电动汽车与电网双向互动下含风电场的动态经济调度研究

Based on the energy saving and emission reduction policies, it is of important theoretical and practical significance to present a model to reduce the power generation costs for the bidirectional complementary coordinating economic dispatch considering electric vehicles to wind power integrated grid. However, electric vehicles and the random factors of the wind power are extremely complex, and the related models and methods cannot meet with the practical needs of the economic optimal dispatch for the grid with electric vehicles and wind power. Therefore, considering the agent, this project aims to establish the bidirectional complementary coordinating economic dispatch models considering electric vehicles to grid, and aims to propose a tight relaxation method and parallel interior point method to solve the presented models. Using the ARMA model and GARCH model, this project is to establish the forecasting models for the wind speed and the wind generation output, and this project is to optimize wind power allocation according to an improved mean-variance model based on Copula function. On the basis of the wind power prediction and allocation, combining with the electric vehicles to grid and wind power, this project aims to build a chance constrained programming model for the dynamic economic optimal dispatch and aims to present an extended cutting plane method for the proposed model. This project is to present economic optimal dispatch models in electric vehicles and wind power integrated system and to study the solution methods, so as to provide scientific basis in formulating the synthetically strategy to exploit the wind, electric vehicles and thermal power, and to expand the application of the optimization theory and methods in economic dispatch.

基于节能减排政策要求，建立更为准确的电动汽车与电网双向互动下含风电场的经济调度模型，以显著降低发电成本，具有重要的理论和实际意义。然而，风电和电动汽车充放电的随机性和复杂性，导致现有相关模型与方法无法满足风-车并网经济调度的实际需要。为此，本项目结合代理商机制，建立电动汽车与电网双向互动下的优化调度模型，提出求解模型的紧松弛法和并行快速内点法；利用ARMA和GARCH模型建立与实际更加吻合的风速与风电功率预测模型，并结合Copula函数建立改进均值-方差模型，进行多风场风电优化分配；结合风电预测和优化分配，建立基于机会约束规划的考虑风-车并网的动态经济调度随机模型，并提出求解模型的广义割平面法。本项目致力于建立电动汽车与电网双向互动下含风电场的经济调度的理论模型，并深入研究模型的有效求解算法，以期为制定切实有效的风-车-火电综合利用策略提供科学依据，拓展优化理论与方法在经济调度中的应用。

本课题以节能减排政策为前提，建立V2G环境下含随机因素的发电经济调度模型，设计模型的快速有效求解方法，从而显著降低发电成本。围绕这一目标，本课题首先基于混合整数规划建立了V2G环境下电动汽车充放电与火电机组组合优化模型；其次，为进一步提高机组组合模型的求解效率，我们采用投影技术消除半连续变量，提出了一种不同于传统机组组合建模的全新的2类状态变量的机组组合模型，所提的更为紧和简洁的2类状态变量模型的求解比其他经典模型更为有效；再次，提出一种基于多cuts和自适应终止的求解机组组合问题的外逼近方法，所提算法相比经典外逼近算法能够以更少的迭代次数和计算时间获得相当质量的次优解；最后，采用CVaR来处理电力市场环境下一系列不确定因素，提出了区域能源提供商的风险厌恶型运行调度模型，并采用所提模型分析智能电网各参与因素与电价波动的相互作用和影响，该模型同时考虑了V2G与电网双向互动、需求响应、排放权交易、随机电价等因素。为有效求解所提的基于复杂混合整数二次约束二次规划调度模型，提出一种紧的多割外逼近方法，通过交替求解混合整数线性规划主问题和非线性规划子问题从而逼近原问题最优解。.本课题的理论研究成果不仅可用于分析处理智能电网、能源互联网中的考虑不确定性的各类调度问题以有效规避风险，也为信息网络在考虑不确定性因素的优化调度研究开拓了新思路。.围绕本课题的研究，发表学术论文11篇，其中SCI收录6篇；向SCI投稿论文3篇，其中一篇正在按照专家审稿意见进行修改；申请发明专利2项；获得软件著作权登记13项；培养硕士研究生2名。