含多种分布式能源的微电网优化及控制策略研究

发布时间：2018-01-06 16:11

本文关键词：含多种分布式能源的微电网优化及控制策略研究　出处：《上海交通大学》2014年博士论文　论文类型：学位论文

【摘要】：利用微电网能够有效地将分布式电源、电力负荷和能量储存装置融合成一体，，有利于可再生能源的接入，充分发挥分布式电源的效益，是未来电网实现高效、环保和优质供电的重要手段。作为大电网的有益补充，微电网一经提出便引起世界各国的广泛关注。本文沿着规划、调度和控制的思路，研究含多种分布式能源的微电网优化问题。主要研究内容包括以下几部分： 1）对含风力发电单元的微电网进行可控分布式电源（Controllable DistributedGeneration，CDG）优化配置。提出基于损耗灵敏因子和电压稳定指标的CDG候选安装位置集合的确定方法。以投资成本、燃料成本、维护成本、购电成本及排放成本之和最小为目标函数，建立考虑负荷因素（负荷类型、负荷水平及不确定性）和能源政策影响的CDG规划模型，引入表示风电出力和负荷功率的随机变量，将运行约束条件描述成概率的形式，采用基于半不变量和Gram-Charlier级数展开理论的微电网随机潮流方法检验规划方案是否满足机会约束条件，应用人工蜂群算法搜索最优配置方案，从而实现CDG类型、位置和容量的同时优化规划。 2）从需求侧响应视角研究含电动汽车换电站（Battery Swap Station，BSS）的微电网经济调度问题。建立基于双层优化的微电网经济调度模型：上层模型的决策机构为微电网调度中心，其目标函数是微电网供电成本最小；下层模型的决策机构为BSS运营商，其目标函数是BSS利润最大；上下层优化模型通过微电网向BSS提供的个性化电价实现互动。采用入侵杂草算法和CPLEX软件分别对上下层优化问题进行迭代求解，得到微电网内部可控微电源的输出功率和个性化电价，达到微电网和BSS双方共赢目的。 3）从供能角度研究含多种分布式能源的微电网优化调度问题。以日电量成本最小、日温室气体排放量最少和全天中最大节点电压偏差最小为目标函数，建立基于机会约束规划的微电网多目标动态调度模型，不仅考虑优化变量在相邻时段间的内在联系，还将考虑风电出力和负荷功率的不确定性，允许调度方案以一定的置信水平满足节点电压和支路功率约束。提出基于随机潮流的克隆选择算法求解该多目标优化模型。在模型求解过程中，采用固定步长和变步长相结合的混合变异策略来提高解的多样性，通过反复克隆、变异、选择等操作，得到满足约束条件的Pareto最优解集，采用模糊隶属度计算总体满意度，解决Pareto最优解的选择问题。 4）对交流微电网的运行控制策略进行研究。以风电机组-燃料电池-并联型蓄电池组所构成的交流微电网为研究对象，根据上述微电源特点分别设计微电网孤岛和并网运行时的控制策略。研究微电网孤岛运行问题时，对并联型蓄电池组采用电压源逆变器控制，使用电压-相角下垂控制以实现按预定比例分配负荷功率，共同为微电网提供电压和频率支撑。研究微电网并网运行问题时，蓄电池组的控制方法由电压源逆变器控制转换为PQ控制，按照设定值输出有功功率和无功功率。燃料电池在两种模式下均采用PQ控制，孤岛运行时其工作状态取决于蓄电池组的荷电状态。在对微电源网侧变换器进行控制时，采用滑模电流内环替代传统PI电流内环，以提高系统控制性能。 5）对孤立直流微电网的运行控制策略进行研究。以光伏阵列-燃料电池-超级电容所构成的低压单极型孤立直流微电网为研究对象，针对采用集中供电方式和分散供电方式的微电网，分别提出考虑光伏阵列、燃料电池和超级电容特性的集中控制策略和分散控制策略。其中，应用扰动观测法实现光伏阵列的最大功率跟踪；利用斜率限制器限制燃料电池输出功率的变化速度，阻止其免受“燃料饥饿”的危害。在集中控制策略中，采用滑模电流内环控制器实现超级电容的快充快放，稳定直流母线电压；在分散控制策略中，利用直流母线电压信号实现微电网不同运行状态之间的自动切换。通过相关算例分析，验证了本文所提出的微电网优化模型和运行控制策略的合理性，以及优化算法的有效性。
[Abstract]:The use of micro grid can effectively distributed power supply, power load and energy storage devices together, have access to renewable energy, give full play to the benefits of distributed power, is an important means to achieve high efficiency of power grid in the future, environmental protection and quality of power supply. As a useful supplement to the power grid, microgrid once it has caused widespread concern all over the world. The planning, scheduling and control method, micro grid optimization with the distributed energy. The main research contents include the following parts:
1) on micro grid power generation unit with wind of controllable distributed power supply (Controllable DistributedGeneration CDG) optimization method is proposed. To determine the position of the CDG candidate loss sensitivity factor and voltage stability index based on set. With investment costs, fuel costs, maintenance costs, purchase of electricity and the minimum cost and emission cost as the objective function considering the factors (load, load type, load level and uncertainty) CDG programming model of impact and energy policy, said the introduction of random variables of wind power output and load power, will run the constraints described in the form of probability, whether the semi invariant and Gram-Charlier series based on the theory of micro grid power flow method, random inspection planning plan to meet the constraints condition of opportunity, application of artificial bee colony algorithm to search the optimal allocation scheme, so as to realize the CDG type, location and capacity of the same Optimal planning.
2) from the perspective of research of electric vehicle charging station from the demand side response (Battery Swap, Station, BSS) of the micro grid economic dispatch problem. The establishment of micro grid economic dispatch model based on bi level optimization: the decision-making body of the model of the micro grid dispatching center, the objective function is the minimum power supply micro cost; decision structure model for BSS operators, the objective function is BSS the maximum profits; the realization of interactive optimization model through personalized price micro grid provided to BSS. The invasive weed algorithm and CPLEX software respectively on the upper and lower optimization iterations, the output power of micro grid internal controllable micro power and micro grid and achieve personalized pricing BSS the two sides win-win goal.
