Research On Multi-satellite TT&C Resource Scheduling Models And Algorithms

With the continuous development of China’s aerospace industry,how to efficiently manage large-scale Tracking Telemetry and Command(TT&C)resources and provide efficient TT&C services for the aerospace industry is facing enormous challenges.Multisatellite TT&C resource scheduling is a scientific problem to study how to allocate TT&C resources to various satellites,and determine the working time of TT&C tasks.The multisatellite TT&C resource scheduling problem in China is a large-scale combinatorial optimization problem with complex constraints.In this paper,the model and algorithm of multi-satellite TT&C resource scheduling problem are studied,and the hybrid framework of exact algorithm and meta-heuristic method is explored.The main work and innovations are as follows:Comparing multi-satellite TT&C resource scheduling with satellite mission planning,there are common features and difficulty,and the characteristics of different types of multi-satellite TT&C resource scheduling problems are summarized.Analyzing the multi-satellite TT&C scheduling problem of China,this paper proposes that it is the multi-satellite TT&C scheduling problem with a time window that can be moved,considering user preferences,considering state constraints and relationship constraints.Considering the goal of task priority,resource preference and resource balance,a mixed integer programming(MIP)model and a constraint satisfaction problem(CSP)model are established.Compared with other studies,the mixed integer programming model established in this paper comprehensively considers the needs of high,medium and low orbit satellites and various complex constraints.Local search is a kind of efficient single-point search algorithm.The algorithm is simple in configuration,moderate in CPU time,and very effective for large-scale problems.Firstly,the core technology of local search algorithm is studied.Secondly,a two-stage local search algorithm with tabu search and late acceptance is proposed.By establishing a taboo table and a late acceptance table,the inferior solution is allowed to be accepted during the search process,and the local search is guided to jump out of the local optimum.The priority of task is considered in first stage of search,and the second stage focuses on optimizing the scheduling scheme for resource preference and resource balance.Considering the poorly convergent of meta-heuristic method and the limited by the scale of the problem for exact algorithm,a hybrid method of the exact method and the meta-heuristic method is proposed.In this paper,the mathematical programming such as branch and bound and constraint programming are analyzed in detail,and the exact method is embedded in the neighborhood search of local search.By solving the subproblems in the search process,the optimization ability of the local search is improved.According to the characteristics of multi-satellite TT&C scheduling problem,five problem decomposition methods are designs in this paper.In the part of experiment,the existing benchmarks are summarized first,and the models and algorithms proposed in the paper are tested.For the problem of multi-satellite TT&C resource scheduling in China,a simulation scenario with 165 LEO satellites and75 MEO satellites and GEO satellites a designed.Daily satellite TT&C and data transmission requirements are also generated.The experimental results verify the effectiveness of local search and hybrid framework proposed in this paper.The results show that TT&C resources of China are relatively sufficient.Therefore,it is necessary to consider user preferences and resource balance for scheduling,and this paper provides some references on scheduling methods.