摘要
采用人工神经网络对冬季运行的空气源热泵冷热水机组特性进行了模拟 ,利用BP算法对网络的连接权值进行学习和调整 ,从而满足给定精度的要求。只要训练样本可靠而充分 。
The performance of the units operating in winter is simulated using the neural networks. In order to achieve high precision, the connection power of the networks is studied and adjusted using the back propagation training algorithm (BP algorithm). The simulated outcome can meet the demands of on line fault diagnosis if the training samples are reliable and abundant.
出处
《流体机械》
CSCD
2002年第5期59-61,24,共4页
Fluid Machinery
基金
哈尔滨工业大学科学研究基金项目 (HIT 2 0 0 0 2 6)
关键词
神经网络
空气源热泵冷热水机组
训练样本
BP算法
neural networks
air source heat pump heater/chiller unit
training sample
back propagation training algorithm
