压缩机热力性能的准确计算，对于使用压缩机制冷空调装置的优化设计起到很关键的作用，而单纯的理论模型难以反映实际的复杂因素，影响计算精度．采用人工神经网络与传统理论模型相结合的方式，建立智能型的压缩机热力计算模型，利用人工神经网络的自学习和泛化功能改善压缩机容积效率和电效率的计算模型精度．神经网络采用多层前向网络（ＭＬＰ），网络训练采用同伦ＢＰ算法．对房间空调器用滚动转子式压缩机启动过程的输入功率变化，以及汽车空调器用变转速往复式压缩机的容积效率进行仿真，并与实验结果对照．结果表明，智能型压缩机模型很好地改善了传统计算模型的精度，而且适应能力更强 The accurate calculation of the compressor thermal performance plays a key role in the optimal design of the compressor refrigeration air conditioning unit. However, the simple theoretical model can hardly reflect the actual complex factors and affect the calculation accuracy. The artificial neural network model is combined with the traditional theoretical model to establish an intelligent computational model of compressor heat and to improve the computational model accuracy of compressor volumetric efficiency and electrical efficiency by using the self-learning and generalization function of artificial neural network. The neural network adopts multi-layer forward network (MLP) and the network training uses homotopic BP algorithm. The change of input power of the air conditioner with rolling rotor compressor and the volumetric efficiency of variable speed reciprocating compressor for automobile air conditioner are simulated and compared with the experimental results. The results show that the intelligent compressor model improves the accuracy of the traditional calculation model well and has better adaptability