详细信息
光照度聚类和支持向量机在路灯节能控制策略中的应用 被引量:6
Application of Illumination Clustering and SVM in Energy-saving Control Strategy of Street Lamps
文献类型:期刊文献
中文题名:光照度聚类和支持向量机在路灯节能控制策略中的应用
英文题名:Application of Illumination Clustering and SVM in Energy-saving Control Strategy of Street Lamps
作者:文俊浩 万园 曾骏 王喜宾 梁冠中
第一作者:文俊浩
机构:[1]重庆大学大数据与软件学院,重庆401331;[2]贵州理工学院大数据学院,贵阳550003
第一机构:重庆大学大数据与软件学院,重庆401331
年份:2019
卷号:46
期号:7
起止页码:327-332
中文期刊名:计算机科学
外文期刊名:Computer Science
收录:CSTPCD;;北大核心:【北大核心2017】;CSCD:【CSCD_E2019_2020】;
基金:国家自然科学基金(61672117);学术新苗项目黔科合平台人才([2017]5789-21)资助
语种:中文
中文关键词:支持向量机;K-means;路灯节能;光照度
外文关键词:Support vector machine;K-means;Energy-saving of street lamp;Illumination
摘要:传统路灯行业主要采用时间、经纬度、光照度等策略控制路灯开关。其中,光照度控制的理论节能效果最佳,但受采集数据误差、安装角度等环境因素影响,节能效果没有达到最大化。针对该问题,提出一种融合光照度聚类和支持向量机算法的路灯节能控制策略。该方法收集光照度、时间、安装角度数据,并使用K-means算法对光照度数据进行聚类,把原本变化剧烈的光照度数据变为5个等级(1-5),然后通过SVM对数据进行学习训练,在不考虑其他外在因素的情况下预测路灯的开关时间。实验研究结果表明,该算法可有效降低路灯的用电量。
The traditional street lighting industry mainly adopts the strategy of time,latitude and longitude,illumination and so on to control street lights,and the theory of illumination control has the best energy saving effect.However,due to the error of light collection,installation angle and other environmental factors,the energy saving effect has not been maximized.Aiming at this problem,this paper proposed a street lighting energy saving control strategy based on illumination clustering and support vector machine algorithm.This method collects the light intensity,time and installation angle data,and uses K-means algorithm to cluster the illumination data and changes the original light illumination data into 5 grades(from 1 to 5).Then,the data is trained by SVM,and the switching time of the street lamp is predicted without considering other external factors.The experimental results show that the algorithm can effectively reduce the power consumption of street lamps.
参考文献:
正在载入数据...