基于深度强化学习的居民实时自治最优能量管理策略

随着居民分布式资源的普及,如何考虑用户多类型设备的运行特性,满足实时自治能量管理需求以达到用户侧经济性最优成为亟待解决的课题。传统基于模型的最优化方法在模型精准构建和应对多重不确定性等方面存在局限性,为此提出一种无模型的基于深度强化学习的实时自治能量管理优化方法。首先,对用户设备进行分类,采用统一的三元组描述其运行特性,并确定相应的能量管理动作;接着,采用长短期记忆神经网络提取环境状态中多源时序数据的未来走势;进而,基于近端策略优化算法,赋能在多维连续-离散混合的动作空间中高效学习最优能量管理策略,在最小化用电成本的同时提升策略对不确定性的适应性;最后,通过实际情境对比现有方法的优化决策效果,验证所提方法的有效性。     

新能源分比例渗透下基于博弈的多主体效益综合评估方法

“双碳”目标下新能源大规模发展,新能源的发展给各能源主体带来了影响,如何量化评估新能源渗透给能源主体带来的影响成为研究重点。首先,基于改进的K-means算法构建基于典型场景集的新能源出力模型;然后,从源-网-荷-储出发,构建火力发电商、电网、用户、储能运营商的决策优化模型;其次,构建考虑新能源渗透的多能源主体效益评估模型;最后,进行算例分析,算例结果验证了模型的有效性。     

基于嵌入式平台的电能信息采集系统研究

针对目前电能信息采集系统存在的通信数据量大、通信费用高的问题,设计了一种基于uC/OS嵌入式实时操作系统的电能信息采集系统,该系统包括智能电表、电能信息采集终端、上位机软件和主站数据中心。系统根据上位机设置的参数下行采集电表中数据,数据在采集终端经过数据压缩处理后,通过远程通信技术上行传输至主站,便于工作人员监测和管理电能数据。文中结合Lempel-Ziv-Oberhumer无损压缩技术和游程编码技术对数据压缩方案进行改进,通过数据压缩对比实验,验证数据压缩方案的可行性。实验结果表明,系统采用的数据压缩技术有效提高了数据传输的通信效率,验证了系统的实用性和有效性。     

基于粗糙集的电池储能电站海量数据处理方法

锂离子电池的储能电站在运行过程中上传的数据量庞大,数据采样率高,增大了实时在线评估工作的难度。如何提取并压缩有效数据,同时保证压缩后的数据具有保真性,成为数据挖掘预处理所面对的主要问题。针对上述问题,选取储能电站特定工况下一个电池簇的数据,利用粗糙集方法,对数据的属性进行约简,根据正态分布的$ 2\sigma $原则对属性值进行二元逻辑划分,将单体电池划分为频繁检测对象和普通检测对象,压缩了在线处理的数据量。最后,对同一电池簇20天内不同工况数据进行验证,证明了该处理方法的有效性。     

电网能量流-经济流耦合分析

依托于同一电力网络的电气源流间的能流信息和市场主体间的交易信息,存在某种天然耦合关系。若不能正确处理,则易导致相关经济流分布信息掌握失准,从而影响系统运行的效能和市场运作的公正性。为此,基于电力网络电能追踪方法正确提供的实时能流信息,考虑上述耦合特性,分析了沿网络源-流对电能配送路径链分布的经济流信息,建立了核算各项费用的实时电价关系式;针对链首和链末2种能量计费方式,提出了具有广泛指导性的费用核算一致性原理,并指出：若不考虑电网实时运行状态影响而直接以上网价、过网价和政府性基金价之和作为下网电价核算相关费用,则会出现不完全满足一致性原理的现象。最后,针对典型交易模式及与其关联的多个网络源-流对,给出了量化其经济流分布的关系式。算例表明,基于考虑耦合特性的实时电价信号分析的经济流,符合核算一致性原理,可为进一步揭示交叉补贴内因提供理论依据。     

Applications of shared economy in smart grids: Shared energy storage and transactive energy

Increasing renewable energy is integrated with power systems and more flexibility is required to address the fluctuation problems caused by renewable energy. Advanced communication and measurement technologies enable the electric grids to approach smart grids which have more solutions for the high penetration of renewable energy. The shared economy as an emerging commercial model has attracted much attention and is widely applied in smart grids. This paper is focused on the state of the art of shared energy storage and transactive energy (TE) which are the typical applications of shared economy in smart grids. The concept, market structure, and demonstration projects of shared energy storage and TE are provided.     

