论文部分内容阅读
尝试用现代人工智能技术来改进现有的制冷系统仿真方法．首先，提炼出与制冷系统仿真结果的量化密切相关的定量参数，然后在已有的定性数值仿真模型的基础上，根据实验数据，采用人工神经网络（ＡＮＮ）方法对仿真模型中的定量参数进行辨识，识别出最佳的定量参数．这不仅有利于提高仿真精度，改善计算稳定性，而且降低了对仿真软件用户的技术要求，有利于仿真技术的实用化．对房间空调器稳态特性仿真的初步结果表明该方法效果良好．
Try to use modern artificial intelligence technology to improve the existing simulation of refrigeration systems. Firstly, the quantitative parameters closely related to the quantification of the refrigeration system simulation results are extracted. Based on the existing qualitative numerical simulation models and the artificial neural network (ANN) method, the quantitative parameters of the simulation model are analyzed based on the experimental data Identify and identify the best quantitative parameters. This not only helps to improve the simulation accuracy, improve the stability of the calculation, but also reduce the technical requirements of the user of the simulation software, which is conducive to the practical application of simulation technology. The preliminary results on the simulation of the steady-state characteristics of the room air conditioner show that the method works well.