摘要
通过选取城市智慧照明规划指标体系中具有典型性和代表性的基础类指标,将中国60座城市的数据进行收集与分级,利用MATLAB平台的LM-BP神经网络对其中50座城市的750项数据进行学习,完成训练后,利用剩余10座城市的数据进行验证,预测结果的准确性达到100%,表明本实验系统具备对城市智慧照明发展基础的分级能力,为研究城市当前阶段智慧照明的评价内容与发展目标提供了一种基础研究方法。
By selecting the typical and representative basic indicators in the urban intelligent lighting planning index system,the data of 60 cities in China are collected and graded,and the LM-BP neural network of MATLAB platform is used to learn 750 items of data of 50 cities.After the training,the data of the remaining 10 cities are used for verification,and the accuracy of the prediction results reaches 100%.It shows that the experimental system has the ability of grading the development basis of urban smart city lighting,and provides a basic research method for the study of the evaluation content and development goals of urban smart lighting at the current stage.
作者
王超
张明宇
王锡铭
吴传德
WANG Chao;ZHANG Mingyu;WANG Ximing;WU Chuande(Tianjin Key Laboratory of Architectural Physics and Environmental Technology,Tianjin University,Tianjin 300072,China)
出处
《照明工程学报》
2021年第5期8-14,共7页
China Illuminating Engineering Journal
关键词
城市照明规划
智慧照明指标
BP神经网络
urban lighting planning
intelligent lighting index
LM-BP neural network
