为了保证配电网优化运行,本文以系统网损最小、节点最低电压幅值最大、开关操作次数最少为目标构造配电网多目标优化重构模型,采用量子人工蜂群算法和帕累托(Pareto)支配关系求解.针对易于陷入局部最优解问题的人工蜂群算法,引入量子理论来进行蜂群的搜索,扩大了全局搜索范围,使其收敛于全局最优解,对于多目标问题,采用拥挤度来控制Pareto最优解的个数.将算法应用于标准IEEE 33节点配电网和加入分布式电源的配电网进行重构测试,实验表明该方法是有效和可行的.
配电网是连接电源系统或输电系统与用户的关键环节,承担着给用户分配和提供电能的作用,而配电网络优化重构作为配电网络优化的重要内容越来越受到电力工作人员的重视.针对广东省博罗县配电网络优化重构项目,提出一种基于粒子群的优化算法,并通过实例分析验证该算法是有效可行的.
基于中长期电力负荷预测受诸多不确定因素的影响,结合广东省博罗县电力负荷的发展情况,建立了一种基于灰色系统理论的中长期负荷组合预测模型,旨在提高博罗电力负荷预测的准确度.对传统灰色预测模型进行了改进,引入二次指数平滑法对原始数据进行修正,构造一个新的样本序列,并构建灰色等维新息预测模型.然后将改进后的GM(1,1)模型与一元二次回归模型进行线性组合,该预测模型精度高,其结果对博罗电力生产实际具有指导意义.
Considering the characteristics of large scale nonlinear mixed integer programming for distribution network reconfiguration,an improved harmony search algorithm is proposed,which controls the parameter evolution of by particle swarm algorithm.First,particle swarm algorithm is introduced to harmony search algorithm to smartly guide the parameter evolution.Then,new disconnecting rules of branch groups and their related concepts are also proposed.By overcoming the constraints on the search capabilities due to fixed parameters,global optimization performance is improved and the generation of infeasible solutions is reduced.Finally,by combining the above method and principles,the accuracy and effectiveness of the proposed algorithm are verified by simulation on IEEE 33-node system and PG&E 69-node system.
In order to solve the problem of distribution network reconfiguration more effectively,an improved bacterial foraging algorithm(BFA)is proposed by combining the characteristics of particle swarm optimization(PSO)algorithm. Firstly,group optimization mechanism of PSO is introduced to BFA. Secondly,considering the defects of BFA in the later stage such as single particle swam and the vulnerability of being easily trapped in a local optimum ,the average variation of reproduction and new dispersal rules are proposed through mechanism analysis,which not only changes the random search of BFA into a guided search but also diversifies the swarm after reproduction,thus the searching speed and global searching ability of the proposed algorithm are effectively improved. Finally,the improved algorithm is ap?plied to distribution network reconfiguration. The simulation result of an example shows that the proposed algorithm can fast and effectively find the optimum solution,thus it is feasible.
A new algorithm for short-term load forecasting based on Kohonen clustering, wavelet packet analy-sis and extreme learning machine( ELM) is proposed. Considering the characteristics of cyclical similarity in power load series, the method choose the days similar with the days to be predicted by Kohonen clustering. The bi-ortho-gonal wavelet is applied to decompose the power load series into sub-sequences at different frequencies. Then, the sub-sequences are sent to ELM network to do the prediction. Finally, the forecasted load is the superposition of all the forecasting results of each sub-sequence. The practical example shows that, compared with traditional single neural network, this combinatorial method has a large improvement in prediction accuracy and operation time and has a certain practicality.
为了更好地解决配电网重构问题,采用改进人工蜂群算法,建立了以系统有功损耗为目标的重构模型.在该改进算法中,利用差分进化操作的思想对人工蜂群算法中的邻域搜索方式进行了改进,直接在离散域进行多维搜索代替原来的单维更新策略,提高了算法的探索能力并且加快了收敛速度.此外,为了克服基于比例适应度选择的不足,用基于适应度排序的选择概率替代基于比例适应度的选择概率,使得种群的多样性得到了保护,从而避免了陷入局部最优.该算法应用于配电网重构能够以较快的速度收敛到最优解,文中的算例结果验证了该算法用于网络重构是可行而且有效的.
In order to further optimize the sitting and sizing of distributed generation, a comprehensive optimi-zation goal was proposed, which aims to minimize the transmission loss and maintain better voltage level, and deter-mine the weight coefficient of different objectives using judgment matrix method. In order to improve the rate of global optimization, the uniform design and opposition-based optimizing were used. Meanwhile, in the improved al-gorithm, to overcome the shortage of selection based on the proportion of fitness, the selection based on fitness ranking probability is adopted instead of depending on the proportion of fitness, which not only protects the diversity of the population but also expands the global search range. Calculation results of standards IEEE 33-bus radial dis-tribution system show that the proposed model is effective and reliable.
