基于免疫遗传优化的可靠性分配方法

Reliability Allocation Method Based on Immune Genetic Optimization

针对现有的可靠性指标分配方法合理性较低、适用范围有限的问题,提出一种基于免疫遗传优化的可靠性指标分配方法。该方法包括以下步骤:排除选型的设备和可靠度指标确定的设备,根据剩余设备的运行工况,进行系统可靠性建模;构建全寿期总成本函数,得到以总成本最低为目标的函数;随机生成初始种群,筛选种群中的抗体,采用遗传算法进行自适应交叉与变异,接着采用免疫算法,提取疫苗,对抗体进行疫苗接种,将适应度高的抗体分入新一代种群,判断新一代种群是否收敛,若是,输出分配结果,否则返回抗体筛选步骤。该方法以全寿期总成本最低为分配目标,分配过程中将免疫算法与遗传算法相结合,适用于复杂系统的任务可靠性分配。

Aiming at the problems of low rationality and limited application range of existing reliability allocation methods, a reliability allocation method based on immune genetic optimization is proposed. The method includes the following steps: excluding the equipment selected and the equipment whose reliability index has been determined, and establishing the system reliability model according to the operating conditions of the remaining equipment is obtained. The total cost function of the whole life is constructed and the function with the lowest total cost as the target is obtained. The initial population is generated randomly, the antibodies in the population are screened, adaptive crossover and variation are carried out by genetic algorithm, and then immune algorithm is used to extract vaccines and vaccinate antibodies. Highly adaptive antibodies are classified into the new generation of population to judge whether the new generation of population converges or not. If convergence occurs, output distribution results, or return to the steps of antibody screening. This method takes the lowest total cost of the whole life as the allocation objective, and combines immune algorithm with genetic algorithm in the allocation process, which is suitable for task reliability allocation of complex system.