Pulverizing system fault diagnosis based on least square support vector machine

Least square support vector machine is an excellent algorithm which can be used to model and classify. If appropriate mapping functions and parameters are selected, the result should be better. An improved particle swarm optimization with changeable inertia parameter and velocity weight is present and then it is used to search better parameter to optimize support vector machine which are used to diagnose faults existed in coal powder producing process. Simulation results show that the improved PSO has higher search precision and global search ability and the faults diagnosis algorithm coupled PSO and LS-SVM has higher diagnosis accuracy rate. This diagnosis is reasonable and applicable.