Research On Fault Feature Extraction Technology And System Development Of Turbine Unit

With the rapid development of the industry, the application of large turbine equipment becomes more and more extensive. The problem of fault diagnosis of turbine equipment drew individuals’increasing concern. State monitoring and fault diagnosis of turbine equipment are the focus in the field of engineering technology at home and abroad. At present, turbine equipment is developing in high speed and automation direction, so it is necessary to ensure its safety and reliability.Based on the turbine system of the offshore oil platform, this paper analyzes and summarizes the fault mechanism and characteristics of the turbine equipment. According to the characteristics of the vibration fault signal, the algorithm flow of empirical mode decomposition is offered. The basic principle of morphological filter is introduced, and the filtering step of the adaptive weighted combined generalized morphological filter is proposed. Based on the nonlinear and non-stationary characteristics of the vibration fault signal of turbine generator unit and the above two algorithms, the empirical mode decomposition method and the adaptive weighted combined generalized morphological filtering algorithm are proposed to solve the problem of vibration fault signal extraction.This paper describes the whole process of the combination algorithm and introduces the method of embedding the algorithm into WinCC. And then the paper introduces the basic concepts, structure and diagnostic process of the expert system and finally designs the vibration fault diagnosis system of turbine generator, combined with features of SIEMENS controller and the function of expert system. The design method and the implemented steps of the condition monitoring and fault diagnosis of turbine unit based on PLC control is created.The research on the vibration fault feature extraction algorithm is of heuristic significance for the design and optimization of the nonlinear and non-stationary signal feature extraction. The proposed method is also applied to the filtering of the contaminated image.