基于人工神经网络的制冷系统工况模拟

针对人工神经网络技术在制冷空调系统中的仿真应用,本文建立了单回路制冷系统的性能仿真系统。通过实验模拟制冷系统在夏季的负荷变动情况,得到了用于神经网络模型训练的样本数据。对制冷系统进行多种神经网络结构的建模,并进行了神经网络中各种结构参数对模型精度影响的分析。利用训练好的双隐层神经网络模型,研究了空调机组性能的影响因素,包括压缩机频率、室内外温度等。模拟结果表明,机组EER随着压缩机频率增加先增加后减少,随着室内温度升高而增加,随着室外温度升高而减少。结果表明,人工神经网络方法是分析制冷机组性能的一种有效途径。

Simulation of Operating Conditions for Refrigeration System Based on Artificial Neural Network

According to the simulation application of artificial neural network technology in refrigeration and air conditioning system, This paper builds the performance simulation system of single loop refrigeration system. Through the experimental simulation of refrigeration system in changing in load in summer, sample data used for training the neural network model are obtained. Build a variety of neural network structure on the refrigeration system, and analysis of the influence of different structure parameters on the accuracy of the neural network model. Using the trained neural network model with two hidden layers, the refrigeration system was simulated in this paper. The factors affecting performance of air conditioning units, which include the frequency of the compressor, outdoor temperature and indoor temperature, were studied in this paper. Simulation results show that EER of the system first increases and then decreases with the increasing frequency of compressor, increases with the increasing indoor temperature, and decreases with the increasing outdoor temperature. The results show that using artificial neural network method to analyze the performance of refrigeration units is an effective way.