Distribution Network Reconfiguration With Distributed Generation

Nowadays, the world is facing a serious problem of energy shortage and environmental pollution. Distributed generation (DG) has many advantages, so it has been developed rapidly and used widely. As a beneficial supplement to traditional centralized power supply, combing DG with the centralized power supply is an important research direction. However, it will have a significant impact on the power system, one of which is the distribution network reconfiguration.Distribution network reconfiguration is a multi objective nonlinear mixed optimization. The integration of DG will change the structure of the network, leading to the changes of the power flow, which makes the reconfiguration of distribution network more uncontrollable. Therefore, the research on network reconfiguration with DG has a great significance.Firstly, the current situation at home and abroad is introduced in the thesis, including mathematical models of distribution network reconfiguration and various algorithms. The various types of the most widely used DG are summarized. By studying the working principle of wind turbines and exploring equivalent circuit model of double-fed wind generator, stochastic model of wind turbines output is set up, using scenario analysis methods. The characteristics of different types of DG are analyzed to deal with the integrated nodes of DG when calculating power flow through back/forward sweep method.Secondly, the traditional particle swarm algorithm is easy to fall into local optimum and premature convergence.In order to overcome those shortcomings, this paper proposes a chaotic simulated annealing binary particle swarm optimization algorithm (CSA-BPSO). Based on the BPSO, chaos algorithm optimization principle is employed. The speed of particle is initiated through combing Logistic mapping with Chebyshev. mapping, which enhances the uniformity of the initial variables. Using the working mechanism of simulated annealing algorithm, this method improves the inertia weight of BPSO, so as to raise the global search ability. Three test functions are used to simulate and the performance of the proposed algorithm is analyzed.Thirdly, a multi-objective optimal model of network reconfiguration with DG is established in this paper. On the one hand, there are two objectives function in the model, including system loss minimization and minimization of maximum voltage deviation. Then, mathematical methods are used to convert it into a single objective function. On the other hand, some constraints are also considered, including flow constraints, DG capacity constraints, node voltage and branch current constraints, distribution network radiating structure constraints.Finally, the distribution network topology is analyzed so as to establish the radial network criterion. Using binary encoding rules based on loop to analyze the initialization method of particle swarm. CSA-BPSO and BPSO algorithm are used to optimize network reconfiguration with constant power type of double-fed wind farm in the occurrence probalities of different scenarios by IEEE33 system, then comparing the results of the two algorithms. The test results show that CSA-BPSO algorithm is better than BPSO algorithm in all optimization goals.