| Optimization problem is a kind of problem throughout ancient and modern civilizations,which is defined as finding the value of decision variables that optimize a certain indicator under the premise of satisfying certain constraints.With the rapid development of human society,the constraints of optimization problems become more and more complex,and the classical optimization methods perform poorly in the face of multi-objective complex optimization problems.At this time,people need a new optimization algorithm with high efficiency,high adaptability and high scalability to solve complex,nonlinear and large scale optimization problems,and the swarm intelligence optimization algorithm is applied.Since it was developed,it has received a lot of attention from scholars and has been successfully applied in many fields of people’s life and production.So far,a large number of swarm intelligence optimization algorithms have been proposed by researchers,such as particle swarm algorithm(PSO),whale optimization algorithm(WOA),sparrow search algorithm(SSA),etc.Although they have obvious advantages compared with classical optimization algorithms,they still have the disadvantages of being premature and falling into local optimal solutions when dealing with practical problems.In this paper,we propose corresponding improvement schemes for the problems in WOA and SSA,and the improved algorithms are successfully applied in practical engineering.The main research works are as follows.(1)In-depth analysis of the algorithm models and the optimization search process of WOA and SSA,and derive the advantages and disadvantages of WOA and SSA and the areas that can be improved.(2)The basic whale optimization algorithm is improved using a hybrid strategy to address some of its own shortcomings.First,it is initialized by the Zaslavskii chaotic mapping graph algorithm to obtain a population with better ergodicity;second,an elite search library strategy is used to enhance the global exploration capability of the algorithm with the ability to jump out of the local optimum;finally,by introducing an adaptive variable speed adjustment factor,the search capability and exploitation capability of the algorithm are effectively coordinated while retaining the advantages of the whale optimization algorithm.The improved whale optimization algorithm is simulated and experimented on 10 benchmark test functions,and then applied to wireless sensor network node localization,and the results show that the improved algorithm outperforms other algorithms,and also verifies its effectiveness in practical engineering.(3)A sparrow search algorithm incorporating forbidden search is proposed and used to solve the robot path planning problem.To address the problems that the sparrow search algorithm is prone to local optimal solutions and insufficient convergence accuracy,Latin hypercube sampling is introduced to initialize the population,which can obtain a uniformly distributed primitive population in the feasible domain interval and make the probability of finding a solution with good diversity and convergence higher;the time-varying Corsi variational operator is used to update the position of the discoverer,which not only maintains the diversity of the population but also improves the convergence accuracy of the algorithm;the taboo search idea is introduced at the convergence The introduction of the taboo search idea at the later stage of convergence can effectively prevent the algorithm from falling into local optimal solutions,thus achieving further performance improvement of the improved sparrow algorithm.The improved algorithm’s performance is significantly improved and its stability is also enhanced by 10 benchmarking functions.Then it is applied to the path planning of mobile robots,which can find the global optimal path efficiently and accurately by combining the actual situation of more and less obstacles,proving the feasibility and effectiveness of the sparrow algorithm incorporating forbidden search in the field of path planning. |
PDF Full Text Request