摘要
针对制冷系统膨胀阀控制回路具有时滞、非线性的特点 ,提出了一种神经元单值预测控制。用一个神经元进行预测 ,并递推估计未来p步的输出值 ,用另一个神经元实现对下一步控制的优化。仿真与应用结果均表明该方法是可行的 ,且具有结构简单、运算量小、速度快的特点 ,控制效果远优于PID控制。
Aiming at expansion valve control loop in refrigeration system, a predictor is constructed using neuron to predict the future p step output, and a control optimization is made using neuron to optimize the next control volume. The simulation results show that this method is feasible, and it is characterized by its simple structure, small calculating amount and fast speed. The control results are better than that of PID control.
出处
《低温工程》
CAS
CSCD
北大核心
2002年第5期29-33,共5页
Cryogenics
