CAJ | 학술논문

针对目前智能算法初期收敛速度难以满足坦克分队武器目标分配（Weapon-Target Assignment，WTA）要求的问题，提出了一种改进人工蜂群算法。该算法结合 NEH 启发式算法和随机方法对种群进行初始化，利用变邻域搜索和模拟退火方法改进了采蜜蜂算法，并简化了跟随蜂算法，提出了一种全局最优限制算法。最后，结合不同规模的 WTA 问题，给出了该算法参数的确定方法。仿真结果表明：改进人工蜂群算法相比于其他算法在初始种群质量和算法初期收敛速度方面具有明显优势，特别适合求解坦克分队 WTA 问题。
침대목전지능산법초기수렴속도난이만족탄극분대무기목표분배（Weapon-Target Assignment，WTA）요구적문제，제출료일충개진인공봉군산법。해산법결합 NEH 계발식산법화수궤방법대충군진행초시화，이용변린역수색화모의퇴화방법개진료채밀봉산법，병간화료근수봉산법，제출료일충전국최우한제산법。최후，결합불동규모적 WTA 문제，급출료해산법삼수적학정방법。방진결과표명：개진인공봉군산법상비우기타산법재초시충군질량화산법초기수렴속도방면구유명현우세，특별괄합구해탄극분대 WTA 문제。
Aiming at the problem that it is difficult for the current intelligent algorithms to meet the re-quirement of tank unit Weapon-Target Assignment (WTA)for faster convergence speed on the early stage,an improved Artificial Bee Colony (ABC)algorithm is proposed.This algorithm adopts a combina-tion method of NEH heuristic algorithm and random method for population initialization,uses variable neighborhood search and simulated annealing method for improving employed bees algorithm and simplify-ing unemployed bees algorithm,and introduces a global optimal limit algorithm.At last,combining dif-ferent scale of WTA problem,the method used to determine algorithm parameters are given.The simula-tion results reveal that the improved ABC has a significant advantage on the quality of the initial popula-tion and the convergence speed on the early stage over other algorithms,and it is particularly suitable for tank unit WTA problem.