数模混合主动配电网的能量优化管理与电池经济性评价

发布时间：2018-12-10 21:32

【摘要】：随着传统能源的枯竭,以微网为载体的可再生能源发电系统由于其灵活、分散、小型、靠近用户和使用清洁能源等特点得到了快速发展。储能系统接入微网,不仅可以平抑可再生能源出力的波动,提高电能质量,还能通过适时吸收和释放功率,移峰填谷,改善电网性能,降低用户的购电成本。合理地进行储能电池的选型与充放电控制,对提升微网运行效率,降低微网运行成本,起着关键作用。本文研究了一种基于模型预测控制的微网能量管理优化策略,以北京交通大学电气楼配网为研究对象,搭建了主动配电网半实物半数字仿真平台,使用电池模型搭建虚拟的储能系统,以微网的月运行成本最小为目标,进行能量管理运行模式和储能配置方式的选择。主要研究内容如下:(1)研究并建立了锂离子电池和铅酸电池的等效电路模型。使用循环次数等效方法评估电池的性能衰减,并对模型的动态特性和循环性能衰减进行了仿真验证,仿真结果验证了模型的精确度及性能衰减指标的正确性。然后,考虑电池运行过程中的实际衰减,建立了储能系统的成本函数,用于储能系统的经济性评估。(2)研究了一种基于模型预测控制,按最大需量计费的微网能量管理优化策略,考虑微网中的负荷需求、各分布式电源的运行成本和约束、储能系统的功率和电量约束等条件,以运行成本最小为优化目标,使用混合整数线性规划模型滚动求最优解。根据得到的最优解对微网中的负荷、储能、分布式发电单元进行优化控制,从而有效提高配网的运行效率,降低运行成本。(3)使用Matlab/Simulink和Labview软件搭建主动配电网半实物半数字仿真平台,该平台融合虚拟的储能系统、能量管理算法计算模块、气象数据采集模块、配网功率数据采集与监控等模块,用于能量管理策略的验证和储能系统经济性评估。(4)基于搭建的半实物半数字平台,使用前文研究的储能电池模型及成本函数、能量管理优化策略,基于北京交通大学电气楼主动配电网的真实环境和实测数据,进行算例验证。计算结果表明,本文研究的能量管理优化算法能够根据负荷需求、可再生能源出力、实时电价以及储能系统的剩余容量实时求解得到储能系统优化的充放电控制策略,有效的降低了微网运行成本。(5)通过对微网运行成本和储能系统的经济性分析,得到了最佳的“电池类型—能量管理运行模式”组合方式。计算结果表明,在微网储能领域,铅酸电池比锂离子电池具有更好的经济性。本文的主要创新点是搭建了半实物半数字的主动配电网实时仿真平台,能够模拟不同主动配电网的“能量管理策略—储能电池”组合方式,通过比较不同组合下的微网总运行成本,得出最优方案,指导微网的实际控制模式与储能配置方式,从而大幅降低实物测试的成本。
[Abstract]:With the depletion of traditional energy, the renewable energy generation system based on microgrid has been developed rapidly because of its flexibility, dispersion, small size, proximity to users and the use of clean energy. The energy storage system can not only reduce the fluctuation of renewable energy output and improve the power quality, but also can absorb and release power timely, shift peak and fill valley, improve the performance of power grid and reduce the cost of electricity purchase. Reasonable selection and charge / discharge control of energy storage battery plays a key role in improving the efficiency of microgrid operation and reducing the operating cost of microgrid. In this paper, an optimization strategy of microgrid energy management based on model predictive control is studied, and the hardware-in-the-loop simulation platform of active distribution network is built with Beijing Jiaotong University electrical building as the research object. The virtual energy storage system is built with the battery model. The energy management operation mode and the energy storage configuration mode are selected in order to minimize the monthly operation cost of the microgrid. The main contents are as follows: (1) the equivalent circuit models of lithium ion battery and lead acid battery are studied and established. The cycle number equivalent method is used to evaluate the performance attenuation of the battery. The dynamic characteristics and cycle performance attenuation of the model are verified by simulation. The accuracy of the model and the correctness of the performance attenuation index are verified by the simulation results. Then, considering the actual attenuation of the battery, the cost function of the energy storage system is established, which is used to evaluate the economy of the energy storage system. (2) A predictive control based on the model is studied. The optimization strategy of microgrid energy management based on maximum demand, considering the load requirement in microgrid, the operating cost and constraint of each distributed power supply, the power and power constraint of energy storage system, etc. The mixed integer linear programming model is used to find the optimal solution by rolling. According to the obtained optimal solution, the load, energy storage and distributed generation unit in the microgrid are optimized and controlled, which can effectively improve the efficiency of the distribution network. (3) using Matlab/Simulink and Labview software to build the hardware-in-the-loop simulation platform of active distribution network. The platform integrates virtual energy storage system, calculation module of energy management algorithm, meteorological data acquisition module. The power data acquisition and monitoring module of distribution network is used to verify the energy management strategy and evaluate the economy of energy storage system. (4) based on the hardware-in-the-loop digital platform, the energy storage cell model and cost function are used. Based on the real environment and measured data of the active distribution network of Beijing Jiaotong University, the optimization strategy of energy management is verified by an example. The calculation results show that the energy management optimization algorithm studied in this paper can solve the optimal charge and discharge control strategy based on load demand, renewable energy output, real time electricity price and residual capacity of energy storage system in real time. (5) by analyzing the operation cost of microgrid and the economy of energy storage system, the best combination of "battery type-energy management operation mode" is obtained. The results show that lead acid battery is more economical than lithium ion battery in the field of microgrid energy storage. The main innovation of this paper is to build a hardware-in-the-loop real-time simulation platform for active distribution network, which can simulate the combination of "energy management strategy-energy storage battery" in different active distribution networks. By comparing the total operating cost of microgrid with different combinations, the optimal scheme is obtained to guide the actual control mode and energy storage configuration of the microgrid, thus greatly reducing the cost of physical testing.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM73

【参考文献】