计算机应用
計算機應用
계산궤응용
COMPUTER APPLICATION
2014年
z2期
166-168,199
,共4页
分数阶PID%神经网络%反向传播算法%最优解%自学习
分數階PID%神經網絡%反嚮傳播算法%最優解%自學習
분수계PID%신경망락%반향전파산법%최우해%자학습
fractional order PID%neural network%Back Propagation ( BP) algorithm%optimal solution%self-learning
为解决分数阶PID控制器参数难于整定的问题，设计了一种基于神经网络的分数阶PID控制器。通过采用反向传播（ BP）神经网络的参数调节策略，可以实现一种五维参数自学习的PID控制器。将分数阶PID控制器数字化，通过BP算法调节神经网络突触权值，经过调整的神经网络输出作为分数阶PID控制器的参数。经过仿真验证，神经网络分数阶PID控制器比传统PID控制器精度提高6倍且控制更加稳定。
為解決分數階PID控製器參數難于整定的問題，設計瞭一種基于神經網絡的分數階PID控製器。通過採用反嚮傳播（ BP）神經網絡的參數調節策略，可以實現一種五維參數自學習的PID控製器。將分數階PID控製器數字化，通過BP算法調節神經網絡突觸權值，經過調整的神經網絡輸齣作為分數階PID控製器的參數。經過倣真驗證，神經網絡分數階PID控製器比傳統PID控製器精度提高6倍且控製更加穩定。
위해결분수계PID공제기삼수난우정정적문제，설계료일충기우신경망락적분수계PID공제기。통과채용반향전파（ BP）신경망락적삼수조절책략，가이실현일충오유삼수자학습적PID공제기。장분수계PID공제기수자화，통과BP산법조절신경망락돌촉권치，경과조정적신경망락수출작위분수계PID공제기적삼수。경과방진험증，신경망락분수계PID공제기비전통PID공제기정도제고6배차공제경가은정。
In order to solve the challenging problem of determing the parameters in fractional order PID controller, a fractional PID order controller based on artificial neural network was proposed. A self-learning PID controller with five-dimension parameters was realized by using parameter adjustment strategy of Back Propagation ( BP) neural network. After the fractional order PID controller was digitized and the synaptic weights were adjusted by using BP strategy, the adjusted outputs of the neural network were used as the parameters of the fractional order PID controller. A series of experiments verify that the artificial neural network fractional order PID controller can increase the accuracy by six times than the traditional PID order controller and is more stable.
