摘要
送风量的精准预测是实现变风量空调蓄冷量精确控制的重要环节。本文根据变风量空调送风量的影响参数,基于深度置信神经网络方法,建立变风量空调送风量的预测模型。将该模型的预测结果同BP、Elman和模糊神经网络的预测结果进行对比,结果表明,深度置信神经网络的预测精度最高,平均绝对相对误差、均方根相对误差和决定系数分别为1.555%、0.789%和0.9975,由此说明本文建立的模型能够精确有效地预测变风量空调的送风量。
The accurate prediction of the supply air volume is an important part to realize the precise control of the cold storage volume of the variable air volume air-conditioning.In this paper,according to the influence parameters of the variable air volume air-conditioning supply air volume,based on the deep-confidence neural network method,a prediction model of the variable air volume air-conditioning supply air volume is established.By comparing the prediction results of this model with those of BP,Elman,and fuzzy neural networks,the results show that the deep-confidence neural network has the highest prediction accuracy,and the average absolute relative error,root mean square relative error,and determination coefficient are 1.555%,0.789%and 0.9975,respectively,showing that the model established in this paper can accurately and effectively predict the supply air volume of the variable air volume air-conditioning.
作者
雷蕾
王宁
郑皓
薛雨
LEI Lei;WANG Ning;ZHENG Hao;XUE Yu(Guilin University of Electronic Technology,Guilin 541004,China;Dalian University of Technology,Dalian 116024,China)
出处
《流体机械》
CSCD
北大核心
2021年第3期85-90,共6页
Fluid Machinery
基金
国家自然科学基金项目(51708146)
广西科技基地和人才专项(桂科AD18281046)
广西自然科学基金项目(2018GXNSFAA281283)。
关键词
变风量空调
送风量
深度置信神经网络
预测模型
ariable air volume air-conditioning
supply air volume
deep-confidence neural network method
prediction model
