Title :
Assessment of Rotor Degradation in Steam Turbine Using Support Vector Machine
Author :
Yan, Jihong ; Ma, Haitao ; Li, Wanzhao ; Zhu, Haiyi
Author_Institution :
Dept. of Ind. Eng., Harbin Inst. of Technol. Harbin, Harbin
Abstract :
Steam turbines are the major equipments in power industries, which play an important role in national economic production. Rotors are the critical component of steam turbines, the assessment of rotor degradation is of great significance to ensure the safe and economic operation of the power plants. In this paper, support vector machine (SVM), a new machine learning technique, is applied to assess rotor life loss severity. Comparing to traditional mechanism methods for rotor residual life evaluation, the SVM model has no limits of the dimension of inputs and less time is spent on computation. Furthermore, the method provides a feasible tool for online rotor condition monitoring and even life prediction based on future operation schedules. The methodology presented in this paper was validated using the data including various starting parameters and corresponding life loss values from Harbin Turbine Company, the evaluation accuracy shows the effectiveness of the method.
Keywords :
condition monitoring; learning (artificial intelligence); power engineering computing; power generation economics; rotors; steam turbines; support vector machines; Harbin Turbine Company; machine learning technique; online rotor condition monitoring; power plant economic operation; power plant safe operation; rotor degradation assessment; steam turbine; support vector machine; Condition monitoring; Degradation; Economic forecasting; Machine learning; Power generation; Power generation economics; Power industry; Production; Support vector machines; Turbines;
Conference_Titel :
Power and Energy Engineering Conference, 2009. APPEEC 2009. Asia-Pacific
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-2486-3
Electronic_ISBN :
978-1-4244-2487-0
DOI :
10.1109/APPEEC.2009.4918199