LAUSR

This paper presents an improved Discrete Salp Swarm Algorithm based on the Ant Colony System (DSSACS). Firstly, we use the Ant Colony System (ACS) to optimize the initialization of the salp colony and discretize the algorithm, then use the crossover operator and mutation operator to simulate the foraging behavior of the followers in the salp colony. We tested DSSACS with several algorithms on the TSP dataset. For TSP files of different sizes, the error of DSSACS is generally between 0.78% and 2.95%, while other algorithms are generally higher than 2.03%, or even 6.43%. The experiments show that our algorithm has a faster convergence speed, better positive feedback mechanism, and higher accuracy. We also apply the new algorithm for the Wireless rechargeable sensor network (WRSN) problem. For the selection of the optimal path, the path selected by DSSACS is always about 20% shorter than the path selected by ACS. Results show that DSSACS has obvious advantages over other algorithms in MCV’s multi-path planning and saves more time and economic cost than other swarm intelligence algorithms in the wireless rechargeable sensor network.