摘要
融合遗传算法的并行搜索结构和模拟退火的可控性概率突跳特性 ,构造出一类高效的混合优化策略 ,该策略适合于多种类型模型的参数估计和控制器参数整定。对典型类型问题的仿真结果验证了混合策略的有效性和初值鲁棒性 。
A class of effective and general hybrid optimization strategy is proposed which reasonably combines the parallel structure of genetic algorithm with the controllable jumping property of simulated annealing. Numerical simulation results on benchmarks demonstrate the effectiveness and robustness on initial states of the hybrid strategy, and the optimization performance are fairly better than those of simple genetic algorithm and classical methods.
出处
《控制与决策》
EI
CSCD
北大核心
2001年第5期530-534,共5页
Control and Decision
基金
国家自然科学基金项目 (60 0 74 0 12 )
国家 973基础研究项目 (G19980 2 0 310 )
关键词
遗传算法
混合策略
参数估计
控制器
参数整定
工业控制
Benchmarking
Computer simulation
Genetic algorithms
Mathematical models
Optimization
Parameter estimation
Robustness (control systems)
Simulated annealing
Tuning
