Study On Fault Diagnosis For Diesel Engine Based On The RBF Neural Network Of Particle Swarm Optimization

Diesel engine has been widely used in such fields as mine power of manyequipments. Diesel engine possesses important and ship as status in national economy.The condition monitoring and fault diagnosis on diesel engine get more and morerecognition of researcher and academe .The spectrum analysis based on FFT, which hasbeen widely used in machine fault diagnosis, can't meet the request of modern machinesfault diagnosis along with the development of signal analysis and the appearance ofmany analysis methods applied to non-stationary.Artificial Neural Networks, a kind of large-scale parallel distributing system, ischaracterized by self-organizing, self-learning, self-adapting and non-linearity. Thosecharacteristics make it have bright prospect in settling complex non-linearity questions.Particle swarm optimization (PSO) is a kind of optimization algorithm based on thetheory of swarm intelligence. It instructs optimization searching by competition andcooperation of each particle in race swarm. As a new rising random global optimizationalgorithm, PSO depends on less empirical parameters, easy to control and has theoreticalparallelism and fast convergence speed, therefore, it has succeed in many fields in the tenyears of its development.In this paper, based on the in-depth analysis of the Radial Basis Function neuralnetwork and PSO theory, PSO is presented to train the Radial Basis Function neuralnetwork and it proves that PSO has powerful network generalization ability and strongeridentification ability. At last, the diesel engine turbocharging system is diagnosed bytrained neural network. The result which meets the fact are got.