摘要
为解决中国城市独居老人健康监护难的问题,本文提出一种基于用电多注意力的独居老人风险识别方法。通过新一代智能电表采集独居老人的历史用电数据,提取独居老人各类型电器的用电特征;通过多注意力机制学习独居老人的电量趋势、负荷波动等特征,建立独居老人用电指纹。在此基础上,将当前独居老人的用电数据与用电指纹数据进行比对,判断独居老人是否存在风险。在某城市社区应用该方法,识别准确率达96.7%。该方法通过电力数据实现了独居老人风险识别,提升了政府的独居老人管理水平。
In order to solve the problem of health monitoring for the elderly people living alone in urban area of China,we propose a risk identification method for the elderly living alone based on multi-attention of electricity consumption.Firstly,the method uses a new generation of smart meters to collect historical power consumption data of the elderly living alone to extract the power consumption characteristics of various types of electrical appliances of the elderly living alone.And then,based on the multi-attention mechanism,it learns the characteristics of electricity consumption trend and load fluctuation of the elderly living alone to establish the electric fingerprint of the elderly living alone.On this basis,it compares the current electricity use data of the elderly living alone with the electricity consumption fingerprint data to judge whether the elderly living alone are at risk.This method has been applied in a city community,with the recognition accuracy of 96.7%.It realizes the risk identification of the elderly living alone with power data,and improves the management level of the government for the elderly living alone.
作者
唐冬来
李擎宇
龚奕宇
钟声
陈泽宇
聂潇
TANG Donglai;LI Qingyu;GONG Yiyu;ZHONG Sheng;CHEN Zeyu;NIE Xiao(Sichuan SGIT Technology Co.,Ltd.,Chengdu 610047,China;State Grid Sichuan Electric Power Company,Chendu 610041,China)
出处
《应用科技》
CAS
2023年第3期134-138,共5页
Applied Science and Technology
基金
国家重点研发计划项目(2016YFB0901102).
关键词
用电特征
多注意力
独居老人
风险识别
负荷波动
用电指纹
智能电表
负荷识别
characteristics of electricity consumption
multi-attention
the elderly living alone
risk identification
load fluctuation
electricity fingerprint
smart meter
load identification
