Ｒesearch on Underground Pipeline Location Based on Multi-modal Sparrow Search Algorithm#br#
DONG Qianhui 1 ，SUN Qian 2* ，LI Kun 1 ，LI Weiyi 3
1． College of Physics and Optoelectronic Engineering，Harbin Engineering University，Harbin 150001，China;2． College of
Information and Communication Engineering，Harbin Engineering University，Harbin 150001，China;3． College of Power and
Energy Engineering，Harbin Engineering University，Harbin 150001，China
摘要 基于多信号分类算法(Multiple Signal Classification，MUSIC)算法的地下目标定位算法在多维遍历搜索过程中存在计算复杂度高、实时性差、无法应用于车载连续探测的问题。针对此问题，本文提出了一种基于多模态麻雀搜索的 MUSIC 地下目标快速定位算法，该算法推导了离地探测模型下的导向矢量模型，将 MUSIC 算法推广到地下管道定位中，并与寻优能力强、收敛速度快的麻雀搜索算法相结合，提高了算法的实时性。此外针对多目标场景下传统麻雀搜索算法无法同时搜索多个谱峰极值的问
题，本文采用聚类算法对麻雀种群进行划分，从而形成多个子种群对目标峰值并行搜索，并结合粒子群算法实现峰值处局部搜索。仿真结果表明，本文所提方案在避免 2D-MUSIC 算法网格量化误差的同时，用时仅为 2D-MUSIC 算法的 0．979%，且与同类算法相比具有更高的精度与搜索成功率。
Abstract: Aiming at the problems of high computational complexity and poor real-time performance whenMUSIC (Multiple Signal Classification ) algorithm is used in multi-dimensional traversal search for underground target location，a multi-modal Sparrow MUSIC fast underground target location algorithm is proposed．In this algorithm，the guidance vector model under the ground detection model is established．Therefore，the application of the MUSIC algorithm is extended to the underground pipeline location．Sparrow search algorithm powerful searching ability and convergence speed，which can be combined with MUSIC algorithm to achieve fast positioning．In view of the problem that the classical sparrow search algorithm cannot search multiple spectral peaks，the cluster algorithm is used to divide the sparrow population，so that the
sparrow population can form multiple subpopulations to search the peaks in parallel．Moreover，the combination
with the particle swarm optimization algorithm ensures the local search at the peaks．The results of simulation and experiment show that the proposed algorithm has high localization accuracy，and compared with the 2D-MUSIC algorithm the average time for target localization is only 0．979%．In addition，compared with the same type of algorithms，the proposed algorithm has a higher search success rate．
