摘要
针对具有非线性、不确定性的受控对象,提出一种兼顾神经元控制器和模糊控制器优点的模糊神经元非模型控制方法。在此控制系统中,模糊控制器产生神经元控制器的输入,由神经元的加权值来调整模糊PID控制器的参数,由神经元控制器产生控制信号对复杂对象进行控制,前置的模糊控制器用来过滤由对象的非线性和不确定性引起的误差大扰动,后置的神经元则在实现模糊PID控制器参数自调整的同时,以非模型控制的方式产生控制作用,从而有效地提高控制系统的鲁棒性、适应性和控制品质。
For a non-linear, uncertain controlled object, proposing a non -model control method of fuzzy neurons which considers neurons controller＇s advantages and fuzzy controller＇s advantages. A fuzzy controller generates inputs of neurons controller and the neurons＇ weight adjusts the parameters of fuzzy PID controller, and then neurons controller generates control signals to control the complex object. The pre-fuzzy controller is used to filter the large disturbance which is caused by controlled object＇s nonlinearity and uncertainty. When rear-neurons adjusts parameters of fuzzy PID controller, it generates control in the method of non -model control, so it effectively improves robustness, adaptability and quality control of the control system.
出处
《现代制造工程》
CSCD
北大核心
2010年第4期10-13,共4页
Modern Manufacturing Engineering
基金
国家自然科学基金资助项目(70471052)
