PSO优化BP神经网络对露点间接蒸发冷却空调机组性能预测研究

Study on Performance Prediction of Dew Point Indirect Evaporative Cooling Air Conditioning Unit Based on BP Neural Network Optimized by PSO

针对传统机组在设计时的实际运行、优化设计、成本投入等问题,基于神经网络理论建模的研究方法,分别建立了BP神经网络和PSO-BP神经网络两种预测模型,通过实际工程的测试得到了相关数据,最后采取理论建模与实测数据相结合的研究手段,得出了PSO-BP神经网络更适合用于解决露点间接蒸发冷却空调机组的性能预测问题的结论,在同一时刻,BP神经网络预测的相对误差为9.6%时,PSO-BP神经网络预测的相对误差为2.21%,精度提高了7.39%。

Aiming at the actual operation of traditional unit when the design,optimization design,the problem such as costs,based on the theory of neural network modeling methods,respectively,set up the BP neural network and PSO-BP neural network two prediction model,the relevant data is obtained by actual engineering test,finally take theory modeling and measured data of the combination of research methods,it is concluded that PSO-BP neural network is more suitable to solve the performance prediction problem of dew point indirect evaporative cooling air conditioning units.At the same time,when the relative error of BP neural network prediction is 9.6%,the relative error of PSO-BP neural network prediction is 2.21%,and the accuracy is improved by 7.39%.