论文部分内容阅读
为了提高智能组卷质量,提出一种基于小生境自适应遗传模拟退火算法的智能组卷策略.该算法动态调整交叉和变异概率进行遗传操作,对中间种群进行小生境选择和模拟退火操作,从而增强了种群多样性,有效克服了遗传算法局部收敛和“早熟”的缺点.文章针对各约束条件建立了组卷数学模型,给出了基于期望平均分的难度分布函数和小生境自适应遗传模拟退火组卷模型.大量测试数据表明,该方法是一种有效可行且实用的组卷方法.
In order to improve the quality of intelligent test paper, this paper proposes an intelligent test paper strategy based on niche adaptive genetic simulated annealing algorithm, which dynamically adjusts the crossover and mutation probabilities to carry out genetic operations and niche selection and simulated annealing of intermediate populations. Enhances the diversity of population and effectively overcomes the shortcomings of local convergence and “precocious ” of GA.In this paper, a paper model mathematical model is established for each constraint, and the difficulty distribution function and niche adaptation based on expected average score Genetic Simulated Annealing Coil Model. A large number of test data show that this method is an effective and practical method of coiling.