摘要
航天器发射高频度常态化和航天产品可靠度的增加,客观上要求航天器发射采用并行调度模式以提高发射资源利用率。针对工序迭代可能引发的航天器发射系统死锁,构建了死锁预测和损失评价机制,建立了最小化任务时间和最小化加权滞后时间综合的目标函数,设计了多任务并行调度模型,开发了多类agent协同工作的优化算法,并使用调度实例验证了模型和算法的可行性和有效性。模型和算法在"天宫一号"和"神舟八号"发射任务中得到初步应用,效果较好。
As the normality of spacecraft launch with high-frequency and an improvement in reliability of space products,using the scheduling mode of parallel multitask to improve the using-rates of launch resource is an objective requirement.To avoid launch system deadlock caused by stochastic iteration of mission processes,the paper constructs the mechanism of forecasting deadlock and evaluating losses,establishes the object function of integrating minimizing completion time and minimizing total weighted tardiness time,designs a parallel scheduling model of multitask,develops a scheduling optimization algorithm of several group agents working together,and verifies the feasibility of the model and algorithm by an example.The application effects of the model and its algorithm in launching ＂Tiangong-1＂ and the ＂Shenzhou-8＂ are better.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2013年第7期1438-1444,共7页
Systems Engineering and Electronics
基金
国家自然科学基金(71071156
70971131)资助课题
关键词
航天器发射
并行调度模型
多类agent算法
随机迭代
死锁
spacecraft launch
parallel scheduling model
multi-agent algorithm
stochastic iteration
deadlock