针对山西煤矿集团煤矿瓦斯抽放站使用独立避雷针防护直击雷这一现状,分析避雷针直击雷防护上所存在的问题及隐患,提出了一种适合于山西煤矿瓦斯抽防站特殊的雷击防护及接地系统方法.该方法采用防雷构架以及上层避雷网取代避雷针以防护直击雷,并采用构架之间搭建避雷线以防护雷电绕击,同时外引环形接地网进行接地网优化以及进行特殊局部冲击优化,采用GPF-94高效膨润土降阻防腐剂,优化冲击电阻,减少冲击残压,能够迅速将雷击产生的雷电流泄入大地,同时采用电磁暂态软件ATP-EMTP对外引环形接地网降低接地网接地电阻效果进行仿真分析,得出整个防雷系统的切实性和可行性.
基于电网运行状态、网络拓扑结构,综合考虑可对线路脆弱性产生影响的继电保护、节点电压偏移、线路地理位置等因素,提出了能更加准确识别电网脆弱线路的综合介数方法。其中,电网运行状态由发电端和负荷端的功率输送、实时潮流及潮流裕度来监测,网络拓扑结构则通过能对线路产生作用的发电机数量及可从线路吸收功率的负荷个数并结合线路介数来分析。以IEEE39节点系统为研究对象,在得出脆弱线路排序之后,使用三种常用的连锁攻击方式进行仿真测试,通过功率传输能力的变化情况来检验分析所提综合介数的有效性。测试结果表明,所提出的综合介数方法能较好地识别脆弱线路。
针对电力负荷序列的非线性、非平稳性等特点,提出了一种基于集总经验模式分解EEMD-样本熵和El-man神经网络的短期负荷预测方法.为了减小电力负荷序列局部分析的计算规模以及提高负荷预测的精度,先利用EEMD-样本熵将原始电力负荷序列分解成一系列复杂度差异明显的子序列;然后在综合考虑温度及日期类型等因素对各子序列影响的基础上,根据各子序列的特点构造不同的Elman神经网络对各子序列分别进行预测;最后将各子序列的预测结果叠加得到最终预测值,并对EUNITE国际电力负荷预测竞赛公布的数据进行仿真实验.仿真结果表明该方法能有效地提高负荷预测的精度.
本文建立了系统有功损耗、节点最低电压幅值及开关操作次数的配电网多目标优化重构模型,并运用多目标粒子群优化算法求解.多目标粒子群算法的关键是如何选取个体的极值和全局极值,本文依据Pareto支配关系对个体极值进行选择,外部存储器就是全局极值的候选解集,计算外部存储器中各粒子与其他粒子的海明距离之和并作为各粒子的适应值,然后采用与适应值呈比例的轮盘赌方式选取粒子的全局最优位置,避免种群多样性的丧失.带时限的粒子全局极值淘汰策略使粒子能跳出局部最优,防止算法早熟收敛,保持了良好的收敛性.通过IEEE 33节点测试系统仿真计算,实验结果表明了该方法的可行性和有效性.
大量的充电站接入电网会对配电网电能质量产生影响,而电动汽车采用不同的充放电模式对电网的影响与谐波污染不同.本文设计了充电机等效充放电模型以及电路中各参数数值,根据V2G原理与充放电机等效模型,研究了充电机充放电过程对电网的谐波影响.通过simulink仿真分析了多辆电动汽车共用一个充电机和多个充电机同时放电时对电网的影响.分析结果表明当采用多台充电机同时充放电,并且每台充电机有多台电动汽车同时放电时,有利于减少充电站对电网的谐波污染.通过对仿真结果的分析提出了几种可行性对策,使谐波影响降到最小.
The effect of reactive power compensation largely depends on the reactive power compensation point selection. From the dynamic analysis methods for non-analytic complex variable power system, the impedance modulus margin index is pointed out to be the most intuitive indicator for voltage stability analysis, and it is used to preliminarily determinate weak line in system. Then this paper analyzes the system instable process under the transient disturbances, and points out that the transient stability margin is the essential indicator describing the stability of the system transient disturbance process, which can be quickly calculated by using extended equal area criterion. On the basis, a kind of dynamic reactive power optimization configuration method based on transient stability margin index is put forward, then the transient stability margin index increase values of various reactive power allocation schemes are compared to get the optimal one. By IEEE 39-bus system and the Guangdong Power Grid 220 kV network system examples, this method is proved to be effective. This work is supported by National Natural Science Foundation of China (No. 61233008) and Guangdong Power Grid Special Research Project (No. 036000QQ00120001).
