对精英加速的改进人工鱼群算法*

人工鱼群算法是一种群智能全局随机优化算法，存在算法收敛精度低和效率差的缺点。为克服这一缺点，利用最速下降法具有运算简单、运算速度较快的特点，提出了对精英加速的改进人工鱼群算法。该算法利用最速下降法对适应度值最好的人工鱼更新，通过人工鱼之间信息交换指导其他人工鱼，提高鱼群整体水平，加快人工鱼群算法收敛速度。数值试验结果表明，所得改进人工鱼群算法不仅运算量减少，而且具有更快的收敛速度和更高的收敛精度。改进算法提高收敛精度和运算效率，相较其他算法具有一定优势。

An improved artificial fish swarm algorithm to accelerate the elite

Artificial fish swarm algorithm is a swarm intelligence global stochastic optimization algorithm, its disadvantages are poor local search capability and low efficiency. The steepest descent method has the advantages of simple operation and fast computation speed, this paper proposes an improved artificial fish swarm algorithm to accelerate the elite. This algorithm uses steepest descent method to update artificial fish with the best fitness value, through exchange of information between fishes guide other artificial fishes, improve the overall level of fish swarm, accelerate the convergence rate of artificial fish swarm algorithm. Numerical test results show that, the improved artificial fish swarm algorithm not only reduces the computational complexity, but also has a faster convergence rate and a higher convergence precision. The improved algorithm improves the convergence precision and computational efficiency, and has some advantages compared with other algorithms.