电动汽车充换电站的优化控制策略研究

发布时间：2018-01-17 14:07

本文关键词：电动汽车充换电站的优化控制策略研究　出处：《山东大学》2015年硕士论文　论文类型：学位论文

【摘要】：为了应对能源与环境危机,电动汽车产业在我国得到了迅猛发展。电动汽车电能补给配套建立的充换电站,成为电网中的一种新的负荷类型,其特点在于具有灵活的电力需求特性。由于电动汽车电池的储能效用,充换电站兼顾了电源与负荷的双重属性,充分挖掘电动汽车充换电站所具有的电能存储功能,利用其柔性、灵活的负荷特性,做出合理的充换电决策,不仅有利于对电力系统负荷波动的平抑,更可以在紧急情况下,为电力系统提供重要的备用支持。因此,研究充换电站与电网运行相协调的优化控制策略具有十分重要的现实意义。要实现充换电站充放电的调控目标,充换电站的能量预测与基于预测结果的优化决策是必不可少的环节。其中,充换电站的负荷预测可以通过对电动汽车的行驶规律以及充换电阈值的研究间接实现,也可以利用充换电站历史运行数据通过规律挖掘手段直接建模实现。而对于充换电站的优化决策,一般可以以充换电站的运行经济性或电网的运行安全性为决策目标,并综合考虑电池容量约束、充电桩功率约束等约束条件,建立充换电站的优化调控模型,得到充换电站充放电功率的优化决策结果。从目前研究现状来看,充换电站的负荷预测及优化控制尚存在以下几个问题：①目前,充换电站负荷预测多采用确定性的预测方法,所得充换电站功率需求序列难以反映预测结果的不确定性,无法为制定鲁棒性的充电策略提供必要的决策依据。②现有充换电站的充换电策略优化多以连续的充电功率为控制变量,然而,受电池荷电状态及电池数量的限制,此种模型给出的最优解往往难以与现实对应。③对电池交换模式下的风、光与充换电站协同运行策略优化的研究较少。在上述背景下,本文首先在充换电站换电需求状态空间离散化的基础上,使用马尔科夫预测方法对充换电站换电需求进行了离散状态概率预测,其所得的换电需求状态的概率分布可为充换电站的运行决策提供更为全面的决策依据；其次,文章提出了基于日前分时电价的双向能量交换下的以电池充放台数为控制变量的充换电站两层规划方法,其中,第一阶段以充换电站的运行经济性为优化目标,第二阶段则以第一阶段求取的最小费用为约束,以满电电池台数最大为目标进行优化,优化过程中考虑了负荷需求的区间不确定性；最后,在分析充换电站电池交换意愿的基础上,文章提出了基于物流网的,以电动汽车充换电站与风电场侧储能系统联合收益最大化为目标的充换电站电池交换模型。所提出的模型与方法均已完成实际系统构建,测试运行效果验证了本文所提出方法的有效性。
[Abstract]:In order to deal with the crisis of energy and environment, the electric vehicle industry has developed rapidly in China. Because of the energy storage utility of electric vehicle battery, the charging power station takes into account the dual properties of power supply and load. Fully excavating the electric energy storage function of the electric vehicle charging and replacing power station, making reasonable charging and switching decision by using its flexible and flexible load characteristics is not only conducive to the stabilization of the load fluctuation in the power system. It is also possible to provide important backup support for power systems in emergency situations. It is of great practical significance to study the optimal control strategy of the charging and changing power station in coordination with the power network operation. The aim of charge and discharge regulation and control of the charging and changing power station should be realized. The energy prediction and the optimal decision based on the prediction results are essential links. The load forecasting of charging power station can be realized indirectly by studying the driving law of electric vehicle and the threshold of charging and switching. It is also possible to use the historical operation data of the charging and changing power station to model the model directly through the rule mining method, and to optimize the decision of the charging and changing power station. Generally, the optimal control model of charging power station can be established by taking the economical operation of charging power station or the operation safety of power grid as the decision goal, and considering the constraints of battery capacity and charging pile power, etc. The optimal decision results of charging and discharging power are obtained. According to the current research situation, the following problems exist in the load forecasting and optimal control of the charging and changing power station. The deterministic forecasting method is often used in the load forecasting of recharge power station, and it is difficult to reflect the uncertainty of the forecast result in the power demand series of the recharge and replacement power station. Can not provide the necessary decision basis for the robust charging strategy. 2. The current charging and switching strategy optimization takes the continuous charge power as the control variable, however. Limited by the charging state and the number of batteries, the optimal solution given by this model is often difficult to correspond to the wind in the switching mode of the battery in the actual situation. There are few researches on the optimization of cooperative operation strategy between light and charging power station. Under the above background, the state space of power exchange demand is discretized firstly in this paper. The discrete state probabilistic prediction of the power exchange demand of the recharge power station is carried out by using Markov prediction method. The probability distribution of the state of the power exchange demand can provide a more comprehensive decision basis for the operation decision of the charging and switching power station. Secondly, this paper puts forward a two-layer planning method for charging and discharging power plants based on the bidirectional energy exchange with the number of battery charging and discharging stations as the control variable, which is based on the time-sharing price before the day. The first stage takes the operation economy of the charging and replacing power station as the optimization goal, the second stage takes the minimum cost obtained in the first stage as the constraint, and the maximum number of full battery stations as the goal. The interval uncertainty of load demand is considered in the optimization process. Finally, on the basis of analyzing the willingness of battery exchange in charged power station, the paper puts forward a logistics network based on it. The battery exchange model of electric vehicle charging power station and wind farm side energy storage system is aimed at maximizing the combined income. The proposed model and method have completed the actual system construction. The test results verify the effectiveness of the proposed method.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U469.72;TM715

【共引文献】