3) from the energy supply of micro grid scheduling problem of containing a variety of distributed energy. With the minimum daily electricity cost, on greenhouse gas emissions and at least in the all day long maximum node voltage deviation minimum as the objective function, the establishment of micro grid scheduling model based on chance constrained programming, not only consider the optimization variables in adjacent relation between period of time, will also consider the wind power output and load uncertainty, allowing the scheduling scheme with a certain confidence level to meet the node voltage and branch power restriction. The trend of the clones were randomly select the multi-objective optimization algorithm to solve the model. Based on the model, the hybrid mutation strategy of fixed step and variable step length combined to improve the diversity of the solution, through repeated cloning, mutation, selection and other operations, Pareto get the optimal solution set, using fuzzy membership The degree of degree of satisfaction is calculated to solve the choice problem of Pareto optimal solution.
4) research on operation control strategy of AC microgrid. The wind turbine fuel cell - battery composed of parallel AC micro grid as research object, design control strategy of microgrid islanding and grid operation according to the characteristics of micro power. Research on Microgrid when using voltage source the inverter control of the shunt battery, using voltage amplitude and phase angle droop control to realize distribution according to the predetermined proportion of load power, to provide support for the voltage and frequency of micro grid. The research of micro grid operation, control method of battery by voltage source inverter control into PQ control, according to the set value of output power and no reactive power. Fuel cells are controlled by PQ in two modes, island operation depends on the working state of the battery state of charge. In the micro power grid side converter When the device is controlled, the inner loop of the sliding mode current is replaced by the inner loop of the traditional PI current to improve the control performance of the system.
5) of operation control strategy for isolated DC micro grid. Low voltage DC unipolar isolated to photovoltaic array - fuel cell and super capacitor composed of micro grid as the research object, the micro grid centralized power supply mode and decentralized power supply mode, are proposed considering the centralized control strategy of photovoltaic array, fuel cell and super capacitor the characteristics and application of decentralized control strategy. The disturbance to achieve maximum power tracking of PV array observation method; velocity change by slope limiter to limit the output power of the fuel cell, to prevent it from burning hazardous material hunger ". In the centralized control strategy, using sliding mode current loop controller to achieve fast charge of super capacitor fast, stable DC bus voltage; in the decentralized control strategy, realize the automatic cutting between different micro grid operation state for using the DC bus voltage signal.
Through the analysis of the related examples, the rationality of the optimization model and the operation control strategy proposed in this paper and the effectiveness of the optimization algorithm are verified.

【学位授予单位】：上海交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM727

【参考文献】