摘要
建立了一种多目标优化的数学模型,并针对标准遗传算法易早熟收敛和进化缓慢的特点,提出了一种改进的组卷遗传算法对模型进行求解.该算法在编码策略、基因修正和算子概率3个方面对标准遗传算法进行了改进.实例分析及仿真验证表明:提出的建模方法将组卷成功率提高到100%,算法运行时间降低到300ms以内,总体上极大地提高了智能组卷任务的执行效率,并能够定量地评估和控制组卷质量.
An optimi zed multi-objective mathematical model was built based on four different test p aper properties.Then,an improved genetic algorithm was proposed to solve this mathematical problem.Three aspects of the genetic algorithm,including the gene encoding,gene modification,and probability of the genetic parameter were impr oved to avoid the disadvantages of premature convergence and slow-evolution.Th e simulation analysis of the experimental data indicates that the success rate i s improved to 100% and the running time of algorithm is limited within 300 ms. So this mathematical model and the improved genetic algorithm promotes the runni ng efficiency of the execution and evaluates and controls the quality of the gen erating result in the task of intelligent-generating test paper.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2013年第5期82-85,共4页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(11271146)
关键词
遗传算法
多目标优化
基因编码
基因修正
遗传算子
智能组卷
genetic algorithms
multiobjective optimizatio n
gene encoding
gene modification
genetic parameter
intelligent-generating test paper
