基于自适应人工鱼群算法的微电网优化运行的研究

发布时间：2018-06-15 07:01

本文选题：微电网 + 自适应人工鱼群算法　；参考：《西安理工大学》2017年硕士论文

【摘要】：微电网中包含多种分布式电源,可以灵活地运行于并网状态和离网状态,有效地解决了新能源发电接入大电网的中出现的问题。但是,分布式电源的多样性和组合的随机性导致了微电网的运行控制难和运行成本高等问题,其中,微电网的优化运行可以降低发电成本和环境污染费用,提升能源利用效率,对微电网系统的经济、可靠运行和环保至关重要,因此,研究微电网的优化运行有十分重要的意义。本文首先介绍了微电网发展的背景和意义,分析了国内外微电网及其优化运行的研究现状,阐述了典型微电网系统的结构特点和工作原理。详细地分析了微电网中的主要微电源:风力发电机、光伏电池、微型燃气轮机、燃料电池及储能系统的运行特性和成本函数,为建立微电网多目标优化运行模型奠定基础。其次,针对基本人工鱼群算法容易陷入局部最优解、收敛速度慢和寻优精度低的缺陷,提出一种改进的自适应人工鱼群算法方法:主要改进了算法觅食行为的视野和步长,人工鱼同时按两种视野进行搜索,确定两个目标位置后,计算两个目标的食物浓度并与当前位置的食物浓度进行比较,选择向食物浓度低的位置移动。通过三个局部极值突出的函数测试,表明改进的算法要优于丛本的人工鱼群算法。针对含风、光、储、微型燃气轮机和燃料电池的并网型和离网型微电网,综合考虑发电成本(燃料成本、投资折旧成本、设备维护成本、并网时的电能交互成本)和环境成本,给出了在求解该模型时的安全运行的约束条件,使系统在一个优化周期内的总运行成本最低,建立了微电网多目标优化数学模型,分别制定出微电网在并网运行时、孤网运行时分时段优化运行策略,采用约束优化自适应人工鱼群算法(AAFSA)对建立的数学模型进行求解,求得一个优化周期内各分布式电源的最佳出力及运行总成本,并与基本人工鱼群算法求得的结果作对比。仿真结果表明:改进的算法具有避免陷入局部最优解、收敛速度快和精度高的特点。
[Abstract]:The microgrid contains a variety of distributed power sources, which can operate flexibly in grid-connected and off-grid states, thus effectively solving the problem of new energy generation connected to large power grid. However, the diversity of distributed generation and the randomness of combination lead to the difficulty of operation control and high operating cost of microgrid. Among them, the optimal operation of microgrid can reduce the cost of generation and environmental pollution, and improve the efficiency of energy utilization. It is of great importance to the economy, reliable operation and environmental protection of microgrid system, so it is very important to study the optimal operation of microgrid. This paper first introduces the background and significance of the development of microgrid, analyzes the research status of microgrid and its optimal operation at home and abroad, and expounds the structural characteristics and working principle of typical microgrid system. The operating characteristics and cost function of the main micro-power sources in microgrid, such as wind turbine, photovoltaic cell, micro-gas turbine, fuel cell and energy storage system, are analyzed in detail, which lays the foundation for the establishment of multi-objective optimal operation model of microgrid. Secondly, aiming at the defects of the basic artificial fish swarm algorithm that it is easy to fall into the local optimal solution, the convergence speed is slow and the optimization accuracy is low, an improved adaptive artificial fish swarm algorithm method is proposed: the visual field and step size of the algorithm foraging behavior are mainly improved. The artificial fish searched according to two kinds of visual fields at the same time, determined the two target locations, calculated the food concentration of the two targets and compared with the food concentration of the current position, and chose to move to the position where the food concentration was low. The results show that the improved algorithm is better than the artificial fish swarm algorithm. For grid-connected and off-grid microgrids containing wind, light, storage, micro gas turbines and fuel cells, the cost of generating electricity (fuel cost, depreciation cost of investment, maintenance cost of equipment, interactive cost of electricity when connected to the grid) and environmental cost are considered synthetically. The constraint conditions for the safe operation of the system in solving the model are given. The total operating cost of the system is the lowest in a single optimization period. The multi-objective optimization mathematical model of the microgrid is established, and the microgrid in grid-connected operation is worked out respectively. In this paper, the algorithm of adaptive artificial fish swarm algorithm (AAFSAA) is used to solve the mathematical model of the isolated network. The optimal output and total operating cost of each distributed power source in an optimized period are obtained. The results are compared with those obtained by the basic artificial fish swarm algorithm. The simulation results show that the improved algorithm has the advantages of avoiding falling into local optimal solution, fast convergence speed and high precision.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM732

【参考文献】