Research On Capacity Optimization Of Grid-connected Microgrid Considering Uncertain Factors

Microgrid is an important means for distributed generation to connect to power grid,because it can improve energy utilization and reduce transmission costs.As an important part of the microgrid design,the capacity optimization configuration of microgrid is based on the weather resources and load data in the region,and a complete model is established,and the capacity of each component in the microgrid is determined by intelligent algorithm.It can ensure the reliability of power supply and reduce the economic input of the whole system.First of all,the research methods,model building and uncertainty analysis problems of microgrid are analyzed,then the mathematical power models of micropower and energy storage batteries required for this study are introduced,and the characteristics are analyzed.Then,the traditional genetic algorithm search process is prone to premature convergence,and later falls into the optimal solution,which makes it difficult to effectively solve the problem of microgrid capacity optimization configuration.In this paper,a parallel wolf swarm-peer selection genetic algorithm(WP-CPGA)is proposed.By changing its selection mechanism and cross-mutation strategy,WP-CPGA can avoid the interference of local optimal solution and premature convergence,and ensure the speed and accuracy.Through three kinds of test function tests,the optimization capability of WP-CPGA is compared with other similar algorithms.The results show that WP-CPGA has better convergence accuracy in solving problems with multivariable and multiconstraint conditions.Thirdly,taking the number of fans,photovoltaic and storage batteries as optimization variables,and taking economy,grid-connected possibility and reliability as indicators,multiple objective functions are selected to establish a deterministic framework for the optimal capacity allocation model of microgrid.The multi-objective optimization problem is transformed into a single-objective optimization problem by fuzzy analytic hierarchy process and solved by WP-CPGA.The results are analyzed from the effects of different weights and algorithms.The results show that the equal weight configuration scheme is only suitable for general occasions,and the target-biased configuration scheme can bring more benefits in running scenarios with clear objectives.Finally,aiming at the serious errors caused by uncertainty prediction and the drawbacks of redundant calculation caused by scene analysis,this paper uses Wasserstein distance index to deal with the uncertainty of wind,light and load.On the basis of multistate modeling,a microgrid capacity allocation optimization model is built under the framework of uncertainty,which takes the lowest annual average economic cost combined with uncertainty operation risk as the objective function.The model is solved by WP-CPGA.Finally,the two configuration results are compared and analyzed under different random scenarios.The results show that the final configuration scheme is better when considering uncertainty factors.