Business models for accelerating phase-out of coal based generation: Developing typologies and a discussion of the relative merits of alternative models

Coal-fired electricity generation accounts for more than 10 Gt, or roughly 30 %, of global CO₂ emissions, with the majority of coal-fired generation capacity located in developing countries. It is possible to significantly reduce emissions from older, expensive and inefficient coal plants at less than US$10/tCO₂e. However, in the absence of suitable business models and financing options, accelerated coal decommissioning may not transpire. Several business models have been tested, such as policy-based closure, buyout, and repurposing, and swapping for renewables. Features of different models can be combined to develop hybrid models. This paper develops typologies for alternative business models so that the available options can be clearly understood and compared. The paper also discusses the potential merits and drawbacks of each model. Rather than offering definitive answers on a suitable approach for coal transition. the paper aims to collate the experience thus far and highlight questions that require further examination.     

Lacking energy storage,and nuclear contribution,wind, and solar photovoltaic electricity is expensive and scarce

Prices of electricity have skyrocketed phased with the uptake of non-dispatchable wind and solar photovoltaic energy supply, and the closure of combustion fuel plants able to produce electricity on demand. At the same time, blackouts have become more likely, with governments requesting citizens not to use electricity when they need to prevent their occurrence. If we were expecting cheap and abundant renewable energy, it seems something is not right. There are, however, solutions. These are the adoption of other energy sources, not only non-dispatchable wind and solar, but also dispatchable hydroelectric, geothermal, and biomass renewable energy, and baseload nuclear energy, plus the development of energy storage, by batteries, pumped hydro, thermal energy storage, and more than that, by hydrogen, to generate enough dispatchable energy to cover the requirements of a growing world.     

考虑综合能源站柔性调控作用的城市配电网多阶段规划方法

随着城市经济发展与终端用能电气化的普及,以制冷用电为首的电负荷需求持续增加且峰谷差情况愈发严重化,导致现有配电网运行负担逐渐加重。在此背景下,将综合能源站(IES)作为多能流耦合节点,在配电网中引入多能协调互补及蓄能移峰功能,计及多个阶段负荷发展及多类型负荷特征,提出在IES柔性调控作用下的配电网多阶段规划方法。基于IES能流结构及其运行机制,分别在不同气候类型地区负荷模拟场景下,以总规划期内投资与运行经济性最优为目标,建立由线路、变电站和IES构成决策对象的配电网双层多阶段规划模型,规划与运行层模型通过决策变量关联达成一体化协同求解。通过修改后的IEEE 33节点配电系统对所提方法的有效性进行校验,并进一步将所提方法应用于某地实际152节点配电系统,分析不同负荷特征下配电网规划运行效益、负荷峰谷差调节效果及IES运行状况,得出计及IES的城市配电网规划方法的实用性结论。     

基于改进三角模糊VIKOR法的用户储能系统配置方案综合评估

针对不确定环境下用户储能系统配置方案评估决策,提出一种基于改进三角模糊多准则妥协解排序(VIKOR)法的评估模型。首先,运用扎根理论构建评估指标集,并采用灰色关联-粗糙集方法进行指标筛选,形成评估指标体系。然后,采用主、客观相结合的方法计算指标组合权重,以提高其准确性,其中主观权重由模糊层次分析(FAHP)法求得,客观赋权法由改进标准间重要性相关性(CRITIC)法求得。最后,考虑指标之间的冲突性与模糊性,运用改进三角模糊VIKOR法对配置方案评估与排序比选。通过仿真分析,验证了本文所建模型的有效性和可行性。