Research On The Improvement Of Sparrow Search Algorithm

Sparrow Search Algorithm(SSA)is an efficient intelligent optimization algorithm inspired by the foraging behavior of sparrows.The SSA algorithm is used in various practical fields because of its lightweight structure,easy-to-understand principle,easy implementation of the process,relatively few parameters that need to be defined in advance during the execution,and good exploration and exploitation capabilities.However,SSA algorithm has problems such as highly dependent on partial individual search and each update of the population is too random,easy to fall into local optimum,and unstable operation results on complex problems.In this paper,we address the limitations of the SSA algorithm and propose two improvement algorithms in terms of improving the population quality and balancing the exploration and exploitation capability of the algorithm.The main innovation points and research works in this paper are as follows:(1)A Differential evolution strategy-based Sparrow Search Algorithm(DSSA)is proposed to address the drawbacks of high initial population randomness and susceptibility to local optima in SSA algorithm.The algorithm uses K-means clustering to classify the initial sparrow population,which reduces the influence of some individuals on the overall algorithm to a certain extent and improves the robustness of the algorithm;a dynamic differential evolution guidance strategy is also designed to guide the sparrow search using a differential evolution strategy with strong exploration capability,which better balances the mining and exploration capabilities of the algorithm.The DSSA algorithm is tested by 10 benchmark functions and compared with the optimization results of several advanced intelligent optimization algorithms,and the results show that the overall performance of the DSSA algorithm optimization is better.Finally,the DSSA algorithm is applied to solve the travel quotient problem,and the results show that the shortest path can be generated with better results when the problem is solved using the DSSA algorithm.(2)The Adaptive Spiral Flight Sparrow Search Algorithm(ASFSSA)is proposed to address the drawbacks of slow convergence speed and poor performance in solving high-dimensional problems of the SSA algorithm.The algorithm first uses Tent chaotic mapping to form a relatively uniform initial population within the solution range;in the discovery phase,it cooperates with the exploration actions of adaptive weights and Levy flight strategy to ensure the diversity of the population as much as possible;subsequently,it introduces a variable spiral search strategy in the following phase to give individuals a certain chance to jump out of the local extremes and enhance their ability to explore better solutions.To a certain extent,it improves the defects such as poor population quality and long convergence time.By analyzing the performance with five advanced intelligent algorithms on 18 benchmark functions with different characteristics,the results show that the ASFSSA algorithm has better convergence accuracy and speed,which verifies the effectiveness of the improved algorithm.