压下的

The same result can also be achieved when optimizing F8 function (Rastrigin's function). Comparing to EGA, KEGA is easy to realize and the increase of computation load can be neglected, but the performance of KEGA can be improved greatly. 2. To enhance the speed of AIA, two new methods of computing the concentration of antibody were developed. Simulation results for F8 and F15 (needle in haystack: type I) functions show that the two methods can improve the running speed of AIA more than 8. 5 times. 3. As to the drawback in defining antibody concentration, a new definition based on Euclidean distance between two antibodies and fitness is proposed. On this basis, a new AIA was constructed. The convergence performance are more better than that of AIA. 4. Combining king crossover with AAIA and DBAIA, two new AIAs, KAAIA and DKBAIA, were proposed.

仿真实验结果表明KEGA实现简单，与EGA相比，其增加的计算开销可以忽略，但性能改善十分明显； 2、针对基于信息熵的免疫算法运行速度慢的缺点，提出了二种能够加快AIA运行速度的新的抗体浓度计算方法，得到两种新的免疫算法，即加速的人工免疫算法(Accelerated Artificial Immune Algorithm，AAIA)和加速的人工免疫算法-2（Accelerated Artificial Immune Algorithm-2，AAIA-2），对测试函数F15(needle in haystack：type I)、F8函数的仿真研究表明，这两种算法的运行速度都是AIA的8.5倍以上； 3、针对AIA的浓度定义存在的缺陷，提出了一种新的基于抗体间欧氏距离和适应度的新的抗体浓度定义，并依据该定义构造了一种基于欧氏距离的人工免疫算法(Euclidean Distance-based Artificial Immune Algorithm，DBAIA)。