Abstract: Backtracking Search optimization Algorithm（BSA） is a new optimization algorithm which is proposed in recent years. For the convergence speed of original BSA is slow and it is easily trapped in local optima, an improved BSA based on the combination of guidance with the best individual and niche technique is proposed to improve the global performance of it. In the method, the strategy with learning from the best individual is introduced in the mutation operator of original BSA to improve the convergence speed of it in the first. In the second, a niche repulsing technology is designed in the paper. The niching radius is determined according to the average minimal distance between every individual and the other individuals, and some parts individuals with high similarity are deleted, some new individuals are generated by a new mutation method which is designed with combining the worst information of current generation, and the new individuals are supplemented in the new population to maintain the constant number of population, the diversity of the population is increased by this operation. The convergence speed and the diversity of population is fully considered in the improved BSA, and the performance of the original BSA is largely improved. 10 typical functions are used in the simulation experiments, and the results are compared with those of other algorithms. The results indicate that the improved algorithm has good performance in terms with the convergence speed and accuracy.

Key words: Backtracking Search optimization Algorithm（BSA）, guidance mechanism, niche technology, mutation strategy

摘要： 回溯搜索优化算法（BSA）是近年提出的一种新型优化算法，针对其收敛速度较慢、易陷于局部最优的缺点，提出了一种基于最优个体引导和小生境技术相结合的改进BSA算法。本方法首先在BSA的变异操作中引入向最优个体学习的策略，以提高算法的收敛速度；其次，设计一种新的小生境排挤技术，根据每个个体到其他个体距离的平均最小值确定小生境半径，排除部分相似性较高的个体；结合群体当前的最差信息，设计一种新的变异方法产生一定数量的新个体补充到新的种群中，维持群体数量的恒定并增强群体多样性。改进的BSA算法充分考虑了算法的收敛速度和群体的多样性，较大地提高了传统BSA算法的性能。对10个典型函数进行仿真测试，并与其他算法结果进行对比，实验结果表明，改进算法在收敛速度与精度方面具有较好的效果。

关键词: 回溯搜索优化算法, 引导机制, 小生境技术, 变异策略