基于遗传算法的运载火箭总线消息调度优化设计

运载火箭控制系统采用1553B总线、按照预先设计的调度表传输周期性消息。针对当前总线消息调度表的设计方法对负载均衡考虑较少的缺陷,在构建1553B总线消息周期调度表数学模型的基础上,引入遗传算法进行最优化设计。该算法将负载率参数的多目标优化问题转换为遗传个体适应度的单目标优化问题,采用“精英保留”遗传策略,有效避免了个体退化,算法稳定高效。数学仿真计算结果表明,总线消息周期调度表求解高效,负载均衡大幅提高,验证了优化设计方法的有效性。

The Optimization Design of Launch Vehicle Bus Messages Scheduling Basedon Genetic Algorithm

The control system of launch vehicle employed 1553B to transmit the static periodic messages. First, the problem of bus message transmission mathematic model was constructed as periodic polling table. In view of the lack of load balancing of existing design schemes of bus messages scheduling, the optimized design method introduced the genetic algorithm to search the best result. The design scheme transmits the multi-objective load parameter optimization to single-objective optimization of the fitness of the evolved individual. The elitist selection policy was used to prevent the individual degradation and to ensure the stabilization and efficiency of the algorithm. The computing results prove that the bus messages scheduling method is easy to solve and improves the load balancing efficiently.