摘要
为探索一种新的工业园区能源管理方案,构建含有垃圾焚烧发电厂的多能源多时间尺度模型.提出了TSA-FEDformer的改进型深度学习算法和针对电能和热能的多时间尺度优化调度方法,将其应用于对历史数据的处理和预测,并以位于江苏连云港的某工业园区为案例进行验证.结果表明,该方法可以有效地降低运行成本和垃圾焚烧发电厂的出力波动,设置的两种场景运行成本相对于常规规则方法分别下降了21.08%和20.19%,垃圾焚烧发电厂出力波动平均变化率减少了1.2%,出力变化率峰值减少了4.28%.
To explore a new energy management solution for an industrial park,a multi-energy and multi-time scale model incorporating an incineration power plant was constructed.By utilizing the improved deep learning algorithm TSA-FEDformer proposed in this study to process historical data and make predictions,a method for optimizing the dispatch of electricity and heat energy across multiple time scales was presented.This approach effectively reduces operational costs and output fluctuations of the incineration power plant.Finally,a case study was conducted using an industrial park in Lianyungang,Jiangsu Province.The results show that the operational costs under two proposed scenarios decreased by 21.08%and 20.19%relative to conventional rule-based methods,while the average fluctuation rate of the incineration power plant output decreased by 1.2%,and the peak output variation rate decreased by 4.28%.
作者
李争
沙弘宇
李宁
韩华林
LI Zheng;SHA Hongyu;LI Ning;HAN Hualin(School of Electrical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China)
出处
《昆明理工大学学报(自然科学版)》
北大核心
2025年第6期111-122,共12页
Journal of Kunming University of Science and Technology(Natural Science)
基金
国家自然科学基金项目(51877070)。
关键词
滚动优化
调度策略
敏感性分析
多时间尺度
rolling optimization
scheduling strategy
sensitivity analysis
multiple time scales
