Research On Faults Recognition And Warning For Rotor Support System Of Steam Turbine Unit

With the large-capacity, high-parameter of steam turbine constantly put into operation, the unit operational reliability requirements are also increasing. Turbine support the important role played by the system after a failure and the impact brought about also attracted attention from all aspects.Among them, an important role to play Turbine support system and its effects after a failure brought about also attracted attention from all aspects.This paper carries out the following work with taking the rotor support system of steam turbine unit as the research object.(1) Based on the fault tree analysis and FMEA analysis, this paper did a deep research on the fault mechanism of support looseness、bearing axial vibration、structural resonance. Make the detailed reasons analysis on the main failure mode, then analyzed the faults and the maintenance measures, established the analyzing of FTA and FMEA for three kinds of faults mode. It formed a relatively complete system of knowledge.(2) Identify failure modes of rotor support system. Identify support systems corresponding monitor measuring points. Support system for signs of typical failure mode extraction and propose different kinds of signs describe the method of quantification. While establishing anomaly detection module and fault identification module. Multivariate state estimate technique was adopted to build steam turbine generator set abnormal detection model. By reconstructing operating data and testing residuals, abnormal information can be detected, so that the subsequent fault diagnosis procedure can be triggered. Evidence theory is taken to fuse symptoms so as to identify the fault mode.(3) Study of fault warning technology of rotor support system. Evaluation of the severity of the fault as the fault status indicators, and quantitative evaluation of individual fault on the basis of the evaluation unit equipment health warning. Fault warning signs quantization method and fuzzy evaluation matrix, the establishment of early warning model, which determine the support system failures severity rating, to obtain the support of early warning system failure, and the corresponding service level.