Modeling and application of marketing and distribution data based on graph computing

Integrating marketing and distribution businesses is crucial for improving the coordination of equipment and the efficient management of multi-energy systems.New energy sources are continuously being connected to distribution grids;this,however,increases the complexity of the information structure of marketing and distribution businesses.The existing unified data model and the coordinated application of marketing and distribution suffer from various drawbacks.As a solution,this paper presents a data model of"one graph of marketing and distribution"and a framework for graph computing,by analyzing the current trends of business and data in the marketing and distribution fields and using graph data theory.Specifically,this work aims to determine the correlation between distribution transformers and marketing users,which is crucial for elucidating the connection between marketing and distribution.In this manner,a novel identification algorithm is proposed based on the collected data for marketing and distribution.Lastly,a forecasting application is developed based on the proposed algorithm to realize the coordinated prediction and consumption of distributed photovoltaic power generation and distribution loads.Furthermore,an operation and maintenance(O&M)knowledge graph reasoning application is developed to improve the intelligent O&M ability of marketing and distribution equipment.

考虑灵活爬坡产品的虚拟电厂两阶段分布鲁棒优化运营策略

随着“碳达峰·碳中和”目标的提出,可再生能源大规模并网对电力系统灵活运行提出了更高要求。虚拟电厂(VPP)通过聚合分布式资源,将灵活爬坡产品(FRP)纳入运营策略,参与多类型电力市场以提升运营收益。基于此,提出了考虑FRP的VPP两阶段分布鲁棒优化(DRO)运营策略。首先,考虑高比例可再生能源渗透率下的FRP市场,结合多类型分布式资源运行特性,提出VPP参与多类型电力市场的优化运营框架;其次,结合基于Wasserstein距离的可再生能源出力不确定集和基于场景分析法的电力市场价格典型场景集,构建考虑FRP的VPP日前投标和日内调度两阶段DRO模型;最后,以某VPP为例对所提模型进行有效性验证。算例结果表明,所提VPP运营策略具有较好的调度经济性与风险鲁棒性,在参与多类型电力市场交易中能够实现运营收益最大化。

考虑分时碳价的虚拟电厂分布鲁棒竞标方法

为了探索双碳目标下虚拟电厂(virtual power plants,VPP)低碳、经济的竞标策略,提出了一种基于分时碳价和分布鲁棒优化思想的VPP参与日前多类市场的竞标策略。从减小电力市场运营成本的角度,提出了一种与电价负相关的分时碳价,采用基准线法为VPP无偿分配碳配额,计算新能源发电的国家核证减排量;综合考虑碳交易市场、能量市场和调峰市场的情况,搭建了VPP参与多类市场的竞标模型;针对风电的不确定性,构建了基于数据驱动的分布鲁棒两阶段竞标模型,并采用列与约束生成算法求解。算例分析表明,分时碳价能够有效降低电力市场运营商的运行成本,提高VPP运营商的收益,所提的VPP竞标模型能给出不同市场的最优投标方案,而分布鲁棒能够兼顾经济性与鲁棒性。

基于分布鲁棒优化的发电商中长期合同电量分解模型

在我国中长期合同交易与现货市场衔接的政策背景下,中长期合同电量必须分解至短时间尺度。针对发电商中长期合同电量的自主分解问题,研究了发电商在现货市场价格严重不确定环境下基于分布鲁棒优化的中长期合同电量分解方法。首先,针对现货价格的严重不确定性,以Wasserstein距离作为分布函数距离测度,构建了现货价格分布不确定性的模糊集。其次,建立了发电商合同分解的分布鲁棒优化模型。以发电商利润最大化为目标,考虑发电商运行约束和合同分解约束,并采用对偶理论等将模型转化为可高效求解的二次规划问题。最后,通过算例仿真并与传统的随机优化和鲁棒优化模型进行对比。结果表明所提模型在处理现货价格不确定性时能有效兼顾发电商合同分解决策的经济性和鲁棒性。

Strategic Bidding in Distribution Network Electricity Market Focusing on Competition Modeling and Uncertainties

