Font Size: a A A

Distribution Network Reconfiguration And Service Restoration Research

Posted on:2013-07-24Degree:MasterType:Thesis
Country:ChinaCandidate:T L ZangFull Text:PDF
GTID:2232330371995732Subject:Power system and its automation
Abstract/Summary:PDF Full Text Request
In electrical power distribution system, the most effective way of optimal control is power distribution network reconfiguration, by operating the switches to optimize the structure of network, transferring the load to change the distribution network power flow, optimizing the distribution network operation mode, restoring power supply of non-fault outage. Therefore, distribution network reconfiguration and service restoration are considered to be the important ways to realize distribution network self-healing function, and there is important practical significance to research and apply them.Intelligent optimization algorithm for distribution network reconfiguration is easy to converge to local optimal solution. Therefore, a hybrid intelligent algorithm called bacterial foraging optimization algorithm with particle swarm optimization strategy (BF-PSO) algorithm is applied to solve distribution network reconfiguration in this thesis. Firstly, the back/forward power flow calculation method and the distribution network reconfiguration mathematical model which aims at the minimum network power loss are introduced. Secondly, the coding scheme, fitness function structure, algorithm parameters setting, algorithm steps, algorithm process and coding reduction principle are introduced in detail. To verify the effectiveness of the proposed method, example tests are conducted on the optimization calculations of the IEEE33and the United States PG&E69node system in Matlab7.10. The results indicate that the BF-PSO algorithm combines the advantages of the BFO and the PSO algorithm which has the strong global and local search capabilities, which can guarantees the diversity of populations and converges to the global optimal solution of reconfiguration. The advantages of solution accuracy and convergence performance of BF-PSO algorithm verify the effectiveness of its application in distribution network reconfiguration.Heuristic search method transforms expert knowledge into the corresponding processing rules, which can significantly reduce the search space. The solution provides better performance of real-time on the distribution network service restoration. In view of this, to implement service restoration of distribution network quickly and accurately, a two-stage service restoration method for distribution network based on the combination of heuristic rules and entropy weight is proposed. In the first stage, by use of heuristic rules, a candidate service restoration scheme set is generated. In the second stage, considering the target of service restoration, five evaluation indices such as the quantities of restored load and transferred load, the margin of load capacity, rate of load balancing and the number of switching operations are introduced in the thesis. The evaluation indices of candidate schemes are calculated, and then these schemes are assessed by entropy weight. Meanwhile, with the introducing of subjective and objective weights, the schemes are synthetically assessed to achieve the preferred and alternative service restoration schemes. Calculation results of a distribution network with six feeders show the effective of the two-stage service restoration method.As to the evaluation for distribution network service restoration scheme, multiple attribute decision making methods are used to evaluate service restoration scheme. As for evaluation based on objective data, the C2R model in Data Envelopment Analysis is used. According to the goals of service restoration, choosing the number of switching operations, the quantities of transferred load and the rate of load balancing as the minimum input index, in the meanwhile, choosing the quantities of restored load and the margin of load capacity as the maximum output index, which transferring the scheme evaluation problem into solving linear programming models. And the sorting of service restoration schemes can be obtained on the basis of the efficiency index values Vk. Then, by the example calculation of the six feeders system, it verifies the C2R model can effectively make the relative efficiency assessment of the service restoration scheme basis of the indices of objective information. As for group decision of service restoration, with the consideration of the group decision-making behavior for power distribution system dispatch, the group decision-making eigenvalue method for service restoration scheme is introduced according to the Perron-Frobenius theorem. And the numerical method is used to calculate the ideal service restoration decision-making expert evaluation value vector X*, namely obtained the sorting of service restoration schemes. Example simulated on5expert’s subjective weights, combining with the objective weights by entropy weight method, can form5experts decision vector. The synthesized evaluation value vector of ideal service restoration decision-making expert is obtained through the group eigenvalue method. The test result indicates that the service restoration group eigenvalue method can effectively integrate various experts’restoration decision making, and the group decision result is consistent with most experts’individual recovery decision-making consistent. As for the consideration of the influence on the service restoration for power load changing, this thesis proposed the interval number representation of distribution network service restoration evaluation indices. The indices’ weights are determined by combination of the subjective and objective information, which uses the analytic hierarchy process method to objective weight and the interval number entropy weight method to subjective weight. On this basis, the interval number grey relation decision-making method is applied to make the interval evaluation for service restoration scheme. The example test indicates that the interval number grey relation decision-making method can effectively solve the distribution network service restoration scheme evaluation with the case of power load changing.
Keywords/Search Tags:distribution network, static reconfiguration, service restoration, intelligentoptimization algorithm, multiple attribute decision making
PDF Full Text Request
Related items