摘要
利用BP网络,以全年运行的空气源热泵冷热水机组性能预测为目标,对半封闭往复式压缩机建立了性能预测模型。以网络的泛化能力作为衡量建模效果的标准,同时指出模型的建立过程就是网络最佳结构参数的搜索过程。计算实例显示,网络的预测效果与训练效果吻合较好,训练代价适中,所建模型可用于压缩机全年运行的性能预测与故障诊断中。
An artificial neural network that uses the error back-propagation training algorithm is proposed to predict the whole year performance of the compressor working in the air-source heat pump heater/chiller unit. The generalization ability of the network is used to judge if the model can be safely used. It is pointed out that the modeling process is optimizing the network architecture parameters. The modeling results indicate that the model can be used in the performance prediction and fault diagnosis of the compressor.
出处
《流体机械》
CSCD
北大核心
2005年第9期21-24,共4页
Fluid Machinery
基金
国家自然科学基金项目(50278021)
关键词
BP网络
空气源热泵冷热水机组
性能预测
back-propagation training algorithm
air-source heat pump heater/chiller unit
performance prediction
