基于LSTM循环神经网络的中小学建筑运营维护期间电力碳排放测算方法研究——以瑞安市滨海实验小学为例
Research on Measurement Method of Electricity Carbon Emissions during Operation and Maintenance of Primary and Secondary School Buildings based on LSTM Recurrent Neural Network——Case Study of Ruian Binhai Experimental Primary School
诸纪萍 1李强 1茅建平 1李超 2邵嘉妍2
作者信息
- 1. 浙江大东吴集团建设有限公司,湖州 313002
- 2. 湖州职业技术学院,湖州 313099
- 折叠
摘要
本文以瑞安小学为案例,利用LSTM循环神经网络对教学楼建筑的电力碳排放进行预测.通过收集和分析能源消耗数据、建筑特征和环境因素,建立了一个基于LSTM循环神经网络的碳排放预测模型,并将预测结果与SVR算法进行了对比.结果表明,该模型在预测准确性和稳定性上表现出明显优势,为碳排放管理和减排决策提供了可靠依据.这一研究为推动建筑碳减排和可持续发展提供了有益的参考和借鉴.
Abstract
This paper uses LSTM recurrent neural network to predict the electricity carbon emissions of a teaching building construction in Ruian Binhai Experimental Primary School.By collecting and analyzing energy consumption data,building characteristics and environmental factors,a carbon emission prediction model is established based on LSTM cycle neural network,and prediction results are compared with the SVR algorithm.The results show that the model has obvious advantages in prediction accuracy and stability,and provides a reliable scientific basis for carbon emission management and emission reduction decisions.This study provides a useful reference for promoting building carbon emission reduction and sustainable development.
关键词
碳排放预测/神经网络/中小学建筑Key words
carbon emission prediction/neural network/primary and secondary school buildings引用本文复制引用
基金项目
浙江省建设厅建设科研项目(2022K051)
湖州职业技术学院横向课题(202HX3025)
出版年
2024
NETL
NSTL
万方数据