Multi-objective Reactive Power Optimization Of Distribution Network With Distributed Power

Reactive power optimization is an effective way to improve the power quality and stability of the power system. In recent years, more and more Distributed Generations (DGs) are introduced into the grid, which has a certain impact on the power flow and voltage of distribution network. In the meantime, it also affects reactive power optimization of the distribution network. Therefore, the research of reactive power optimization of the distribution network containing DGs has great significance.On the premise of studying basic theory of Differential Evolution algorithm (DE), this paper discusses how to improve it with quantum encoding and Artificial Bee Colony algorithm (ABC). Then reactive power optimization of the distribution network containing DGs is studied. The work includes:1. This paper analyzes the difference between the distribution network containing DGs and the traditional distribution network. Some common DG technologies are introduced. Then we study the impact of DGs on the distribution network from a theoretical perspective.2. DE algorithm is widely used in the field of optimization in recent years. It has many advantages in that its parameters are simple to set and its calculation is very easy. However, the traditional DE algorithm has a premature convergence and poor local search ability. We use the idea of quantum encoding and ABC algorithm to improve DE algorithm and an improved quantum differential evolution algorithm (IQDE) is proposed. In the improved algorithm, the idea of quantum encoding increases the individual diversity while the accelerating evolutionary operation of the follower and the random search operation of the scouter respectively improves local search ability and global search ability of DE algorithm.3. In order to verify the performance of the proposed algorithm in reactive power optimization, with the objective of minimum active power loss, the improved algorithm and DE algorithm are applied to 14 bus system for simulation. The results show that with less convergence time and smaller population size, IQDE algorithm can obtain an even or better optimization effect compared with DE algorithm.4. This paper then studies reactive power optimization of the distribution network containing DGs. A simulation is done on 30 bus system with DGs. The results prove that IQDE algorithm has better performance than DE algorithm. By the statistics of voltage changes of network system nodes before and after DG mergence, we notice that the introduction of DG is likely to cause a voltage close to the upper limit. So it is necessary to consider both minimum voltage deviation and minimum active power loss as the objective functions.5. In order to achieve multi-objective reactive power optimization, a multi-objective IQDE algorithm based on Pareto optimal solutions is proposed. We study the construction method of non-dominated solution set and the preservation method of non-dominated solutions, as well as the crowding distance which controls the size of elite set. Then through 30 bus system simulation, we can reach a conclusion that the algorithm can obtain a uniformly distributed Pareto front. It simultaneously optimizes two objective functions while meeting both economic and security requirements of the power grid.