基于BP神经网络的横流开式空气源换热塔性能预测

Performance Prediction of Cross Flow Open Air Source Temperature Heat Transfer Tower Based on BP Neural Network

摘要: 本文基于BP神经网络预测吸热工况下空气源换热塔的热性能，通过改变空气源换热塔的循环溶液流量、风量和进口溶液温度，记录不同工况下的测量参数，利用BP神经网络处理试验数据。网络采用三层结构，隐含层神经元个数为5个，以溶液流量、溶液进口温度、风量、干球温度和盐球温度为输入参数，空气源换热塔吸热效率为输出值。吸热效率预测值和实测值的相关系数、平均相对误差、均方根误差分别为0.995、1.3775%、6.178 × 10⁻³。结果表明，BP神经网络可以准确预测空气源换热塔吸热工况下的性能，对空气源换热塔热泵系统的运行和设计有重要意义。

Abstract: This paper predicts the thermal performance of air source heat exchange tower under endothermic condition on BP neural network, by changing the circulating solution flow rate, air volume and inlet solution temperature of the air source heat exchanger tower, the measured parameters under dif-ferent working conditions were recorded, and the test data were processed by BP neural network. The network adopted a three-layer structure, and the number of hidden layer neurons was 5. The solution flow rate, solution inlet temperature, air volume, dry bulb temperature and salt bulb tem-perature were taken as input parameters, and the heat absorption efficiency of the air source heat exchanger was taken as output value. The correlation coefficient, mean relative error and root mean square error of predicted and measured endothermic efficiency are respectively 0.995, 1.3775% and 6.178 × 10⁻³. The results show that BP neural network can accurately predict the performance of the air source heat exchanger under the condition of absorption, which is of great significance to the operation and design of the heat pump system of the air source heat exchanger.