| The ship auxiliary boiler system is an important part of ship auxiliary machinery,which is used to provide ships with saturated steam for non-primary power purposes such as driving steam auxiliary machinery and supplying auxiliary heat sources.Its trouble-free operation is of great significance to the nornal navigation of ships.In this paper,the fault diagnosis method is researched on the three types of faults that are common in ship auxiliary boiler system,such as slight blockage of the flue of the auxiliary boiler,dirty oil supply line filter,and small leakage of the main boiler safety valve.In view of the problems of incomplete fault knowledge,small number of fault samples,and difficult evaluation of fault diagnosis system for ship auxiliary boiler system fault diagnosis,this paper adopts the research idea of the combination of simulation technology and fault diagnosis technology,Steam drum water level and other five,parameters,based on which the ship auxiliary boiler system normal working condition sample database and fault working condition sample database are established,and then the ship auxiliary boiler system fault diagnosis method is studied and discussed.First of all,this article uses the modular modeling method,centralized parameter analysis method and other modeling methods to establish the mathematical models of the three main parts of the furnace part,steam drum part,water cooling wall and downcomer part of the ship auxiliary boiler system.Based on the mathematical models of each part,the simulation model of each part was designed and implemented using MATLAB/Simulink simulation software.From the static and dynamic verification results of the model,the simulation model accurately simulates the static and dynamic operation process of the ship's auxiliary boiler system,and can be used as a data source for the normal working condition data sample database of the ship's auxiliary boiler system.Next,this article is based on the knowledge that the auxiliary boiler flue is slightly blocked,the oil supply line filter is dirty,and the boiler main safety valve has a small amount of leakage.Three types of faults were set and verified in the boiler system simulation model.It can be seen from the results of the fault simulation that the change trend of the dynamic change curve of the fault monitoring point of the model is basically the same as the description in the previous studies,so it can be used as the data source of the sample database of fault conditions of the ship's auxiliary boiler system.Then,based on the support vector machine algorithm and the ship auxiliary boiler system data sample database,a fault diagnosis model of the ship auxiliary boiler system is established,and the fault diagnosis model of the ship auxiliary boiler system with unoptimized parameters is tested and evaluated.It can be seen from the fault diagnosis results that the fault diagnosis model of the ship auxiliary boiler system without optimized parameters shows certain adaptability to the fault diagnosis of the three common faults of the ship auxiliary boiler system,but there is still much room for optimization.Finally,based on the grid search method and particle swarm optimization algorithm,the kernel parameter gamma and penalty factor C in the support vector machine fault diagnosis model are optimized,and the fault diagnosis model obtained by parameter optimization and the fault diagnosis of unoptimized parameters are diagnosed.The model is compared with the fault diagnosis effect.Finally,it is concluded that for the ship auxiliary boiler system,the auxiliary boiler flue is slightly blocked,the oil supply line filter is dirty,and the boiler main safety valve leaks a small amount of three common faults.The fault diagnosis method based on the support vector machine algorithm has It is quite feasible,and in the parameter optimization algorithm of the fault diagnosis model,the particle swarm optimization algorithm has a better optimization effect than the grid search method. |
PDF Full Text Request