含分布式电源的配电网电压功率优化调整研究

发布时间：2018-06-11 23:38

本文选题：配电网 + 分布式电源　；参考：《昆明理工大学》2017年硕士论文

【摘要】：随着我国配电系统的快速发展,用电负荷呈现出不确定性与多样性,在配电网中则更容易引起电压质量问题。配电网作为电力系统中电压等级最低的部分与用户紧密联接,电压质量是作为电力系统正常稳定运行的重要保证,它代表着电力系统运行、维护和管理水平的高低,影响着工业生产质量和人民生活的水平。实际配电网中,负荷在时间特性上快速而不均衡的发展变化、在空间特性上随着季节性发展变化,经常导致局部地区出现电压不合格现象,而随着分布式电源的接入,在一定程度上既可减少系统网损,也可以治理局部电压偏低现象,例如,日负荷的高峰期,部分地区或因负荷过重、功率因数低而导致部分电压出现越下限的情况,若此区域接有分布式电源,利用分布式电源发出的无功来调整电压水平,使电压偏低现象得到治理。但在某种情况下,负荷的低谷期,全网出现电压越上限的情况,此时若有分布式电源接入,致使接入点产生大量的无功功率,最终导致电压越上限。因此,在含有分布式电源的配电系统中,利用分布式电源和其他的调节手段进行电压与功率的优化调整。鉴于分布式电源接入配电网会造成系统潮流的复杂变化,本文从分布式电源的节点类型、容量等级、空间分布、渗透率四个方面来分析分布式电源对系统的网损和电压造成的影响。节点类型通过将分布式电源划分为PI节点、PQ(V)节点、PV节点和PQ节点来反映;容量等级通过在同一节点接入不同容量的同类型分布式电源来反映;空间分布通过在不同节点位置接入相同容量的同类型分布式电源来反映;渗透率通过多个节点接入相同类型的分布式电源来反映。本部分对于分布式电源的潮流计算分析是通过牛顿-拉夫逊法在辐射型IEEE33节点系统上进行测试的,优化方法采用基因遗传算法,最后得出本文的研究目的之一—分布式电源对系统的网损和电压造成的影响。本文的优化类型为系统网损和电压偏差的多目标优化问题,通过分布式电源与有载调压变压器分接头、无功补偿装置进行优化系统网损与治理电压偏差。采用的方法为标量化优化算法和矢量化优化算法,对于标量化算法利用人工智能的基因遗传算法对多目标问题进行优化,得到优化结果、收敛性、计算时间等参数;对于矢量化算法,首先利用规格化法平面法先对多目标问题转化成单目标问题,然后利用跟踪轨迹内点法进行优化,求出帕累托优化解集,最后求出多目标优化的折中解。将两种方法在IEEE33节点、21节点系统上进行仿真验证,对比两种算法的求解时间、收敛度和优化结果等。
[Abstract]:With the rapid development of power distribution system in China, power load presents uncertainty and diversity, and voltage quality is more likely to be caused in distribution network. As the lowest part of the power system, the distribution network is closely connected with the users. The voltage quality is an important guarantee for the normal and stable operation of the power system. It represents the level of operation, maintenance and management of the power system. It affects the quality of industrial production and the standard of living of the people. In the actual distribution network, the load develops rapidly and unevenly in the time characteristic, and the spatial characteristic changes with the seasonal development, which often leads to the local voltage disqualification phenomenon, and with the access of the distributed power supply, To a certain extent, it can not only reduce the network loss of the system, but also control the phenomenon of low local voltage. For example, during the peak period of daily load, part of the area or because of excessive load and low power factor can cause part of the voltage to exceed the lower limit. If there is a distributed power source in this area, the voltage level can be adjusted by reactive power generated by the distributed power source, so that the phenomenon of low voltage can be eliminated. However, in some cases, in the low period of load, the voltage of the whole network is higher than the upper limit, at this time, if there is a distributed power access, the access point will produce a lot of reactive power, and finally the voltage will exceed the upper limit. Therefore, in the distribution system with distributed power supply, the voltage and power are optimized by using distributed power generation and other regulation methods. In view of the complex changes in power flow caused by the access of distributed generation to the distribution network, this paper analyzes the node type, capacity level and spatial distribution of distributed power supply. The influence of distributed power generation on system network loss and voltage is analyzed in four aspects: permeability. The node type is reflected by dividing the distributed power supply into Pi node / PQV) node / PV node and PQ node, and the capacity level is reflected by the same type of distributed power source with different capacity connected to the same node. Spatial distribution is reflected by access to the same type of distributed power source of the same capacity at different node locations, and permeability is reflected by multiple nodes accessing the same type of distributed power source. In this part, the power flow analysis of distributed power generation is tested by Newton-Raphson method on the radiation IEEE33 bus system, and the genetic algorithm is used to optimize the method. Finally, one of the research purposes of this paper is the influence of distributed power generation on the network loss and voltage of the system. The optimization type in this paper is the multi-objective optimization problem of system network loss and voltage deviation. Through the tap of distributed power source and on-load voltage regulating transformer, reactive power compensation device is used to optimize the system network loss and deal with voltage deviation. The methods used are scalarization optimization algorithm and vectorization optimization algorithm. The genetic algorithm of artificial intelligence is used to optimize the multi-objective problem, and the optimization results, convergence, calculation time and other parameters are obtained. For the vectorization algorithm, the multi-objective problem is first transformed into a single-objective problem by using the normalized plane method, and then the Pareto optimization solution set is obtained by using the tracking trajectory interior point method. Finally, the compromise solution of the multi-objective optimization is obtained. The two methods are simulated on the IEEE 33-bus / 21-bus system, and the solution time, convergence and optimization results of the two algorithms are compared.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM714.2

【参考文献】