Fault Diagnosis Method Of Rolling Element Bearings Based On Multi-Source Information Fusion Bayesian Networks

The steam turbine is one of the main equipment in the thermal power plant, the control of the steam turbine runs through the whole process of power generation, its advantages and disadvantages of the performance directly impact on the electrical network operation’s safety, economy and stability.The familiar reheat steam turbine is regarded as a researchful object in this thesis. Based on the complete analysis about modeling and simulating technology of DEH,the modeling-simulating method of combining the differential equation-Oriented and Physical Object-Oriented was used, regarding to the shortage of traditional single method. In order to improving more easily, the modeling-simulating method of Physical Object-Oriented was used for failure prone control system, which made its structure clear and salability strong. The modeling-simulating method of differential equation-Oriented was used for the complexity about steam turbine, which could simplify the mode and get better control effect.Then based on the principle in the second paragraph, the Control system module library was established. It is convenient to establish the simulating model of DEH for the library and easy to research the system and fault diagnosis of DEH for the simulating model。On this basis, according to the fuzzy control theory, the established model of the steam turbine control system has been optimized, which leaded to choosing the fuzzy parameter self-tuning PID controller. Under the external disturbance, the controller’s dynamic characteristic is simulated in MATLAB/SIMULINK environment, which provides a research direction to improve the control quality of the steam turbine governing system.The research in this thesis has a further improving control quality of the steam turbine, providind reference for put forward and researching the design methods withoptimizing performance of the unit control system.