论文部分内容阅读
两层结构的相容控制框架可以解决复杂冲突多目标控制问题,但其需要克服的困难是多目标直接优化、第一层所选区间目标的实现和在线实时优化控制。采用多目标遗传进化算法解决了直接优化多目标的工具问题。为能实现第一层中所选择的控制目标,第二层中设计了不同于第一层中形式的指标,并加入选择函数以反映实际系统的需要。结合预测控制思想的特点,提出了在线迭代遗传算法,解决了在线优化速度问题。以预测控制动态过程中误差、能耗两个目标为例对相容预测算法进行了说明。
The two-level structure of compatible control framework can solve the problem of multi-objective control of complex conflicts. However, the difficulties to be overcome are multi-objective direct optimization, the realization of the target of the first level of selection and the online real-time optimization control. Multi-objective genetic evolution algorithm is used to solve the problem of direct optimization of multi-objective tools. In order to achieve the control objectives selected in the first tier, indicators different from those in the first tier are designed in the second tier and selection functions are added to reflect the needs of the actual system. Combined with the characteristics of predictive control, an online iterative genetic algorithm is proposed to solve the problem of online optimization speed. To predict and control the dynamic process of the two objectives of error, energy consumption as an example of compatible prediction algorithm is described.