住宅区供热负荷预测研究
|
摘要
本文分析了住宅区供热负荷影响因素,利用BP神经网络算法进行预测,建立了两种BP神经网络结构,一种输入样本是前几日供热负荷值和室外温度,另一种输入样本是室外温度、室外风速、室内用热系数及前几日供热负荷值,二者网络结构输出都是当天供热负荷值;对某住宅区域供热负荷进行了供热负荷预测与验证,比较了采用不同神经网络结构进行供热负荷预测的误差。结果表明住宅区域供热负荷变化规律具有明显的周期性,同时每天逐时供热负荷变化趋势线形态存在不同,需要选择合适的影响因素和样本个数,提高供热负荷预测的准确性。
This paper analyzed the influence factors of heating load in residential area. The BP neural network algorithm was used for forecast, and two BP neural network structure were built.One input sample was heating load value and outdoor temperature for the previous days, the other one was outdoor temperature, outdoor wind speed, indoor heat coefficient and heat supply in the last few days. The two network structure output were both heating load value on the day. Besides,the heating load of a residential district was predicted and verified heating load, and the error of heating load forecasting with different neural network structure was compared. The result shows that the regulation of heating load changes in residential area is periodic. It is necessary to choose appropriate factors and sample number to improve the accuracy of heating load forecasting due to diffident changing trend of heating load each day.
This paper analyzed the influence factors of heating load in residential area. The BP neural network algorithm was used for forecast, and two BP neural network structure were built.One input sample was heating load value and outdoor temperature for the previous days, the other one was outdoor temperature, outdoor wind speed, indoor heat coefficient and heat supply in the last few days. The two network structure output were both heating load value on the day. Besides,the heating load of a residential district was predicted and verified heating load, and the error of heating load forecasting with different neural network structure was compared. The result shows that the regulation of heating load changes in residential area is periodic. It is necessary to choose appropriate factors and sample number to improve the accuracy of heating load forecasting due to diffident changing trend of heating load each day.
引文
[1]Henrik Aalborg Nielsen, Henrik Madsen. Modelling the heat consumption in district heating systems using a grey-box approach[J]. Energy and Buildings, 2006, 38(1):63-71.
[2]Daniela Popescu, Fiorina Ungureanu, Abel Hemarndez-Guerrero. Simulation model for the analysis of space heat consumption of buildings[J].Energy, 2009, 34(10):1447-1453.
[3]胡文斌,杨昌智,孙一坚.短期供热负荷的灰色拓扑预测[J].煤气与热力,1999,19(3):51-54.
[4]张晓彤,刘金祥,陈晓春,张建.区域建筑冷(热)负荷的影响因素敏感性分析及预测方法研究[J].建筑科学,2013,29(08):77-82.
[5]康胜文.两种热负荷预测方法的比较[J].区域供暖,2004(2):22-24.
[2]Daniela Popescu, Fiorina Ungureanu, Abel Hemarndez-Guerrero. Simulation model for the analysis of space heat consumption of buildings[J].Energy, 2009, 34(10):1447-1453.
[3]胡文斌,杨昌智,孙一坚.短期供热负荷的灰色拓扑预测[J].煤气与热力,1999,19(3):51-54.
[4]张晓彤,刘金祥,陈晓春,张建.区域建筑冷(热)负荷的影响因素敏感性分析及预测方法研究[J].建筑科学,2013,29(08):77-82.
[5]康胜文.两种热负荷预测方法的比较[J].区域供暖,2004(2):22-24.