A probability uncertainty method of fault classification for steam turbine generator set based on Bayes and Holospectrum

With the rapid development of the machinery and the increasing complexity of the steam turbine generator set, it is a great challenge for the safe and reliable operation of the steam turbine generator set. The uncertainties of fault classification and complicated working conditions become important research fields of steam turbine generator. A probability method on the uncertainty reasoning of fault classification for steam turbine generator is proposed in this paper based on the 2D-holospectrum and Bayesian decision theory. Firstly, Bayesian decision theory is adopted for the preliminary fault estimation on actual risk loss by calculating the loss expectation of each decision. Then, the area ratio of overlap region in 2D-holospectrum and the evidence theory can give the probability of the fault. Framework and model of the uncertainty reasoning are also described in this paper. Finally, the model is verified by the experiment of the rotor vibration on test rig. The results show that the method proposed is feasible for reasoning under imperfect information condition.