基于滚动优化的电-热-气-冷系统多时间尺度低碳运行
Multi-Timescale Low-Carbon Operation of the Electric-Heat-Gas-Cooling Combined Supply System Based on Rolling Optimization
徐楠 1陈斌 2黄伟 2靳梓康 3王义3
作者信息
- 1. 国能神东煤炭集团有限责任公司机电管理部,陕西榆林 719300
- 2. 许继电气股份有限公司,河南许昌 461000
- 3. 郑州大学电气与信息工程学院,河南郑州 450001
- 折叠
摘要
电-热-气-冷多能联供型微网对实现能源可持续发展具有重要的应用价值.针对多能联供系统碳排放量较高和负荷模型预测不准确问题,提出了一种基于滚动优化的电-热-气-冷系统多时间尺度低碳运行策略.首先,建立电-热-气-冷系统设备模型.其次,构建日前与日内两阶段模型,在日前调度阶段引入含赏罚因数的碳交易机制,通过将卷积神经网络(convolutional neural networks,CNN)与双向长短期记忆网络(bi-directional long short term memory,Bi-LSTM)进行结合对风光功率进行预测,并以运行成本最低为目标进行优化.之后,建立日内多时间尺度的优化调度模型,以调度成本最低为目标进行求解.最后,以某市综合能源系统为研究对象进行分析.结果表明,所提出的方法能够有效减少碳排放,提高负荷模型预测的准确度的同时实现多能联供系统的低碳经济运行.
Abstract
In view of the high carbon emissions of the multi-energy cooperation system and the inaccurate prediction of the load model,this paper proposes a multi-time scale low-carbon operation strategy of the electricity-heat-gas-cooling system based on rolling optimization.Firstly,the equipment model of the electrical-thermal-gas-cooling system is established.Secondly,a two-stage model of day-ahead and intra-day is constructed,and a carbon trading mechanism with reward and penalty factors is introduced in the day-ahead scheduling stage The wind and solar data are predicted by combining convolutional neural networks(CNNs)and bidirectional long short term memory(Bi-LSTM)networks and optimize with the goal of minimizing operating costs..Thirdly,an intraday multi-time scale optimization scheduling model is established,and the rolling optimization method is adopted to solve the problem with the goal of lowest scheduling cost.Finally,a comprehensive energy park in a certain city is taken as the research object for analysis.The results show that the proposed method can effectively reduce carbon emissions,improve the accuracy of load model prediction,and realize the low-carbon economic operation of multi-energy cogeneration system.
关键词
电-热-气-冷/多能协同/阶梯式碳交易机制/多时间尺度/滚动优化/CNN-Bi-LSTMKey words
power-heating-gas-cooling/multi-energy coordination/stepped carbon trading mechanism/multiple time scales/rolling optimal/CNN-Bi-LSTM引用本文复制引用
基金项目
国家自然科学基金(62203395)
河南省博士后科研启动基金(202101011)
出版年
2024
NETL
NSTL
万方数据