Modeling And Solution Research On Dynamic Economic Dispatch Of Power System With Wind Power

Economic dispatch is a core issue in the optimal operation of the power system.In recent years,with the continuous improvement of "energy saving and emission reduction" requirements,the traditional dispatch mode that takes economic benefits as the single objectivie must be changed.Economic emission dispatch(EED),which can take into account both economic indicator and environmental protection,has gradually become a research hotspot.At the same time,as the proportion of wind power in the power system increases year by year,the uncertainty of wind power will inevitably bring a non-negligible impact on the operation of the power grid.How to analyze and model the uncertainty of wind power,and establish a reasonable optimal dispatch strategy for wind power,and promote the consumption and utilization of wind power,has urgent needs and important practical significance.In view of the above problems,the research of this paper mainly focuses on the modeling of dynamic economic dispatch(DED)and dynamic economic emission dispatch(DEED)considering wind power connection,the corresponding solution algorithm,and the solution to large-scale unit commitment etc.,the main research content and results are as follows:(1)Calculate the Weibull distribution parameters of regional wind speed by time period,thereby obtain the probability density function(PDF)and cumulative distribution function(CDF)of wind power output.Based on this,it is proposed to use the difference between the conditional expectation of wind power output and the scheduled wind power output to determine the positive and negative spinning reserve required for wind power incorporation.Chance constraint is imposed on the output of wind farms and its deterministic form is transformed according to the CDF of wind power,thereby constructing a DED model including wind power connection.In order to solve this non-convex objective function,a joint solution algorithm based on differential evolution(DE)algorithm and sequential quadratic programming(SQP)is designed.An adaptive mutation factor is adopted in the DE algorithm,which can improve the global search ability of DE algorithm,and the SQP method can further optimize the solution obtained by DE algorithm.(2)On the basis of the single-objective DED model in previous charpter,a DEED model including wind power is extended by introducing the pollution emission objective of thermal units.In order to solve the model,a hybrid multi-objective optimization algorithm combining DE algorithm and particle swarm optimization(PSO)algorithm is proposed.The algorithm is implemented based on Pareto dominance theory and dynamic external archive set.The algorithm adopts a dual-population evolution mechanism that can make full use of the advantages of DE and particle swarm algorithm.The calculation method of crowding distance and the clipping rule of Pareto solution set get improved.Two different test cases demonstrate the validity and rationality of the proposed model and algorithm.(3)In order to avoid the problems of poor model adaptability and accuracy in traditional parametric modeling methods,based on non-parametric kernel density estimation(KDE)technology,the probability distribution of wind power are its forecast error are accurately modeled.A segmented statistics method on wind power forecast data is used to estimate the confidence interval of wind power output and the upper and lower bounds of the forecast error.According to the established probability model of wind power output,the prediction confidence interval,and the upper and lower bounds of the prediction error,a DED model containing wind power is constructed.A hybrid algorithm combining the advantages of Bat algorithm(BA)and PSO evolution is designed to solve the built model,and an individual crossover mechanism is introduced in the algorithm's evolution to solve the problem of falling into local optimality easily existed in BA and PSO.Finally,test cases are performed to verify the effectiveness of the model and algorithm.(4)The more comprehensive considered model for DED based on non-parametric KDE technology is extended to the corresponding DEED model.In order to effectively solve this DEED model with multi-objective,high-dimensional,nonlinear and strong constraints,an improved multi-objective brain storm optimization(IMOBSO)algorithm is proposed based on the basic brain storm optimization(BSO)algorithm,where measures such as random clustering centers,differential mutation and crossover operations are introduced to enhance the converging and diverging operation of BSO.Finally,the effectiveness of the algorithm and model is verified by a classic simulation example,and the results are analyzed in detail.