基于神经网络波浪补偿平台垂向位移的直接多步预报

作者：魏栋; 叶家玮; 陈远明; 张波

来源：2007两岸三地博士生NBIC学术论坛, 杭州, 2007-10-01.

　　BP algorithm is a typical supervised learning algorithm,but neural network cannot be trained on-line by it.For this reason,a new algorithm (TD-DBP),which wascomposed of temporal difference (TD) method and dynamic BP algorithm (DBP),was proposed to overcome the restriction.TD-DBP algorithm can make Elman network train on-line incrementally.The gradient descent momentum and adaptive learning rate TD-DBP algorithm can improve the training speed and stability of Elman network effectively.U sing the collected real time data,the modified TD-DBP algorithm was able to realize direct multi-step predictions for vertical displacement of wave compensating platform.

• 单位
  华南理工大学