Prediction Research On Failure Of Diesel Engine Based On STORM Framework With Big Data And Improved APPSO-BP Algorithm

At present,diesel engine,as a power energy equipment,plays a vital role in a wide range of fields,which greatly improves social and economic benefits.In order to reduce or even avoid the loss caused by its failure,the study of appropriate fault prediction method becomes the focus of attention.At first,this paper studies the status quo and development of diesel engine fault prediction technology,then introduces the empirical mode decomposition algorithm in theory,asynchronous particle swarm optimization(pso)algorithm and BP neural network algorithm,and illustrates the use of empirical mode decomposition algorithm from the principle of processing on the diesel engine cylinder head vibration signal and then use asynchronous particle swarm optimization(pso)algorithm and BP neural network algorithm for the feasibility of the diesel engine fault prediction.Secondly introduced the distributed system,a giant database HBase,and flow processing principle and application of the Storm,the framework and the big data and flow processing framework combined with a diesel engine fault prediction technology,constituted based on improved APPSO-diesel engine fault prediction algorithm of BP neural network algorithm.At last,through simulation experiment has set up a platform diesel engine fault prediction,verified by the experimental simulation of diesel engine based on APPSO-BP neural network algorithm is efficiency and prediction accuracy of fault prediction algorithm,at the same time confirmed the feasibility of the diesel engine fault prediction platform.Moreover,the comparison experiment proves that the combination of big data technology and flow treatment frame is more efficient and fast for diesel engine failure prediction.The main characteristic of this paper is to use the optimization of the intelligent prediction algorithm combined with big data technology,not only ensure the reliability of the data and store the mass,and the real-time data mining ways,on this basis to achieve real-time monitoring and fault prediction of diesel engine fault,and builds the framework based on large data Storm APPSO-BP neural network algorithm of diesel multiple failure prediction platform,through the simulation experiments confirmed the platform running accuracy,for diesel multiple fault forecast increased diesel multiple fault prediction method in the field of technology type.