【作者】 田文奇；

【导师】 和敬涵；

【作者基本信息】 北京交通大学 ， 电气工程， 2013， 博士

【摘要】 电动汽车以其绿色环保、节能高效的特点取代传统燃油汽车已成为低碳经济发展的必然趋势。随着电动汽车的推广普及,规模化电动汽车在时间和空间上的无序随机充电行为将增加电网运行中的不确定影响因素,给电网的经济调度和运行控制带来新的挑战。因此,有必要根据电动汽车充电的功率需求和能量需求特性,研究电动汽车有序充电调度方法,减小规模化电动汽车充电对电网的影响,同时提高充电设施的运行效率和电动汽车用户充电的便利性。本文围绕电动汽车换电站有序充电调度方法,基于时间尺度、空间尺度、时空双尺度开展研究,主要工作如下：(1)电动汽车在时间上的无序随机充电行为可能会使电网负荷出现“峰上加峰”的现象,本文利用充电负荷在时间上的可调度性,以改善电网负荷曲线、降低换电站购电成本为目标,在满足车辆换电需求的前提下,从时间尺度出发提出了电动汽车换电站有序充电时间调度方法。方法考虑了约束条件的随机性,采用粒子群及其变异算法求解有序充电时间调度数学模型,对换电站的日充电计划进行优化。基于奥运换电站实际运行数据的仿真算例表明该方法能够使换电站在满足车辆换电能量需求的同时避开电网负荷高峰期充电,达到避峰填谷、降低换电站运行成本的目的。(2)电动汽车用户从自身角度出发对换电位置的选择具有随机性和盲目性,可能会增加用户换电成本且不利于电网和换电站的经济运行,本文以同时满足用户侧和充电设施侧充电负荷空间分配要求为目标,对电动汽车用户换电位置进行引导,从空间尺度出发提出了电动汽车换电站有序充电空间调度方法。采用粒子群及遗传算法对比求解有序充电空间调度数学模型。基于深圳充电站实际运行数据背景的仿真算例以及基于校园配电网百辆车示范运行项目的算例表明该方法既能够满足电动汽车用户对换电位置的要求,又能够促进电网经济运行,提高换电站充电设施利用率。(3)为了实现同时在时间和空间两个尺度上对电动汽车充电行为进行引导和调度,提出了电动汽车换电站有序充电时空联合调度方法。方法以换电站日计划充电电量为联接变量,将实时的有序充电空间调度作为联合调度方法优化循环的内环,将非实时的有序充电时间调度作为联合调度方法优化循环的外环。基于校园配电网百辆车示范运行项目的仿真算例验证了方法的有效性,实现了时空两个尺度的引导和调度。更多还原

【Abstract】 Electric vehicle (EV) with its environmental protection, energy saving and high efficiency features to replace the traditional fuel vehicles has become the inevitable trend of the development of low carbon economy. With the popularization of electric vehicle, the out-of-order random charging behavior of large-scale EV users in time and space will increase the uncertain influence factors in power grid operation and bring new challenges to power grid economic dispatch and operation control. Therefore, it is necessary to research EV coordinated charging method according to the EV charging power and energy demands characteristics to reduce the large-scale EV charging effects on the grid and improve the efficiency of charging infrastructure and the charging convenience of EV users.The dissertation focus on EV battery swapping station coordinated charging dispatch method based on temporal dimension, spatial dimension and temporal-spatial dimensions. The main works and specific researchs are summarized as follows:(1) Out-of-order random charging behaviors of large-scale EV users in time will increase power grid peak load. Based on meeting charging demand of electric vehicles and the charging load’s schedulability at the time, this paper proposed EV battery swapping station (BSS) coordinated charging temporal dispatch method from the time dimension to improve power grid load curve and reduce the charging cost of BSS. With the stochastic constraints particle swarm optimization algorithm and its improved algorithm are taken to solve coordinated charging temporal dispatch model and achieve BSS optimized daily charging plan. Simulation example based on operation data of Olympic Charging Station show that the method can make EV charge in load valley period and meet EV energy demand at the same time. Economy of BSS operation is improved and peak load is shifted.(2) The choices of electric vehicle users for charging position from their own point of view are blind and random and may increase the users’ changing cost. It is not conducive to the economic operation of power grid and BSS.In order to satisfy the user’s and BSS’s charging load space allocation requirements, this paper proposed EV battery swapping station coordinated charging spatial dispatch method from the spatial dimension to guide EV user’s charging position. Particle swarm optimization algorithm and genetic algorithm are taken to solve coordinated charging spatial dispatch model contrastively. Simulation example based on the converted actual operation data of Shenzhen charging station and campus grid100vehicles demonstration project show that the method can meet the EV user’s charging position requirements and promote the the power grid economic operation and BSS utilization rate.(3) In order to achieve EV charging behavior guidance and scheduling both in temporal and spatial double dimensions at the same time, BSS coordinated charging temporal-spatial dispatch method is proposed. The method joined temporal dispatch and spatial dispatch together on the BSS daily plan charging energy. Real-time coordinated charging spatial dispatch as temporal-spatial dispatch method’s inner loop optimization and non real-time coordinated charging temporal dispatch as temporal-spatial dispatch method’s outer loop optimization. Simulation examples based on campus grid100vehicles demonstration project demonstrate the effectiveness of the proposed method. EV charging behavior guidance and scheduling are realized from the dimensions of time and space.更多还原