反向蚁群算法在故障监控中的应用.pdfVIP

Proceedings of the 29th Chinese Control Conference July 29-31, 2010, Beijing, China 反向蚁群算法在故障监控中的应用* 马海平，阮谢永，金宝根 绍兴文理学院物理与电子信息系, 绍兴312000 E-mail:Mahp@usx. edu. cn 摘 要: 为提高飞机发动机故障监控的实时性, 将蚁群优化算法应用于故障诊断监控参数的优选问题中. 针对现有蚁 群算法进化速度慢的不足, 本文提出了一种反向蚁群优化算法, 该算法利用反向学习的加速性能, 将其运用到蚁群算法 信息素更新过程中来加快算法的进化速度, 改 算法优化搜索能力, 从而缩短计算时间. 同时该算法具有过程简单, 易 于实现的优点. 几种典型函数的测试结果以及监控参数的优选实验表明提出的新算法是有效的, 具有一定优越性. 关键词: 进化算法, 蚁群优化, 反向学习, 故障监控 Oppositional Ant Colony Optimization Algorithm and Its Application to * Fault Monitoring MA Haiping, RUAN Xieyong, JIN Baogen Department of Physics and Electrical Engineering, Shaoxing University, Shaoxing 312000, P.R.China E-mail: Mahp@usx. edu. cn Abstract: In order to improve the real time of aircraft engine fault monitoring, it applies ant colony optimization (ACO) to select feature parameters of fault monitoring. To tackle the slow nature of ACO, an oppositional ant colony optimization (OACO) is presented in this paper. Utilizing the acceleration performance of opposition-based learning (OBL), it employs OBL for pheromone updating to accelerate the evolutionary process, improve the searching capability, and shorten the computing time. Also it has some merit including simpleness and easy implement. Through benchmark functions and monitoring parameter selection problem, it demonstrates that the proposed algorithm is effective and superior. Key Words: Evolutionary Algorithms, Ant Colony Optimization, Opposition-Based Learning, Fault Monitoring 程中，通过这些初步研究已显示出ACO