基于人体舒适度指数的高峰季节空调负荷预测方法  

AIR CONDITIONING LOAD FORECASTING METHOD IN PEAK SEASON BASED ON HUMAN BODY AMENITY INDEX

作　　者：韩平平[1] 丁静雅 吴红斌[1] 仇茹嘉 徐斌 吴家毓 Han Pingping;Ding Jingya;Wu Hongbin;Qiu Rujia;Xu Bin;Wu Jiayu(Anhui Province Key Laboratory of Renewable Energy Utilization and Energy Saving(Hefei University of Technology),Hefei 230009,China;State Grid Anhui Electric Power Company Limited Research Institute,Hefei 230601,China)

机构地区：[1]新能源利用与节能安徽省重点实验室(合肥工业大学),合肥230009 [2]国网安徽省电力有限公司电力科学研究院,合肥230601

出　　处：《太阳能学报》2025年第3期141-150,共10页Acta Energiae Solaris Sinica

基　　金：安徽省自然科学基金(2008085UD10);国家自然科学基金区域创新发展联合基金(U19A20106)。

摘　　要：提出一种基于综合人体舒适度指数的高峰季节空调负荷预测方法,从而获得更加准确的空调负荷数据参与电网调控。首先,考虑到不同季节的负荷增量影响和数据样本范围,分别利用最大负荷比较法和基准负荷比较法得到更具可信度的空调负荷数据;其次,计算包含温度、相对湿度和风速指标的主客观综合权重,构建考虑时空分布特性的人体舒适度模型,并验证其与空调负荷之间的关联性;最后,基于综合人体舒适度指数提取建模样本数据,并将其作为神经网络的输入,建立空调负荷预测模型。理论分析和算例验证表明所提方法在不同情景下可有效提高空调负荷预测精度。This paper proposes an air conditioning load forecasting method in peak season based on the comprehensive human body amenity index,so that more accurate air conditioning load data can be provided for power grid regulation.First,considering the impact of load increment in different seasons and the range of data samples,the maximum load comparative method and the reference load comparative method are used to obtain more reliable air conditioning load data.Then,the subjective and objective comprehensive weights,which includes temperature,relative humidity,and wind speed indicators are calculated.A human body amenity model is constructed to consider spatiotemporal distribution characteristics,and its correlation with air conditioning load is verified.Finally,utilizing on the comprehensive human body amenity index,the modeling sample data is extracted and used as input to the neural network to establish an air conditioning load forecasting model.Theoretical analysis and numerical cases demonstrate that the proposed method can effectively improve the accuracy of air conditioning load forecasting in different scenarios.

关 键 词：分布式发电 空调 负荷预测 人体舒适度指数 双向长短期记忆网络 

分 类 号：TM732[电气工程—电力系统及自动化]