Wang Shaohong

Flue Gas Turbine Condition Trend Prediction Based on Improved Echo State Network

Fault prediction is the key technology for ensuring safe operation and scientific maintenance of large equipment. As the running of flue gas turbine has nonlinear characteristics, echo state network (ESN) was introduced to predict the condition trend of the turbine. Singular value decomposition was used to improve the linear regression algorithm of ESN, and the prediction workflow was given. Condition trend prediction results showed the effectiveness of the proposed method.

The Construction and Application of Remote Monitoring and Diagnosis Platform for Large Flue Gas Turbine Unit

Large flue gas turbine unit is the key equipment in Catalytic Cracking unit of oil-fining plant and it plays an important role in energy saving. As operating in variable condition and high temperature harsh environment, the fault rate of the unit is relatively high. Once faulty happens, enormous economic loss will be caused so it is very important to make condition monitoring and diagnosis. The remote monitoring and diagnosis technology is a new fault diagnosis mode combining with computer technology, communication technology and fault diagnosis technology. Making large flue gas turbine unit as research object, this paper introduces different modes of condition monitoring and diagnosis system, then elaborates overall structure design of the remote monitoring and diagnosis platform constructed, and analyses application of the platform for the unit in detail. The platform can take full advantage of technical support and data sharing to perform remote monitoring and fault diagnosis as well as prediction effectively, improve success rate of fault diagnosis for the unit greatly, and provide technical means to achieve predictive maintenance for large unit.

Research on state prediction of flue gas turbine based on elman neural network

In the light of the characteristics of Elman neural network model which can be approximate to the arbitrary non-linear function and its ability to reflect the dynamic characteristics of the system, this paper provides a state prediction model of flue gas turbine by applying Elman neural network and makes prediction of the overall vibration value. Compared to traditional static BP network prediction model, examples show that Elman neural network model has simple structure and wonderful dynamic characteristics. This model can accurately predict the state of flue gas turbine, with high convergence rate and precision. It has a good performance in non-linear time series prediction, indicating that this model is feasible in the state prediction of flue gas turbine.