基于遗传优化决策树的建筑能耗短期预测模型  

Short-term Prediction Model of Building Energy Consumption Based on Genetically Optimized Decision Tree

作　　者：丁飞鸿 刘鹏 卢暾[1,2,3] 顾宁[1,2,3] 丁向华 杨宝明[4] 戴文祺 邹超君[4] DING Feihong;LIU Peng;LU Tun;GU Ning;DING Xianghua;YANG Baoming;DAI Wenqi;ZOU Chaojun(School of Computer Science,Fudan University,Shanghai 201203,China;Shanghai Key Laboratory of Data Science,Fudan University,Shanghai 201203,China;Shanghai Institute of Intelligent Electronics and System Research,Shanghai 201203,China;Shanghai Luban Software Co.,Ltd., Shanghai 200433,China)

机构地区：[1]复旦大学计算机科学技术学院,上海201203 [2]复旦大学上海数据科学重点实验室,上海201203 [3]上海智能电子与系统研究院,上海201203 [4]上海鲁班软件股份有限公司,上海200433

出　　处：《计算机工程》2019年第6期280-289,296,共11页Computer Engineering

基　　金：上海市科委项目“信息化技术与调适技术的集成研究与示范”(16DZ1202402)

摘　　要：使用线性回归模型预测公共建筑能耗数据时,存在不确定性影响因素和精度偏差问题。为此,建立一种遗传优化决策树模型。采用遗传算法优化梯度提升决策树的子树生成过程,以模型预测的R-Square值作为迭代的评估标准,从而达到能耗预测的目的。实验结果表明,与传统的回归预测模型相比,该模型预测精度较高。When using linear regression models to predict public building energy consumption data,there are uncertainties and precision deviations.Therefore,a Genetically Optimized Decision Tree(GODT) model is established.The Genetic Algorithm(GA) is used to optimize the subtree generation process of the Gradient Boosting Decision Tree (GBDT).The R-Square value predicted by the model is used as the evaluation criterion of the iteration to achieve the purpose of energy consumption prediction.Experimental results show that the prediction accuracy of this model is higher than that of the traditional regression prediction model.

关 键 词：建筑能耗 回归预测 梯度提升决策树 遗传算法 遗传优化决策树 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]