摘要
针对电力系统模型及运行参数不确定性 ,在用户电力控制器中引入神经网络控制策略 ,应用神经网络设计了线性控制器 ,所提出的神经网络采用改进BP算法 ,改进BP算法比传统BP算法多了两个变量 ;学习速率和动量因子。将计算得的补偿电流经过神经网络训练实现预定目标 ,由MATLAB仿真结果可观察到经神经网格控制的用户电力控制器可实现系统侧的控制目标。并且其控制算法更为简单、有效。
This paper presents the artificial neural network (ANN) used in a custom power controller.The neural network to be used in a custom power controller is trained by an improved BP algorithm.The proposed BP algorithm is different from the conventional BP algorithm in the two more varible factors,i.e.a learning rate factor and a momentum factor.The compensating current is trained to reach the target.The emulator in MATLAB program shows that the ANN controlled controller can achieve the controlling target at the system side,and that the improved algorithm is simpler but more effective than the conventional one.
出处
《贵州工业大学学报（自然科学版）》
CAS
2001年第1期71-75,共5页
Journal of Guizhou University of Technology(Natural Science Edition)
基金
贵州省科委基金!(96 ) 310 6号
贵州省教委 99年科研基金
