基于网络优化算法的LAMOST焦面板观测模型

A Model for Observation Planning on Focal Plane in Lamost Project Based on Network Optimization Algorithms

在大天区多目标光纤光谱天文望远镜LAMOST系统中,为了快速获得高效率的观测规划,根据天文统计数据和工程设计状况,利用网络优化的最大流算法,构造了固有观测范围内多轮任务分配的数学模型,基于VC实现和优化了该算法模型,并对观测完备率、光纤利用率、时间复杂性、计算资源需求等方面进行全面分析。实验结果表明,算法模型,符合LAM-OST的基本观测要求,能够高速有效地实现一个焦面板上4000个观测单元对相应星象数目的观测分配。问题的解决,同时对未来解决大天区观测范围内复杂条件下的超大规模观测任务分配问题提供了重大参考。

In LAMOST project, in order to achieve an efficient observation plan in a short time, taking advantage of Max Flow Algorithm, one of the Network Optimization Algorithms, a mathematical model of multi - turn detection task assignment in a certain observation range is estabhshed, according to astronomic statistic and current engineering design data. By VC programming, the paper realizes and optimizes the model, and gains a complete analysis of the observation completion rate, the fibre - utilization rate, the time complexity and resource requirement of the algorithm. The outcome of experiment shows that, the method for task planning can achieve a highly - efficient plan for a single focal plate with 4000 observing units and 16000 target stars timely and meet the basic requirements of the project. The in - depth research of this method is also of great help to the future research of seeking a solution for detection task assignment in complex and huge observation range.