不可靠测试条件下基于NSGA-Ⅱ的多目标测试优化选择

针对测试不可靠因素严重影响测试优化选择结果以及现有方法不能很好解决多目标测试优化选择等问题，提出基于第二代非支配排序遗传算法(NSGA-Ⅱ)的多目标测试优化选择的方法。首先，描述了测试不可靠条件下多目标优化选择问题的数学模型；其次，在该数学模型下，将系统给出的故障检测率和隔离率作为约束条件，将测试代价、漏检率和虚警率作为优化目标，建立了多目标优化问题；然后，提出带有精英保留策略的NSGA-Ⅱ对多目标问题进行优化选择，利用NSGA-Ⅱ能够得到一组Pareto最优解，可根据实际需求选择最优的测试组合；最后，针对某装备进行实例分析，得到3组最优解，可以满足不同需求下的最优选择，验证了所提数学模型与多目标优化算法的可行性与有效性。 

Multi-objective test optimization selection based on NSGA-Ⅱ under unreliable test conditions

Since test optimization selection plays a vital role in the test design of various equipment systems, in the testability design of various types of equipment, test unreliable factors seriously affect the optimization of test selection. First, this paper describes the mathematical model of the multi-objective optimization selection problem under unreliable test conditions. Second, under this mathematical model, the test cost, missed detection rate, and false alarm rate are used as the optimization goals, and the fault detection rate and isolation rate are constraints. Thus, a multi-objective optimization problem was established. Third, the NSGA-Ⅱ algorithm, a fast Non-dominated multi-objective optimization Sorting Genetic Algorithm-Ⅱ with an elite retention strategy, was proposed to optimize the proposed multi-objective problem. Using the NSGA-Ⅱ algorithm, a set of Pareto optimal solutions are obtained, and the optimal test combination can be selected according to actual needs. Finally, an example analysis is performed on a certain equipment, three sets of optimal solutions are obtained, which can meet the optimal selection under different needs, and the feasibility and effectiveness of the mathematical model and multi-objective optimization algorithm are verified.