本文应用遗传算法，采用基于“链码”的编码方式和数学形态学的变异算子，设计了一种连续体结构拓扑优化的新方法。
A novel algorithm for topology optimization of continuum structures has been designed on the basis of "chain-code" encoding and mathematical morphologic operators.
利用用户认知的不确定性设计定向变异算子。首先，采用主成分分析法辨识用户认知的不确定性；
A directional mutation operator is designed by using the user's uncertain cognitive. The main ingredient analysis is used to identify the user's uncertain cognitive.
通过设计粗粒度并行遗传算法和交叉、变异等算子，提高了算法的计算效率和性能。
The computing efficiency and the performance of the algorithm are improved by a designed parallel algorithm, the crossover and the mutation operators.
并且为上面的改进设计了相应的选择算法、交叉算子、变异算子和免疫算子。
Besides, the thesis also designs the corresponding selection algorithm, crossover operator, mutation operator, and the immune operator for the improvement of the above problems.
针对作业车间调度问题，提出了最小化空闲时间的处理过程及其变异算子，设计了一种自适应遗传算法。
Classic scheduling benchmark problem test shows:the self-adaptive measure can efficiently keep current populations diversity, can use very small population size; shortest idle time .
针对作业车间调度问题，提出了最小化空闲时间的处理过程及其变异算子，设计了一种自适应遗传算法。
Classic scheduling benchmark problem test shows:the self-adaptive measure can efficiently keep current populations diversity, can use very small population size; shortest idle time .
应用推荐