工业仪表与自动化装置
工業儀錶與自動化裝置
공업의표여자동화장치
INDUSTRIAL INSTRUMENTATION & AUTOMATION
2014年
2期
7-10
,共4页
PID参数%蛙跳算法%惯性权重%最优解
PID參數%蛙跳算法%慣性權重%最優解
PID삼수%와도산법%관성권중%최우해
PID parameters%SFLA(shuffled frog leaping algorithm)%inertia weight%optimum solution
针对非线性、大延迟、时变的控制系统,传统的PID控制效果不理想,为此提出用蛙跳( shuffled frog leaping algorithm,SFLA)仿生优化算法整定PID参数,但传统蛙跳算法易陷入局部最优,收敛速度慢,因此提出权重改进的蛙跳算法( weight improved shuffled frog leaping algorithm,WIS-FLA)。该算法引入线性递减惯性权重修正最差青蛙的更新策略,可以平衡算法的全局搜索和局部搜索。通过两个经典控制系统的仿真测试,结果表明,WISFLA算法可以平衡算法的全局搜索和局部搜索,比SFLA和PSO( particle swarm optimization)的寻优能力强,迭代次数少,更适合PID参数的整定优化。
針對非線性、大延遲、時變的控製繫統,傳統的PID控製效果不理想,為此提齣用蛙跳( shuffled frog leaping algorithm,SFLA)倣生優化算法整定PID參數,但傳統蛙跳算法易陷入跼部最優,收斂速度慢,因此提齣權重改進的蛙跳算法( weight improved shuffled frog leaping algorithm,WIS-FLA)。該算法引入線性遞減慣性權重脩正最差青蛙的更新策略,可以平衡算法的全跼搜索和跼部搜索。通過兩箇經典控製繫統的倣真測試,結果錶明,WISFLA算法可以平衡算法的全跼搜索和跼部搜索,比SFLA和PSO( particle swarm optimization)的尋優能力彊,迭代次數少,更適閤PID參數的整定優化。
침대비선성、대연지、시변적공제계통,전통적PID공제효과불이상,위차제출용와도( shuffled frog leaping algorithm,SFLA)방생우화산법정정PID삼수,단전통와도산법역함입국부최우,수렴속도만,인차제출권중개진적와도산법( weight improved shuffled frog leaping algorithm,WIS-FLA)。해산법인입선성체감관성권중수정최차청와적경신책략,가이평형산법적전국수색화국부수색。통과량개경전공제계통적방진측시,결과표명,WISFLA산법가이평형산법적전국수색화국부수색,비SFLA화PSO( particle swarm optimization)적심우능력강,질대차수소,경괄합PID삼수적정정우화。
For the nonlinear,large delay,time-varying control system,conventional PID control result is not satisfactory. So PID self-turning based on SFLA( shuffled frog leaping algorithm) is presented. But because of the problems of SFLA such as local optimality and slow convergence rate. Therefore we propose a weight improved shuffled frog leaping algorithm. In this algorithm,which introduce the linear decreasing adaptive inertia weight to correct the poor frog update strategy can balance the global search and local search. The simulation results of experiments on two classical control system show that compared with the SFLA and PSO ( particle swarm optimization ) , the WISFLA can balance the global search and local search,and its optimization ability would be stronger,the number of iterations less,more suitable for the tuning of the PID parameters optimization.
