Improved Pareto Optimization Method For Multi-target Power Generation Scheduling Of Wind Power Grid-connected Systems

With the continuous development of intelligent power network,wind energy plays an important role in the power system.Under the premise of the safety and stable operation of power grid,the dispatch of multi-objective wind power grid seems important with the consideration of wind power grid dispatch cost,emission and network loss.So it is of great practical significance to study the optimization method for multi-objective generation dispatch of wind energy integrated electric power system.This paper establishes a multi-objective model for wind power grid dispatch.There are several constraints to meet: the amplitude and phase of nodal injection power constraints;the power balance constraints;the node voltage constraints;the power angle constraints and the spinning reserve constraints.After these constraints being met,the introduction of adaptive penalty coefficient can help better to describe the relationship between the cost of wind power output and deviation.The mathematical model takes into account the penalty cost and transmission line loss,which can reflect the actual operation of the power system more comprehensively and truly.In my paper,I put forward an improved differential evolution algorithm.This algorithm sets up three sub-populations to carry out parallel search.In order to promote the restructuring and exchange of information among populations,they can learn from each other and evolve together in their own population.In the process of mutation operation,chaotic logic is used to generate random sequence that guides the feasible solution to approach towards the global optimal solution.At the same time,use the Boltzman probability distribution to guide new individuals to move towards disposable solution,shortening the crowding distance between the sub-individual and the optimal individual.The zoom factor of non-uniform mutation can control iterative step size,which can ensure the convergence speed of the algorithm.A mid-piecewise normalized normal constraint method was put out in this paper.First,connect two extreme solutions of target function as Utopia line,and there is a point intersection between the Pareto front and pedal line which walks through the midpoint of Utopia line(the midpoint of the Pareto solution set).Then,connect intersection point and the two extreme solutions to obtain the two sub-Utopia line,and find the midpoint of the Pareto line in the subUtopia line by using midpoint segmentation strategy.A series of discrete Pareto optimal solutions are obtained to a suitable degree,which can effectively overcome the uneven distribution of Pareto solution caused by slope change.A multi-objective optimization method is proposed for the improve differential evolution algorithm and the mid-piecewise normalized normal constraint.By way of normal constraint multi-objective optimization problem is transformed into a single objective optimization problem,and then use the improved differential evolution algorithm to obtain the corresponding solution under normal constraint,which can also bring out the advantages of both methods.Use the Matlab2014 software platform to write the algorithm simulation program and the wind power grid connected IEEE-30 node system is simulated,and the optimal compromise solution is obtained by using fuzzy set theory.Simulation results show that the algorithm can obtain a uniform Pareto solution set,and has better global searching ability.The feasibility of the algorithm is thus verified.