【作者】 汪超；

【导师】 王昕；

【作者基本信息】 上海交通大学 ， 电工理论与新技术， 2011， 硕士

【摘要】 电力系统无功补偿与无功平衡,是保证电压质量的基本条件,对保证电力系统的安全稳定与经济运行起着重要作用。为此,需要对电网进行合理的无功电源规划,选择合适的目标函数和控制手段,用优化方法制定无功规划方案。本文首先介绍了无功功率对配电网电压损耗、有功网损以及功率因数的影响,说明无功补偿的重要性。随后,本文以有功损耗功率最小作为目标函数,将节点电压越限和发电机无功出力越限作为罚函数,建立无功补偿在配电网中优化配置的数学模型。针对上述多目标优化问题,提出了基于黄金分割的混沌粒子群算法进行求解。该算法通过黄金分割评判准则,按照适应度的高低,将粒子群中的粒子分成标准粒子和混沌粒子两类,同时解决了粒子群优化过程中容易陷入局部最优和混沌算法重复搜索部分解的问题,从而可以更有效的搜索到全局最优解,成功地提高了无功优化问题的求解速度,使算法能更好地适应问题的求解。在此基础上,为了减少大量混沌计算带来的冗余度,结合云理论,提出了基于黄金分割的云混沌粒子群算法。该算法利用云理论的快速搜索能力,将粒子分成标准粒子、云粒子和混沌粒子,从而克服了原算法大量运算带来的冗余度增加的问题,有效地提高了算法的速度。最后,为了进一步解决混沌算法计算量大的问题,本文提出了基于黄金分割的混沌云粒子群算法。该算法引入云理论将上述混沌粒子进一步改进为混沌云粒子,从而最大程度地克服混沌算法的冗余度问题,仿真结果证明了该算法的有效性。更多还原

【Abstract】 Reactive power compensation and reactive power balance in the power system is one of the basic conditions to ensure the voltage quality. It plays an important role in guaranteeing the power system’s safe, stable and economy operation. In this respect, reactive power planning on the grid is needed. To realize this purpose, selecting the appropriate objective function, optimizing reactive power plan with the optimization method is very important.First, the impacts of voltage loss, active power, loss power factor which is caused by the reactive power will be described to illustrate the importance of reactive power compensation. Then, a mathematical model in the distribution network will be established. Its objective function is designed to be the cost of active power loss of the system, in which the node voltages beyond limited and the generator reactive power output beyond limited are considered in the way of penalty function.In order to calculate the model, a new method called the Golden section Chaotic Particle Swarm Optimization (GCPSO) will be proposed. According to the level of fitness, it adopts the golden section principle to divide all the particles into standard particles and chaos particles, which solves the problems of easily falling into local optimum if only using PSO and repeat searching part of the solution if only using chaotic. So it is more effective in searching the global optimal solution and faster in calculation than the PSO method.Additionally, in order to reduce the redundancy caused by chaos, based on cloud theory, a new algorithm called the Golden section Cloud-Chaos Particle Swarm Optimization (Cloud-Chaos GPSO) will be presented. Utilizing the fast searching capability of cloud method, it divides all the particles into standard particles, cloud particles and chaos particles. The cloud particles are designed to replace parts of the chaos particles to overcome the issue of redundancy and improve the speed effectively.Finally, in order to solve the problem of the redundancy caused by chaos better, a new algorithm called the Golden section Chaos-Cloud Particle Swarm Optimization (Chaos-Cloud GPSO) will be presented. It improves the chaos particles above into chaos-cloud particles to overcome the problem of redundancy furthermore. The simulation results show the effectiveness of the algorithm.更多还原