Research On The Distributed Generation-Based Black-Start Scheme For Distribution Network

In recent years, the worldwide has witnessed a number of power system blackouts. The blackouts not only cause huge loss in economy, but also further affect people’s life. To build the feasible black-start scheme under blackouts, so as to restore the power supply quickly and stably has important practical significance.In this paper, we realize the black-start of distribution network based on DG and achieve the self-recovery of distribution network under blackouts. We adopt the parallel recovery strategy and divide the distribution network into several subnets. Then, each subnet is restored by the respective DG. And the optimal recovery path is determined. Finally, each restored subnet is connected at synchronization points to form the main distribution network. The main work of this paper includes three aspects:1) The definition and classification of distributed generation are introduced. The paper discusses the working principle of solar photovoltaic power generation, wind power generation, micro gas turbine, fuel cell. And their corresponding mathematical models are established. The distributed generations are compared with the traditional large units. The advantages of DG as black-start power and the corresponding evaluation index are summed up. From the eight aspects of voltage, power loss, power quality, reliability, relay protection, operation planning, fault location and operation analysis, the influence of distributed generation access to the distribution network is discussed.2) The equivalent load model is applied to simplify the complex distribution network. The top-bottom hierarchical model for DG-based distribution network is built to simulate the restoration sequence of radial distribution network in the event of failure. Considering the limited capacity of DG, we propose a partitioning method based on breadth-first search algorithm to determine the maximum power range of DG and synchronization points between subnets. A synchronization method between subnets based on power transmission is adopted to solve the synchronization problem brought by parallel recovery strategy.3) The path optimization of the partitioned distribution network is discussed. According to the partition recovery target, the number of path between the power point and the load node is to measure the recovery speed, the product of node weight and load capacity is to measure the load recovery. And the optimization model is established. The optimization problem is solved by the binary particle swarm optimization algorithm (BPSO). What’s more, the influence of inertia weight and learning factor on the convergence of algorithm is discussed, which improves the existing binary particle swarm optimization algorithm. The proposed approach is evaluated through simulation experiments by using the IEEE 69-bus network as an example and the numerical result verifies the effectiveness of the solution.