【作者】 冯怀玉；

【导师】 李国庆；

【作者基本信息】 东北电力大学 ， 电力系统及其自动化， 2011， 硕士

【摘要】 对于一个大型的互联电力系统,其区域间的可用输电能力不仅是电力系统安全、可靠运行的一个重要技术指标,也是作为引导电力市场交易,充分利用现有输电资源、防止输电阻塞等方面的指挥棒。随着人们对自然资源的保护意识的日益增强和土地使用的限制以及电网中的各种负荷水平日益增大等因素。因此如何准确、快速地计算出电力系统区域间的可用输电能力,使系统在满足安全性和可靠性约束条件的前提下,最大限度地满足各区域的用电负荷需求以及最大程度的提高其经济性,这一问题已经成为电力系统研究人员所关注的重要课题。本文建立了基于最优潮流并计及系统静态安全约束条件的可用输电能力计算模型,同时尝试采用一种源于大自然新兴的仿生类智能优化算法——模拟植物生长算法求解该模型。该算法是以植物的向光性机理作为寻优机理模式,其核心思想就是植物的生长环境是由可用输电能力问题的可行域形成,而植物进行光合作用的光源则是由全局最优解来充当,这样就生动的模拟了真实植物的向光性原理,同时也建立了植物枝叶在不同的光线强度环境下快速向光源生长的动力机制。该算法采用将目标函数和约束条件分开处理,且无需将约束条件以罚函数的形式并入目标函数中;对初始点的选取也没有过多的要求;兼具方向性和随机性的搜索机制等特点。通过引入自适应动态调整步长机制,提高解的准确性,增强算法的全局寻优能力。以IEEE-30节点标准系统为例对其进行建模,并采用Matlab7.1进行编程仿真计算,将计算结果与Benders分解法、改进粒子群算法进行比较分析,表明该算法具有较强的全局寻优能力和较快的收敛速度,从而验证了该算法用于可用输电能力计算的合理性和有效性。更多还原

【Abstract】 In the an interconnected power system, Available Transfer Capability is not only a technical target to meet safe and reliable operation, but it is also a guide to guide the power market transaction,be full use of existing transmission resources and prevent transmission congestion.With increasing awareness of the protection of natural resources and land availability consideration and the load level of modern power systems has become more and more serious. Therefore, the problem of how to accurately and efficiently calculate the inter-regional Available Transfer Capability and make the system to meet safety and reliability constraints, and to meet the needs of load to the greatest extent, which has caused widespread concern of system researchers.In this paper, the Available Transfer Capability model is built based on Optimal Power Flows, and state safety constraints are considered. Also,a kind of new bionic intelligent optimization algorithm which occurs in nature is trying to solve this model.-Plant Growth Simulation Algorithm. The algorithm regards the plant phototropism mechanism as optimization model,its core idea is that it considers the feasible region of ATC as plant growth environment,and it regards the global optimal solution as the light sourse for photosynthesis of plants.In this way,the principles of true plant phototropism can be vividly imitated,establishing the dynamic mechanism that plant leaves rapidly growth to the light at the different light intensity. Plant Growth Simulation Algorithm with the advantages of parallel search mechanism, easy to deal with constrained optimization problems;and using the objective function and constraints separately, and without the constraints of the form of a penalty function into the objective function. Therefore,it is a simple calculation model and has few demands to the selection of initial points and has strong global optimization capability, good robustness, simple principle and easy implement etc.The introduction of adaptive step size mechanism is made better to improve the accuracy of solutions and global convergence performance of the algorithm.The verification results by IEEE-30 bus system show the algorithm has the stronger global convergence performance and the quicker convergence speed compared with the results gained by benders decomposition and improved particle swarm optimization (IPSO), and it is proved to be rational and available.更多还原