Multi-objective Power Grid Planning Optimization Considering Power Quality Of Fan Access

Large-scale wind power integration promoted the rapid development of power industry and the level of country economic, also caused a great impact on the normal operation and planning of the grid, especially a series of power quality problems which can not be ignored, where the voltage fluctuation and flicker are the main negative impacts. In order to ensure the power quality after the grid planning in line with national standards, this requires that the power quality problem caused by wind farm connected need to be considered in grid planning.Grid planning has proven to be a complex system optimization problem with multiple objectives, nonlinear and solved with definitely difficult. Aiming at artificial selection target weight of traditional optimization algorithm and local convergence of analog binary cross operator, the normal distribution crossover(NDX) is introduced into NSGA2 algorithm. On the one hand ` NSGA2 algorithm uses the fast non-dominated sorting, the congestion level and elitist screening Pareto optimal solution set so as to achieve each objective function can be relatively large(or small), and it can overcome the shortcomings of traditional multi-objective optimization algorithms, showed a strong advantage in the field of multi-objective optimization; On the other hand, NDX operator uses a normal distribution and discrete restructuring operations to improve the solution space exploration and development capabilities, to further expand the search space. A combination of both can improve the global search performance and convergence, and to obtain high-quality and evenly distributed Pareto optimal solution.Since flicker level at the point of common connection of wind farms is directly related to the grid equivalent impedance of PCC`and thus directly related to grid structure, therefore, on the basis of considering wind farm grid power quality problems, this article constructs the wind farm network planning model with targets of total line length and value of flicker at the PCC. Taking into account the wind and load forecast uncertainty, with DC probabilistic load flow calculation method based on Monte Carlo, and chance constrained programming method to determine whether the plan violates branch power constraints, in order to check reliability of optimization program.In order to verify the effectiveness of the proposed model and methods, with IEEE-6 nodes and modified IEEE-24 bus system, using traditional genetic algorithms and two different NSGA2 algorithms to optimize, comparative results show that NSGA2 algorithm showed great superiority than traditional genetic algorithm to deal with multi-objective optimization problem in planning, And compared to the classical NSGA2 algorithm, Proposed algorithm also has high decision-making efficiency, and the resulting optimization plan which contains wind farm grid, ensure the economy, also make power quality network achieve higher quality.