摘要
为解决飞行动作识别规则的自动提取问题,提出一种基于改进粒子群优化算法的飞行动作规则提取方法。在对关键飞行参数特征量进行符号化的基础上,利用基于改进的动态惯性权重策略的离散二进制粒子群算法对符号化的各飞行参数特征量进行组合寻优,以找到能够完全表达飞行动作的识别规则。仿真实验表明,应用该方法得到的飞行动作识别规则简洁、有效,在实践中有良好的应用前景。
In order to resolve the auto-extraction problem of flight action rules acquisition, this paper proposes a flight action recognizing method based on modified Particle Swarm Optimization（PSO）. At the foundation of signed key flight parameters, it uses a discrete binary version of particle swarm algorithm based on modified dynamic inertia weight to find the optimization combination in the key flight parameters, and finds out the rule which can express the flight actions perfectly. Simulation result shows that the flight action rule found by the method is concise and available. The approach is quite promising.
出处
《计算机工程》
CAS
CSCD
北大核心
2008年第20期221-223,共3页
Computer Engineering
关键词
粒子群优化算法
惯性权重
飞行动作识别
规则提取
Particle Swarm Optimization（PSO）
inertia weight
flight action recognition
rules extraction
作者简介
王新亮（1980-），男，硕士研究生，主研方向：飞机健康状态监控.E-mail：wxl1226@yahoo．com．cn
倪世宏，教授、博士生导师.
