Resea Rch On Fault Diagnosis Of Gas Turbine Based On Neural Network

Heavy-duty gas turbine is one of the key equipment of gas power plant. The normal operation of the gas turbine related to the safety and economy of the entire gas power plant, and therefore the gas turbine condition monitoring and fault diagnosis is of great significance.In this paper, the operation principle and main structural components of gas turbine are introduced, and the common faults and its causes of gas turbine are described in detail. Based on this, the research contents mainly include the following three aspects:(1) This paper carried out a fault statistical analysis for the domestic gas turbine, and aim to provide a reference that can quickly locate the fault position, find the causes of the fault, and find a solution for the staff in the future in the similar fault.(2) This paper introduces a new neural network diagnosis method——The Self-Organizing Feature Map and related theory of it, and makes a detailed description on its use in Matlab.(3) At the end of this paper, an analysis based on the SOM method was made on the data that came from gas turbine test rig existed, and it is concluded that this method can be effectively applied to the gas turbine fault diagnosis.