基于决策树算法的园区照明系统能耗优化控制方法
投稿时间:2022-03-22  修订日期:2022-04-26  点此下载全文
引用本文:
摘要点击次数: 171
全文下载次数: 0
作者单位邮编
赵宇冰* 国网甘肃省电力公司 730000
刘育平 国网甘肃省电力公司 
卜小伟 国网甘肃省电力公司 
祝存平 国网甘肃省电力公司 
周进艳 国网甘肃省电力公司 
杨建伟 国网甘肃省电力公司刘家峡水电厂 
吴相荣 国网思极飞天(兰州)云数科技有限公司 
中文摘要:针对园区照明系统能耗数据缺失、区域能耗难以控制的问题,助力企业节能、减排,提出一种基于决策树算法的园区照明系统能耗优化控制方法。预处理园区照明系统能耗数据,通过属性选择统一能耗数据属性值,采用基于马氏距离的k均值填充法填补能耗数据缺失值;选取C4.5决策树搭建园区照明系统能耗回归预测模型,预测获取园区照明系统总能耗,构建园区照明系统能耗控制目标函数,衡量照度及照明设备亮度二者跟随时间推移产生的规律,以此判断园区照明系统设备的下一时刻亮度,实现对园区照明系统的能耗优化控制。实验结果表明,所提方法可有效填补缺失能耗数据,并精准预测园区照明系统能耗值,预测精度高达99.16%;优化控制后的年度耗电量相比上一年度缩减两倍,优化控制效果优秀。
中文关键词:决策树算法  园区照明系统  能耗预测  优化控制  缺失值填补  照明设备
 
Optimal control method of energy consumption in campus lighting system based on decision tree algorithm
Abstract:Aiming at the problems of lack of energy consumption data of park lighting system and difficult to control regional energy consumption, and helping enterprises to save energy and reduce emissions, a park lighting system energy consumption optimization control method based on decision tree algorithm is proposed. Preprocess the energy consumption data of lighting system in the park, unify the attribute value of energy consumption data through attribute selection, and fill in the missing value of energy consumption data by K-means filling method based on Mahalanobis distance; Select C4 5. The decision tree establishes the regression prediction model of the energy consumption of the lighting system in the park, predicts and obtains the total energy consumption of the lighting system in the park, constructs the energy consumption control objective function of the lighting system in the park, and measures the law of the illumination and the brightness of the lighting equipment with the passage of time, so as to judge the brightness of the lighting system equipment in the park at the next moment and realize the optimal control of the energy consumption of the lighting system in the park. The experimental results show that the proposed method can effectively fill in the missing energy consumption data and accurately predict the energy consumption of the lighting system in the park, with a prediction accuracy of 99.16%; The annual power consumption after optimization control is reduced by twice compared with the previous year, and the optimization control effect is excellent.
keywords:Decision Tree Algorithm  Campus Lighting System  Energy consumption prediction  Optimal control  Missing value filling  Lighting equipment
查看全文   查看/发表评论   下载pdf阅读器