machine and the special type of shipping. On the one hand,it highly improve product quality and labor productivity;On the other hand,if one part of it is fault,it must be dangerous in people security.The process of machinery fault diagnosis includes the acquisition of information and extracting feature and recognizing conditions of which feature extraction and condition identification are the priority. EMD is a self-adaptive time frequency analysis method that is applicable to non-linear and non-stationary processes perfectly thus EMD has a high signal-noise ratio.According to the non-stationary vibration signal characteristics of mini diesel engine. EMD and AR method is introduced into mini diesel engine fault diagnosis. Mahalanobis distance have better generalization and guarantee the local optimal solution is exactly the global optimal solution. The experimental results demonstrate the proposed diagnosis approach in which EMD,AR and Mahalanobis distance are combined is effective.The outline of the work is as follows:1. By sound collection system,it can collect sound data in 100% efficiency,80% efficiency or 75% efficiency of the mini diesel engine.In the lab it simulate six faults,which include the abnormal of the drain tap,the low of the in gas,the obstruction of the oil-tube and so on.2. The concept of IMFs energy entropy is proposed. The energy of sound signals in different frequency bands will change when faults occur, thus one or more kinds of energy variation in frequency components indicate fault occurrence.EMD method and AR method are used to analysis sound signals,which is collected the mini diesel engine in the lab. More kinds of energy variation in frequency components is put up by AR method.3.By the Mahalanobis distance it can be used to put up to classify the result of EMD method and AR method analysis,which it carry out to be intelligent diagnoses in the the sound signal of the mini diesel engine.