Abstract: The uniform table is designed by the good lattice point method may not be the best, because it cannot generate table of large factors, when n is even; meanwhile it generates only in the subspace of solution space. Based on the above issues, this paper introduces intelligent algorithm, including the Particle Swarm Optimization(PSO) algorithm and the Improved Simulated Annealing(ISA) algorithm. Comparison test results show that intelligent algorithm can generate uniform table of large factors according to the situation. Compared with the Good Lattice Point(GLP) method, both of the intelligent algorithms can achieve better uniformity table, and the deviation of uniform design table can be smaller constructed with ISA algorithm than PSO algorithm.

Key words: uniform design table, Good Lattice Point(GLP) method, Improved Simulated Annealing(ISA) algorithm, Particle Swarm Optimization(PSO) algorithm, uniformity

摘要： 当 为偶数时，使用好格子点法不能产生因素数较大的设计表，且只能在解空间的子空间内产生设计表，所产生的设计表无法保证最均匀。针对上述问题，引入智能计算方法，包括粒子群优化算法和改进的模拟退火算法。对3种算法的优化性能进行比较，结果表明智能算法可以对较大范围的因素数产生设计表，与好格子点法相比，智能算法能够得到均匀性更好的设计表，并且相对粒子群优化算法，改进模拟退火算法构建的均匀设计表的偏差更小。

关键词: 均匀设计表, 好格子点法, 改进模拟退火算法, 粒子群优化算法, 均匀性

CLC Number: 

• TP18