摘要
短期热负荷预测对于城市集中供热系统的安全经济运行有着重要的作用,提高预测的精度是广大学者研究的首要任务。长期以来,研究者尝试了多种不同类型的神经网络对短期热负荷进行预测。由于集中供热系统的热负荷受外界温度等因素的影响较大,因而在进行神经网络建模时考虑天气因素才能使预测结果更为精准。该文提出一种基于天气预报的改进BP神经网模型来预测供热系统的短期热负荷。在Matlab平台上通过API接口调用天气预报中的温度值,作为神经网路的一部分输入,和数据库中的热负荷值一起,同时进行网络训练,并考虑一些不确定干扰因素。实验表明,加入天气预报的神经网络模型使热负荷预测达到了更为精准的效果。
Short-term heat load forecasting plays an important role in the safe and economic operation of the urban district heating system,and improving the accuracy of prediction was the first task for many scholars＇ study. For a long time,some researchers have tried different types of neural network to forecast the short-term heat load. As a result of the strongly influence which caused by the external factors such as temperature for district heating system heat load,considering the weather factors can make the prediction results more accurately when making neural network model. This paper presents a improved BP neural network model based on weather forecast which can forecast the short-term heat load. The weather forecast,as a part of the neural network input,gained through the temperature API interface and are trained with the historical data of the heat load at the same time. The program were ran on the Matlab platform. Some uncertain interference factors were also considered in the experiment. The experimental result shows that the neural network model based on weather forecast can make the heat load prediction more accurately.
出处
《自动化与仪表》
2015年第5期5-8,共4页
Automation & Instrumentation
基金
国家自然科学基金项目(61463040)
内蒙古自然科学基金资助项目(2012MS0910)
