Random search method for relay satellite single-access antenna scheduling based on collision resolution

Abstract

The invention belongs to the satellite scheduling field and relates to a relay satellite single-address antenna scheduling random search method based on conflict resolution. The method comprises the following steps: (S1) matching the relay satellite resources for the task according to the alternative service time window information submitted by the task; 12) generating an initial solution of a relay satellite single-access antenna scheduling random search accord to a task resource matching result; (3) evaluating the fitness of the current solution, finding out the key task causing the conflict, and generating a perturbation -deletion pool of the task; (4) randomly perturbing or deleting tasks in the perturbation-deletion pool, and gradually iteratively resolving task conflicts; (S5) updating resource and task sets: deleting the resource and task without conflict and updating the resource and task set. The method of the invention has high speed, is suitable for large-scale relay satellite dispatching scene, adopts double iterative structure, and generates a dispatching scheme with high quality.

Description

technical field [0001] The invention belongs to the field of satellite scheduling, and in particular relates to a random search method for relay satellite single-site antenna scheduling based on conflict resolution. Background technique [0002] Tracking and Data Relay Satellite (abbreviation: TDRS), referred to as relay satellite, mainly provides data relay, continuous tracking and orbit measurement and control services for medium and low orbit spacecraft. Since its establishment, the relay satellite system has played an irreplaceable role in supporting information-based operations, national defense scientific research experiments, and civil applications. With the continuous expansion of the scope of application, the number of tasks faced by the relay satellite system has increased year by year, and the difficulty of application and scheduling has gradually increased. How to realize the flexible management and efficient utilization of relay satellite resources has become t...

AI Technical Summary

Problems solved by technology

For this model, the above algorithms are difficult to use directly, and there are mainly the following problems: First, the applicability of the intelligent optimization algorithm to high-dimensional problems is obviously stronger than that of the exact solution algorithm [16] , but it is still difficult to apply to the ultra-high dimension of the model. Excessively high dimension will make the solution matrix very sparse, which makes the algorithm face difficulties in the search of the solution space, the mutual learning between population individuals and the evolution of the population, and the operation of the algorithm is also difficult. It will take a lot of time; second, the model is a high-dimensional mixed integer optimization problem with complex constraints. The above algorithm is generally applicable to integer optimization problems or continuous optimization problems, and is not very applicable to mixed integer optimization problems [17] , there are difficulties in operations such as encoding, crossover, and mutation; third, the model constraints are complex, and it is difficult to construct multiple feasible solutions as solution populations and maintain the feasibility of solutions during the evolution process; fourth, the model is different from general optimization The problem is different, the quality of the solution is only reflected in the overall level of the scheme, there is no difference in the quality of a certain decision variable or the scheduling scheme of a certain task, which is different for the differential evolution algorithm, which emphasizes mutual learning between population individuals In other words, the effect will be greatly affected by the

Although the common intelligent optimization algorithm cannot be directly applied to the solution of the model in this paper, it is based on the random search idea under the guidance of the objective function value or some phenomena. [18] , still has a strong reference value and reference significance for solving high-dimensional optimization problems under complex constraints

Images