基于改进LHS含分布式能源的电力系统概率潮流计算

发布时间：2018-05-24 08:45

本文选题：概率潮流计算 + 拉丁超立方算法　；参考：《兰州交通大学》2017年硕士论文

【摘要】：随着化石能源的逐渐枯竭以及全球和国家低碳环保的政策需要,风力发电、太阳能发电等清洁低碳能源在新能源比例中越来越重。但是由于风能和太阳能具有较大随机性和间歇性,这些分布式能源的出力方式会导致随机性变化,特别是在大规模的新能源并网以后,例如发电机输出功率波动、负荷功率变化等对电网的安全运行造成的影响更加明显,这些存在问题对传统的潮流计算提出了新的挑战。传统潮流电力系统分析长期以来建立在确定性潮流计算的基础之上,但是在实际网路中,系统的参数,网络的拓扑结构、母线的负荷等都是不确定的值,这就需要考虑通过大量的计算,耗费时间成本。而且这很难反应整个体统的总体状况,传统的确定性潮流计算已无法满足上述问题,因此考虑随机因素影响的概率潮流计算法得到了发展和应用。研究概率潮流计算的前提是分析概率潮流所用到的算法,因此本文先分别分析了蒙特卡罗法,拉丁超立算法及其优缺点。在分析拉丁超立方算法的基础上,对该算法中排序步骤进行改进。提出一种基于随机行走原理的拉丁超立方算法,并将该方法应用在概率问题求解中,仿真结果证明了该算法所具有的优越性。以下是论文完成的主要内容:(1)本文针对目前常见概率算法进行研究分析,重点阐述目前算法存在的缺陷和不足,最后指出所改进方法对于解决所存在问题的必要性。(2)针对阐述拉丁超立方算法中存在的缺点,将随机行走算法引入要改进的算法中。本文通过MATLAB软件进行编程,以蒙特卡罗法为参考值,利用测试函数在加入含风电的IEEE-14节点和IEEE-118系统的条件下,比较基于随机行走算法的拉丁超立方算法(Random Walk Latin hypercube Sampling,RWLHS)和基于施密特正交化法的拉丁超立方算法(Gram-Schmidt Latin hypercube Sampling,GSLHS)两种算法,从而得出RWLHS算法的有效性。(3)其次将本方法引用到含光伏电场的电力系统中,建立了负荷、发电机和光伏电场的模型,通过改进的方法分析系统节点加入分布能源前后节点电压的变化和支路潮流的变化。
[Abstract]:With the depletion of fossil energy and the need of global and national low-carbon environmental protection policies, clean and low-carbon energy sources, such as wind power and solar power, are becoming more and more important in the proportion of new energy sources. But because wind and solar are more random and intermittent, the way these distributed sources of energy are produced can lead to random changes, especially after large-scale new sources of energy are connected to the grid, such as generator output power fluctuations. The influence of load power change on the safe operation of power grid is more obvious. These problems pose a new challenge to the traditional power flow calculation. The traditional power flow analysis is based on the deterministic power flow calculation for a long time, but in the actual network, the parameters of the system, the topological structure of the network and the load of the bus are all uncertain values. This requires consideration of time-consuming costs through a large number of calculations. And it is very difficult to reflect the overall situation of the whole system. The traditional deterministic power flow calculation can no longer meet the above problems, so the probabilistic power flow calculation method considering the influence of random factors has been developed and applied. The premise of studying the calculation of probabilistic power flow is to analyze the algorithms used in probabilistic power flow, so this paper first analyzes the Monte Carlo method, Latin superposition algorithm and its merits and demerits respectively. Based on the analysis of the Latin hypercube algorithm, the sorting steps in the algorithm are improved. A Latin hypercube algorithm based on random walk principle is proposed and applied to probabilistic problem solving. The simulation results show the superiority of the algorithm. The following is the main content of this paper: (1) this paper studies and analyzes the common probability algorithms, focusing on the shortcomings and shortcomings of the current algorithms. Finally, the necessity of the improved method for solving the existing problems is pointed out. (2) aiming at the shortcomings of the Latin hypercube algorithm, the random walk algorithm is introduced into the improved algorithm. In this paper, the MATLAB software is used to program, the Monte Carlo method is used as the reference value, and the test function is used under the condition of adding the IEEE-14 node and the IEEE-118 system with wind power. The Latin hypercube algorithm based on random walk algorithm, Random Walk Latin hypercube sampling RWLHS, and the Latin hypercube algorithm Gram-Schmidt Latin hypercube sampling GSLHSbased on Schmidt orthogonalization are compared. The validity of RWLHS algorithm is obtained. Secondly, the method is applied to the power system with photovoltaic electric field, and the model of load, generator and photovoltaic electric field is established. The changes of node voltage and branch power flow before and after adding distributed energy are analyzed by the improved method.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM744

【参考文献】