Research On Dynamic Decomposition Optimal Dispatch Considering The Confidence Risk Of Wind Power-photovoltaic-hydropower

As the proportion of wind power,photovoltaics and hydropower in the modern smart grid has increased significantly,the multiple uncertainties presented by the integrated grid-connected power generation of wind-solar-hydropower have brought more and more significant risks to the decision-making of power system optimization and dispatch.However,the strong random fluctuations of wind power,photovoltaic and hydropower output and the large differences in grid-related characteristics of power generation equipment limit their large-scale grid connection and complete consumption.At the same time,this phenomenon has brought about huge problems of abandoning wind,light,and water,which poses a huge challenge to the optimization of the power-generation side and the safe operation of the distribution network side.In order to more reasonably and flexibly evaluate the risks brought by the multiple uncertainties of wind power-photovoltaic-hydropower to the optimal dispatch process,this paper will carry out research on this.The main work is summarized as follows:(1)Select an appropriate uncertainty distribution function to fit the predicted wind speed,solar irradiance and river flow data,and obtain the relationship between these uncertain variables and the corresponding wind power,photovoltaic,and small hydropower output.The uncertainty confidence interval and the available output histogram are used to calculate the confidence risk index of the estimated power deviation of the wind power-photovoltaic-hydropower output;Based on the classification opportunity constraint method,a specific calculation formula for the confidence risk of the wind power-photovoltaic-hydropower output over/under-estimation power deviation is proposed;Introduce the system transformer ratio and reactive power compensation device capacity into the decision variables,and use this method to control and adjust the output of the power generation equipment to achieve the purpose of making full use of the controllable resources of the smart grid;Finally,an economic/risk multi-objective source-network coordinated optimal dispatch model is constructed on the basis of taking into account the multiple uncertainties and confidence risks of wind power-photovoltaic-hydropower.(2)In order to solve the complex model established in this paper efficiently and reliably,an innovative multi-objective dynamical decomposition evolutionary(MODDE)algorithm is proposed,which combines the efficient dominant feasible solution constraint processing method with multi-objective evolutionary algorithm based on decomposition using allocation strategies in dynamic resources,which makes the algorithm solution process more efficient,maximizes the use of the population information of infeasible domain solutions,and solves the problem that the calculation amount of each sub-problem in conventional MOEA/D is the same,which leads to the inefficiency of the optimization process.In terms of selecting the pareto optimal compromise solution,an improved technique for order preference by similarity to an ideal solution(TOPSIS)is proposed.Compared with the traditional TOPSIS method,the relative proximity calculation formula of each scheme and the positive and negative ideal solutions is improved,which can be more reasonable.And objectively evaluate the pros and cons of each program.(3)The effects of burr distribution,log-normal distribution and gumbel distribution in fitting wind power-photovoltaic-hydropower uncertainties are compared,and the simulation calculation is carried out with the improved IEEE30 node system connected to the wind power-photovoltaic-hydropower unit.The case analysis is carried out from the three aspects of different confidence levels,different dispatch methods and different optimization algorithms,and the following conclusions are drawn: the dispatch model in this paper still can take into account risks and costs in the face of uncertain variables of different proportions.This verifies the accuracy and feasibility of the model;When the source-network is coordinated and optimally dispatched,the voltage distribution of each node in the network is more balanced,which can well improve the problem of large overall voltage deviation in conventional dispatch,and reduce the harm of conventional dispatch to many electrical equipment and the safety,stability and economic operation of the power grids;the superiority of the MODDE algorithm is higher than other optimization algorithms,and its performance and computational efficiency are better.The models and algorithms proposed in this paper provide new ideas for electric vehicle energy storage,load demands response,and multi-objective intelligent optimization algorithms,which can be further integrated into source-grid-load-storage coordination optimization calculations.