CAJ | 학술논문

　　智能组卷是一个包含多重约束条件的目标优化问题，遗传算法的群体搜索策略可以为多目标优化提供较好的解决方案。但传统的遗传算法在组卷过程中存在收敛速度慢、收敛性较差等缺点，组出的试卷质量不高。提出一种新的元胞遗传组卷算法，将群体中的所有元胞按照一定的演化规则演化之后，再进行遗传操作，并把该算法应用到智能组卷中。实验结果表明，新的元胞遗传组卷算法与传统的遗传组卷算法相比，可以有效地提高收敛速度，并能进一步改善收敛性，组出的试卷更加符合人们的要求。
　　지능조권시일개포함다중약속조건적목표우화문제，유전산법적군체수색책략가이위다목표우화제공교호적해결방안。단전통적유전산법재조권과정중존재수렴속도만、수렴성교차등결점，조출적시권질량불고。제출일충신적원포유전조권산법，장군체중적소유원포안조일정적연화규칙연화지후，재진행유전조작，병파해산법응용도지능조권중。실험결과표명，신적원포유전조권산법여전통적유전조권산법상비，가이유효지제고수렴속도，병능진일보개선수렴성，조출적시권경가부합인문적요구。
The intelligent test paper construction is a multiple constraints objective optimization problem. The groups search strategy of the genetic algorithm can provide a better solution for multi-objective optimization. Traditional genetic algorithm has shortcomings such as slow convergence and poor convergence in test paper process. The error of the test paper is relatively large. This paper proposes a new intelligent test paper construction method based on cellular genetic algorithm. Compared to the traditional test paper construction method, the novel approach can effectively improve the convergence rate, and further improve the convergence in intelligent test paper. The result is more in line the requirements of users.