论文部分内容阅读
针对压缩映射遗传算法(CMGA)操作效率太低,收敛至最优解迭代次数太多的问题,采用了近亲交叉回避策略改进压缩映射遗传算法,不但保证收敛到全局最优解,而且提高了算法的收敛速度和操作效率。为了能对具体被控对象的性能进行有选择性的控制,分析了ITAE积分性能指标作为目标函数的缺点,在目标函数中增加了超调量、控制量和上升时间等综合因素,得到了性能更好的目标函数,应用于改进压缩映射遗传算法的适应度函数,并把以上改进算法的模糊规则优化应用于地板采暖系统,与模糊控制、未改进的压缩映射遗传算法优化模糊控制进行比较,提高了系统的控制效率,简化了模糊控制器的设计难度。仿真结果证明该方法在地板采暖系统中的有效性。
Aiming at the problem that the operation efficiency of CMGA is too low and converges to the problem that the optimal solution has too many iterations, the near-relative cross-avoidance strategy is adopted to improve the compression mapping genetic algorithm, which not only guarantees convergence to the global optimal solution, but also improves the algorithm Convergence speed and operational efficiency. In order to control the performance of a specific controlled object selectively, the shortcoming of the ITAE integral performance index as an objective function is analyzed. In the objective function, the integrated factors such as overshoot, control amount and rise time are added to obtain the performance A better objective function is used to improve the fitness function of the compression mapping genetic algorithm. The fuzzy rule optimization of the above improved algorithm is applied to the floor heating system, compared with the fuzzy control and the uncomplicated compression mapping genetic algorithm to optimize the fuzzy control. Improve the control efficiency of the system, and simplify the design difficulty of the fuzzy controller. Simulation results show the effectiveness of this method in the floor heating system.