Research On Array Layout Of Wind Farm Based On Kinetic-molecular Theory Optimization Algorithm

Kinetic molecular theory optimization algorithm is an artificial physical law optimization algorithm based on the theory of molecular thermal motion in physic,and it is also a computation intelligence method by simulating the attraction and repulsion between molecules.In view of the advantages and disadvantages of the KMTOA,several strategies have been designed to improve the performance of the KMTOA,such as M-elite coevolutionary strategy ? memory mechanism and crystallization principle.Profit from the enrichment and perfection of the theoretical basis of kinetic molecular theory optimization algorithm,this paper constructs a single and multi-objective optimization algorithm to practical application to solve the layout scheme of the wind farm array efficiently.This related research work are as follows:(1)The relationship and difference between the KMTOA and other algorithms are analyzed in detail,the characteristics and potential of the KMTOA are also explained.This paper explores the topological structure and clustering phenomenon of the KMTOA.In view of the advantages and disadvantages of the KMTOA,the weak connection theory in sociology is introduced to improve the topological structure of the KMTOA.Then,the ability of the KMTOA to escape from the local optimum is improved by the multi-subgroup co-evolution strategy and chaotic perturbation method.The improved KMTOA has better performance in single?multi-modal and migration functions.The test results on the migration function show that the improved KMTOA can overcome the inherent defects of the original algorithm's topological structure.(2)In order to solve the multi-objective optimization problem,a multi-objective kinetic molecular theory optimization algorithm based on decomposition and dominance criteria(BCE-KMTOA)is proposed.By improving the topology of the KMTOA,the problem that the optimal individual cannot adapt to multi-objective optimization in the algorithm updating formula is solved skillfully,and the optimal individual selection strategy based on decomposition is constructed.This algorithm preserves the evolutionary mechanism of the KMTOA,and the test results on the DTLZ and ZDT function test set show that the BCE-KMTOA can well solve the multi-objective optimization problem.(3)The mathematical model of wind farm array layout problem is established,and the single objective function of wind farm power output based on wake effect is constructed.By discretizing the continuous single objective function,the difficulty of solving the optimization problem of wind farm array layout is reduced.Then the improved single objective KMTOA is used to solve the problem.In order to obtain more optimal schemes for wind farm array layout,and to enable decision makers to take into account the power output and efficiency of wind farms,a two-objective wind farm array layout optimization model was designed.This model abandons the coordinate-based wind farm model,and then applies the proposed bi-criteria multi-objective KMTOA to solve the multi-objective wind farm array layout optimization problem.