供暖期间多电源优化调度策略及预测分析  

Optimal scheduling strategy and predictive analysis of multiple power sources during heating season

作　　者：李东雄[1] 杨文旭[1] 马凯 丰胤东 成治瑾 Li Dongxiong;Yang Wenxu;Ma Kai;Feng Yindong;Cheng Zhijin(Shanxi University,Taiyuan)

机构地区：[1]山西大学,太原

出　　处：《暖通空调》2025年第2期32-37,共6页Heating Ventilating & Air Conditioning

摘　　要：风光发电是山西能源转型发展的重要组成部分,冬季供暖期是风光发电出力较高的季节,由于风光发电具有季节性、随机性和间歇性特点,使得风光发电随装机容量的增加对电网峰谷差的反向调峰作用进一步增强。因此,合理调配电力资源,促进风光电力消纳,降低弃风弃光率很有必要。本文提出在目前由风电、光电及火电作为主要电源的发电系统中加入氢储能调峰站进行调节,用来平抑新能源发电的波动性,保障电网系统的安全稳定运行,同时在目前的负荷侧加入新能源汽车集群来增大负荷侧需求,推进源网荷储一体化。计算出山西省风电机组、光电机组月平均出力系数和保证容量系数,并分析其年特性;建立对应的以风光联合出力最大化和火电机组运行出力最小化为目标的电力系统调度数学模型,对其进行算例应用分析验证;基于长短期记忆网络模型预测山西省供暖期一日内各机组的出力,通过粒子群算法优化模型,模拟加入氢储能后的风电、光电机组出力,并最终验证模型的可行性。Wind power generation and solar power generation are essential parts of energy transition and development in Shanxi.During the winter heating season,they generate higher output.The seasonal,random,and intermittent natures of wind power and solar power increase their ability to counteract peak-to-valley differences in the grid as installed capacity grows.Therefore,it is crucial to efficiently allocate power resources,promote the integration of wind power and solar power,and reduce their curtailment rates.This paper suggests integrating hydrogen energy storage power stations into current power systems that mainly use wind power,solar power,and thermal power to mitigate the fluctuations in renewable energy generation and ensure the safe and stable operation of the electrical grid.It also proposes adding clusters of electric vehicles to increase demand on the load side,promoting an integrated approach to generation,network,load and storage.This paper calculates the monthly average output coefficients and guaranteed capacity coefficients for wind and solar power units in Shanxi province,analyses their annual performance,and develops a mathematical model for power system scheduling aimed at maximizing the output of wind and solar power units while minimizing the operating costs of thermal power units.The model is validated through case studies.Using the long short-term memory(LSTM)network model,it predicts the 24-hour power output of each unit in Shanxi province during the heating season.The particle swarm optimization(PSO)algorithm is used to further optimize the model,simulate the output power of wind and solar power units with added hydrogen energy storage,and ultimately verify the model s feasibility.

关 键 词：新能源 氢储能 出力预测 电力调度 风电 光电 

分 类 号：TM73[电气工程—电力系统及自动化] TU832[建筑科学—供热、供燃气、通风及空调工程]