针对传统遗传算法收敛速度慢且容易未成熟收敛引起的核动力船舶设备故障诊断响应延迟、误诊、漏诊问题，提出一种基于信息熵的免疫遗传算法用于核动力船舶设备的故障诊断：利用已知船舶设备故障征兆集合，选用概率因果模型，引入信息熵免疫遗传算法，求解具有最大后验概率的故障集合。某船用核动力蒸汽发生器与液压泵故障仿真结果表明，基于信息熵免疫遗传算法优化的概率因果模型不受故障样本的限制，具有较好的通用性，且模型故障诊断精度较高、寻优速度快。本方法同样适用于其他领域的故障诊断问题。
Because the slow convergence rate and premature convergence of the traditional genetic algorithm, fault diagnosis of nuclear power equipment has been delayed response, misdiagnosed, missed diagnosed and so on. An immune genetic algorithm based on information entropy for fault diagnosis of nuclear power systems was proposed:By using the known nuclear power fault symptom sets, the probability causal model was selected, and the information entropy immune genetic algorithm was introduced to solve the fault sets which had the maximum a posteriori probability. The simulation results of the nuclear steam generator and hydraulic pump faults showed that the probabilistic causal model based on the information entropy immune genetic algorithm was not restricted by the fault samples. This method had good generality, high accuracy, and fast searching speed. the process can also be applied to fault diagnosis in other fields.
2017,39(5): 118-122 收稿日期：2016-11-09
DOI：10.3404/j.issn.1672-7619.2017.05.023
分类号：TL383
基金项目：国家科技重大专项资助项目（2013ZX06002001-003）
作者简介：刘锐(1985-),女,工程师,主要从事核动力设备故障诊断研究
参考文献：
[1] 冯长宝, 韩忠伟, 孙沫莉. 遗传算法和神经网络在船舶柴油机故障诊断中的应用[J]. 舰船科学技术, 2016, 38(6):40-42.
[2] 钱琴. 神经网络在船用柴油机性能诊断中的应用[J]. 舰船科学技术, 2016, 38(6):34-36.
[3] 段孟强, 袁灿. 基于入侵性野草算法的核动力装置故障诊断[J]. 原子能科学技术, 2015, 49(4):719-723.
[4] MO K, LEE S J, SEONG P H. A dynamic neural network aggregation model for transient diagnosis in nuclear power plants[J]. Progress in Nuclear Energy, 2007, 49(3):262-272.
[5] EVSUKOFF A, GENTIL S. Recurrent neuro-fuzzy system for fault detection and isolation in nuclear reactors[J]. Advanced Engineering Informatics, 2005, 19(1):55-66.
[6] MA J, JIANG J. Applications of fault detection and diagnosis methods in nuclear power plant:a review[J]. Progress in Nuclear Energy, 2011, 53(3):255-266.
[7] PENG Di, GENG Zhi-qiang, ZHU Qun-xiong. A multilogic probabilistic signed directed graph fault diagnosis approach based on Bayesian inference[J]. Industrial & Engineering Chemistry Research, 2014, 53(23):9792-9804.
[8] PENG Y, REGGIA J. A probabilistic causal model for diagnostic problem solving, part I:integrating symbolic causal inference with numeric probabilistic inference[J]. IEEE Trans SMC, 1987, 17(3):146-162.
[9] PENG Y, REGGIA J. A probabilistic causal model for diagnostic problem solving, partⅡ:Integrating symbolic causal inference with numeric probabilistic inference[J]. IEEE Trans SMC, 1987, 17(3):395-406.
[10] 周国强, 王雪青, 刘锐. 基于改进广义极值分布的核管道最大腐蚀深度预测[J]. 中南大学学报(自然科学版), 2013, 44(5):1926-1931.
[11] 傅平, 罗可. 基于信息熵的免疫遗传算法聚类分析[J]. 计算机工程, 2008, 34(6):227-229.
[12] 孔衍, 任鑫, 王川, 等. 基于改进遗传算法的核动力装置故障诊断研究[J]. 原子能科学技术, 2012, 46(11):1357-1361.
