PROBABILISTIC MODEL-BASED DEGRADATION DIAGNOSING OF THERMAL SYSTEM AND SIMULATION TEST

摘要：

This paper proposed a probabilistic model-based approach to diagnose the possible parameters deviations that cause energy system degradation. It is competent for differentiating the deviations that is usually indiscernible in conventional physical model-based analysis. Probabilistic model combines domain knowledge and statistical data. Its diagnostic output provides a probabilistic confidence level for optimum operation. Operators own experience can also contrast with the model output to improve operation availability. A prototype model is tested on a full-scope simulator to verify its practical availability.

关键词： Diagnosis degradation probabilistic model Bayesian networks thermal system

作者: LI-PING LI JIN MA NING ZHAO ZHENG ZHAO JI-ZHEN LIU

作者单位: School of control and engineering, North China Electric Power University, Baoding 071003, China School of Power and Energy, North China Electric Power University, Baoding 071003, China

会议类型: 国际会议

会议名称: 2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

会议地点: 大连

会议语种:英文

页码: 1483-1486

在线出版日期: 2006-08-13（万方平台首次上网日期，不代表论文的发表时间）