| Optimization is an important activity in the development of human history and civilization,which involves problems in many fields.With the development of science and technology,the complexity of optimization problems is increasing,and the traditional optimization methods are faced with great challenges when solving complex problems.Intelligent optimization algorithm is a kind of search algorithm which simulates nature evolution and is used to solve complex optimization problems.It uses artificial intelligence technology to find the best solution for complex problems.Different from traditional optimization methods,intelligent optimization algorithm can adapt to optimize the objective function,which has higher robustness and reliability,and has better optimization effect.The development of intelligent optimization algorithm provides new ideas and methods for solving complex optimization problems.Differential Evolution(DE)is a kind of population-based intelligent optimization algorithm,which is regarded as one of the most popular optimization algorithms to solve complex optimization problems.It has the advantages of fast convergence speed,fewer control parameters,and robust optimization results.However,the traditional DE algorithm suffer from search stagnation and premature convergence.If the algorithm falls into a stagnant state and cannot continue to search for the optimal solution or falls into a local optimal solution due to the lack of population diversity,which will affect the performance of DE algorithm.Due to the significant impact of population structure on the search behavior of DE algorithm,constructing a specific population structure can greatly improve the performance of DE algorithm.Therefore,this paper combines multiple neighborhood topologies with DE algorithm for different types of optimization problems,and the effectiveness of the proposed algorithm is verified by experiments.The main work of this paper is as follows:(1)A differential evolution algorithm with ternary search tree is proposed in this paper.Based on the three-dimensional container loading problem,this paper divides the remaining space of loaded cargo into three dimensions: upper,left and right,and designs a ternary search tree model to generate the initial population for the algorithm.The three child nodes of the model correspond to the three dimensional spaces,and carry out boundary constraints on the individual population to ensure that the generated solutions can form a packing sequence and ensure the effectiveness of the packing scheme.The proposed algorithm was tested on the Bischoff and Ratcliff instances of threedimensional container loading problems.The experimental results show that the differential evolution algorithm with ternary search tree has certain advantages over other algorithms in solving three-dimensional container loading problems.(2)A differential evolution algorithm with adaptive neighborhood topology is proposed in this paper.For solving the problem of balancing global search ability and local search ability of differential evolution algorithm,this paper constructs a small-world neighborhood topology to avoid large-scale global search,so as to find the convergence direction quickly.Novelty search strategy is introduced to enhance population diversity and jump out of local optimal.In addition,two adaptive scaling factors are designed to balance the exploration and exploitation capabilities of the algorithm.The proposed algorithm is compared with other advanced differential evolution algorithms on the benchmark test function.The experimental results show that the differential evolution algorithm with adaptive neighborhood topology has better performance in convergence accuracy and convergence speed.(3)Solving the transportation optimization problem of open-pit mine development.In this paper,a constraint optimization model of open-pit mining transportation is constructed around the open-pit mining transportation process.The mining capacity,receiving capacity and transportation behavior of open-pit ore are taken as constraint conditions,and the minimum transportation cost is taken as the optimization objective.The differential evolution algorithm with adaptive neighborhood topology proposed is used to solve the constraint optimization model of open-pit mining transportation.In the selection operation of the algorithm,the degree of "feasibility" is introduced to select the individuals entering the next generation of evolution,making full use of the favorable information of excellent infeasible individuals to guide the whole evolution process.The simulation results show that the proposed adaptive neighborhood topology differential evolution algorithm can provide theoretical methods and technical support for the optimization of open-pit mine transportation process. |
