Study On Pulverizing System Fault Diagnosis In Thermal Power Plant

As an important part of the thermal power system, the coal pulverizing system takes the responsibility to supply the boiler fuel. Pulverizing system runs in harsh working environment. It possesses complicate structure of system. In addition, coal feeder and mill continuous high load operation for a long time. These factors result in higher failure rates. It is necessary that we research coal pulverizing system fault diagnosis and realize automatization of fault diagnosis and forecast in order to meet these requirements to guarantee the pulverizing system safe and reliable operation, improve the safety and economy of thermal power system.This thesis mainly focus on medium speed mill pressurized direct-firing pulverizing system as the main research object. The coal pulverizing system is described in aspects of its composition, operational principle and main specifications. Through studying on the running characteristics of the coal pulverizing system and research of relevant data, the common faults in the coal pulverizing system reason and phenomenon are analyzed in this paper. It summarizes several typical characteristic of the coal pulverizing system fault signs and fault parameters. These lay a further development of the coal pulverizing system fault diagnosis and monitoring system.In this paper, improved nonlinear state estimation method of fault diagnosis, Neural network fault diagnosis method based on data mining and neural network fault diagnosis method based on data mining were investigated. The article describes in details the principle, process, results of the three methods and sums up out of their respective suitable uses.Finally, the paper uses the improved nonlinear state estimation algorithm as a forecasting method,using particle swarm optimization support vector machine algorithm for fault diagnosis methods.In this paper,the condition monitoring and fault diagnosing system for the coal pulverizing system has been designed, which based on Virtual Instrument technology. The system can warn and diagnose faults correctly. Whats more,it has a positive impact on the safe and economic operation performance of units.