Developing the electricity market at the distribution level can facilitate the energy transactions in distribution networks with a high penetration level of distributed energy resources(DERs)and microgrids(MGs).However,the lack of comprehensive information about the marginal production cost of competitors leads to uncertainties in the optimal bidding strategy of participants.The electricity demand within the network and the price in the wholesale electricity market are two other sources of the uncertainties.In this paper,a day-ahead-market-based framework for managing the energy transactions among MGs and other participants in distribution networks is introduced.A game-theory-based method is presented to model the competition and determine the optimal bidding strategy of participants in the market.Robust optimization technique is employed to capture the uncertainties in the marginal cost of competitors.Additionally,the uncertainties in demand are modeled using a scenario-based stochastic approach.The results ob-tained from case studies reveal the merit of considering competition modeling and uncertainties.

Geographical Distribution Of Chinese Car Market

Ⅰ. Brief Account of Geographical Divi-sion The formation and development of the do-mestic car market are heavily influenced by the development of geographic economic momen-tum. On the basis of the general economic

基于多层利益共享的虚拟电厂参与电碳市场分布鲁棒优化模型

风电和光伏的高渗透率增加了新型电力系统对灵活性资源需求.虚拟电厂作为一个特殊电厂聚合了可控分布式电源、新能源、储能、碳处理、负荷等各类资源,“对内协同”可以实现内部资源协同调控,“对外统一”可以参与外部电碳市场获利.基于此,本文创新地提出了虚拟电厂参与电碳多类市场分布鲁棒优化模型.为了刻画风电和光伏的不确定性,构造了基于Wasserstein距离的分布模糊集和基于数据驱动的误差不确定集.为了兼顾经济性和鲁棒性,考虑内部运行成本以及外部参与多类市场收益,构建了最坏分布下期望收益最大的两阶段鲁棒优化模型,并提出模型求解方法.为了保证联盟动态平衡,提出了多层利益分配方法.最后,算例分析表明:在“对内协同,对外统一”的经营模式下,有效激发虚拟电厂内各类资源的潜力,参与多个市场后获取共享利益,实现了多方互利共赢.所提模型具有数据驱动、快速求解、灵活可控、经济实用等优越性.多层利益分配方法能够简便、有效地将共享效益传导至各主体.

Local Energy and Planned Ramping Product Joint Market Based on a Distributed Optimization Method

High penetration of renewable energy generation(REG)in the distribution system increases both the power uncertainty at a given interval and the power variation between two intervals.Reserve markets addressing power uncertainty have been widely investigated.However,there is a lack of market mechanisms regarding the power variation of the load and REGs.This paper thus defines a planned ramping(PR)product to follow the net load variation and extends the local energy market to include the trading of PR products.Players are economically compensated for their PR products.Bidding models of dispatchable generators and flexible load aggregators in the joint market are investigated.To solve the market problem in polynomial time,a distributed market clearing method is developed based on the ADMM algorithm.The joint market is tested on a modified IEEE 33-bus system.It verifies that introducing the PR market can encourage flexible loads to provide more PR service to accommodate the net load variation.As such,the ramping cost of dispatchable generators is reduced by 29.09%in the test case.The planned energy curtailment from REG is also reduced.The computational efficiency of the proposed distributed clearing method is validated by comparing it with a centralized method.

基于场景树概率驱动的电动汽车聚合商能量-调频市场分布鲁棒投标策略

电动汽车聚合商(electric vehicle aggregator,EVA)的能量-调频市场投标策略决定着电动汽车灵活性的市场价值,而EVA的投标决策过程面临着市场侧诸多不确定因素的影响。为此,针对市场价格及调频信号的不确定性建模问题,提出了一种考虑不确定因素多时间尺度相关性的分布鲁棒建模方法。首先,提出一种基于分层聚类法的场景树模型,刻画不同市场不确定因素的时序相关性。其次,建立了一种基于两阶段优化的EVA参与能量-调频市场的多时间尺度投标决策模型,并基于混合范数距离构建场景树概率的模糊集,实现EVA投标决策在分布鲁棒优化框架下求解。然后,利用列和约束生成法解决构建的max-min-max-min 4层鲁棒问题。最后,通过仿真验证了所建模型在解决两阶段不确定性优化问题和提升投标策略经济性方面的优势。