Study On Vibration Fault Diagnosis Of Hydroelectric Generating Set Based On Stochastic Resonance And Multidimensional Permutation Entropy

With the vigorous development of hydropower cause in our country,the proportion of hydroelectric power and capacity of the unit are increasing.If the unit is in the event of accident,which not only affect the safety of hydropower station or power plant itself,but also will have a significant impact to the stability of the operation of power grids.In hydroelectric generating set,about 80% of the failures are reflected in the vibration signal,therefore,it is particularly important to carry out the research on vibration fault diagnosis of hydroelectric generating set.Therefore,in this paper,the theory of stochastic resonance(SR)and multidimensional permutation entropy(MPE)is introduced into fault diagnosis of vibration of hydroelectric generating set.And the method of vibration fault diagnosis of hydroelectric generating set based on stochastic resonance and multidimensional permutation entropy is proposed,which provides a new train of thought and method for vibration fault diagnosis of hydroelectric generating set.This paper first discusses the background,significance and the purpose of hydroelectric generating set vibration fault diagnosis research,and introduces the domestic and foreign research present situation,development trend and existing problems of the vibration fault diagnosis about hydroelectric generating set.Secondly,in view of the defects that traditional de-noising method is easy to damage the useful components,the paper introduced the theory of stochastic resonance and the method of signal de-noising based on stochastic resonance is proposed,the simulation results show the superiority of the method,in addition,this paper studies the main factors affecting the stochastic resonance output,which is the key to achieve the best de-noising effect.Again,on the base of the permutation entropy principle,this paper introduces the determination of two important parameters in the permutation entropy method,and the numerical verification is carried out;In view of the capacity limitations that the permutation entropy extracts the signal feature,the feature extraction based on multidimensional permutation entropy is proposed,and the simulation results show the superiority of the method.Then,as to the problem of particle swarm algorithm(PSO)is easy to fall into local optimum,the particle swarm algorithm is improved,and the improved particle swarm optimization algorithm is used to optimize the parameters of support vector machine,The fault diagnosis model is established which based on improved particle swarm optimization and support vector machine.At the same time,the simulation results show that the model of based on improved particle swarm optimization algorithm and support vector machine is better than the genetic algorithm and support vector machine.Finally,the actual monitoring data of hydropower station is analyzed according to the above theory and method,the simulation results show that the method which can more accurately diagnose the fault type of the unit,consistent with the actual fault of the unit,has high diagnostic accuracy.