Bzsed on the pzrtition of power system,the trznsient stzbility mzrgin index which czn zccurztely determine the system stezdy stzte under trznsient disturbznce wzs used to improve the trzjectory sensitivity of rezc-tive power optimizztion configurztion method. And z rezctive power optimizztion method considering trznsient stzbil-ity of the trzjectory sensitivity configurztion wzs put forwzrd. Ensuring zll system nodes’ trznsient stzbility zs z pre-requisite ,the new method could zvoid the limitztion of compensztion effect when the finzl compersztion points come from the szme pzrtion. According to studying the disturbed process of nodes whose trznsient stzbility is wezk,the czlculztion of trzjectory sensitivity index is reduced,zchieving the purpose of determining rezctive power configurz-tion points,which considers both wezk links znd quzntitztive compensztion system. The proposed scheme is veri-fied to be effective by IEEE 39 points system znd 220 kV power system network of Guzngdong province.
Both EPON and GPRS technology are introduced,their technical characteristics and advantages are analyzed.Based on the dominating EPON communication technology and supporting by the GPRS wireless communication,a distribution automation communication system is established.The distribution network automation level of the base is improved,and the base can run in a safety,economic,reliable and effective way.
In recent years, distributed generation has become a hot topic at home and abroad, and researchers have attached great importance to the new methods of reconfiguration in case of failure. This paper proposes to adopt improved Ant Colony System Algorithm for reconfiguration of the distributed power. Power loss and minimum load balance index as objective function, the ACS algorithms capability of avoiding being trapped in local optima can be enhanced. The algorithm can do better in global search and avoid the problem of the premature convergence. When ACS algorithm is applied to the IEEE33 node system, the simulation results show that the method can not only effectively reduce power losses, but also decrease load balancing index , thus improving the quality of power supply.
In this paper, to solve the unit commitment problem economically and quickly, an improved firefly algorithm (FA) and particle swarm optimization (PSO) algorithm including both discrete and continuous parts was proposed. The starting and shutdown state of units were optimized according to using discrete binary real coded firefly algorithm. The inheritance from earlier state and constraints to later period of time for running time and shutdown time were considered in repair strategy. The continuous PSO was used in units' load dispatch after the process of deciding starting-stopping states, where constraints of power balance, unit ramp, spinning reserve and lower and upper limits were considered. Deal with the upper limits using Penalty function. To solve the constraints of power balance, zooming in the particles of each dimension variables same proportionally, and through sharing the penalty terms among particle swarms the constraints of units' ramping rates are settled. Simulation results of 10-unit and 24-hour system show that the proposed method is correct, effective and predominant.
An advanced control strategy is proposed on the basis of the three-phase grid-connected photovoltaic(PV) systems,which is virtual synchronous generator,making grid-connected PV systems possess external characteristic of the synchronous generator based on the idea for operation of mimic synchronous generator.Firstly,a mathematical model of photovoltaic power generation is established,and then a mathematical model of virtual synchronous generator is created,and lastly photovoltaic power generation system structure is built on Matlab simulation software platform,while a simulation control module of virtual synchronous generator is also set up.Finally,feasibility of grid-connected photovoltaic based on virtual synchronous is verified by simulation results.
Due to the fault-caused trip out of single transmission line that connects the small hydropower system with power grid, the small hydropower system becomes an islanded system and a large amount of power surplus in the islanded system leads to low success rate of autoreclosure of the faulted line. For this reason, an approach to carry out the autoreclosure of such a transmission line is proposed: according to equal area criterion, the maximum allowed reclosure frequency and reclosure time can be calculated by which the system stability can be sustained, thus since the trip out of the single tie line, when the relative shifting of EMFs at both sides of the faulted line went through one or several integral periods and the amplitude of both EMFs are equal, the autoreclosure is put into force. Simulation results of actual system show that the theoretical analysis of the proposed approach and the calculation are correct. To maintain transient stability of small hydropower system, by means of adopting optimal reclosing-time capturing autoreclosure combined with the measure of high-frequency generator tripping the maximum permissible reclosure time between 4sec to 5sec can be achieved, thus the success rate of autoreclosure can be obviously improved. The proposed principle and approach are available for reference to technical renovation of autoreclosure devices for small hydropower systems.