A multi-modal multi-target satellite task scheduling method based on a ring topology

Abstract

The present application relates to the technical field of satellite scheduling, in particular to a multi-modal multi-objective satellite task scheduling method based on a ring topology, comprising the following steps: S1: taking the minimization of task request failure rate and ground station load imbalance rate as the optimization target, constructing a multi-modal multi-objective optimization scheduling model of satellite task scheduling, and determining the problem target space; S2: determining a task available time window set according to user task request, a ground station set, a satellite-ground visible time window set and a problem constraint set, and constructing a problem decision space; S3: adopting a knowledge evolutionary algorithm iterative optimization to explore and maintain more equivalent Pareto optimal solution sets, and obtaining a high-quality scheduling scheme set, wherein the knowledge evolutionary algorithm is based on a ring topology structure decision population updating strategy. The present application can realize multi-modal multi-objective optimization of the problem, so that a group of satellite task scheduling schemes considering multiple optimization targets and differentiated from each other can be obtained after a single run.

Description

Technical Field

[0001] This invention relates to the field of satellite scheduling technology, specifically a multimodal, multi-objective satellite mission scheduling method based on ring topology. Background Technology

[0002] Satellite mission scheduling is a critical issue for ensuring the on-orbit safety of spacecraft and the completion of their missions. However, traditional satellite mission scheduling methods mostly rely on heuristic rules to construct a feasible scheduling scheme, focusing on meeting mission requirements more effectively through conflict resolution. Their problem modeling ignores the multi-objective optimization characteristics of satellite mission scheduling, making it difficult to comprehensively reflect the increasingly complex new requirements of satellite mission scheduling. Furthermore, the single scheduling schemes generated also ignore the multimodal optimization characteristics of the problem, i.e., the situation where multiple different schemes correspond to the same optimization objective value, failing to provide decision-makers with a flexible scheme selection space.

[0003] Optimization methods based on swarm intelligence theory, as metaheuristic algorithms, possess characteristics such as global optimization, parallelism, and self-organization. However, under the pressure of global environmental selection, their populations tend to prematurely converge to easily searchable local Pareto solutions, making it difficult to guarantee a balance between the population's convergence in the goal space and the diversity in the decision space. Summary of the Invention

[0004] The purpose of this invention is to provide a multimodal, multi-target satellite mission scheduling method based on ring topology to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A multimodal, multi-objective satellite mission scheduling method based on ring topology, characterized by the following steps:

[0007] S1: With the optimization objectives of minimizing the mission request failure rate and the ground station load imbalance rate, a multimodal and multi-objective optimization scheduling model for satellite mission scheduling is constructed to determine the problem objective space;

[0008] S2: Based on the user task request, the ground station set, the satellite-to-ground visible time window set, and the problem constraint set, determine the available time window set for the task and construct the problem decision space;

[0009] S3: Iterative optimization using a knowledge evolution algorithm is employed to explore and maintain more equivalent Pareto optimal solution sets, thereby obtaining a high-quality scheduling scheme set. The knowledge evolution algorithm is based on a ring topology decision population update strategy.

[0010] Preferably, the multimodal multi-objective optimization scheduling model in step 1 is as follows:

[0011] (1);

[0012] In formula (1): f 1(x) and f 2(x) represents the mission request failure rate and ground station load imbalance rate of satellite mission scheduling scheme x, respectively; R and A They represent sizes of | R | and | A |The task request set and the ground station set; L (x a ) represents the ground station in scheme x. a The workload; Let x be the average load of each ground station in scheme x.

[0013] Preferably, step S2 specifically includes:

[0014] The user task request set R = { r , s , st , et , dut , prt , swt , θ max}, where r Indicates the subtask number; s For the sub-mission target satellite; st The earliest start time of the subtask; et The latest end time for the subtask; dut Duration of the subtask; prt and swt These are the subtask preparation time and switching time, respectively. θ `max` represents the maximum elevation angle requirement for the subtask;

[0015] The ground station collection A = { a , fbst , fbet},in a Indicates the ground station number; fbst and fbet These are the set of start times and end times for disabling the corresponding ground stations, respectively.

[0016] The set of time windows visible between the satellite and the ground VTW = { vst , vet , θ},in vst , vet and θThese represent the start time, end time, and highest elevation angle of the visible time window, respectively.

[0017] Set the maximum elevation angle constraint c1:

[0018] (2);

[0019] Set time window constraint c2:

[0020] (3);

[0021] The problem decision space Ω is described as follows:

[0022] (4);

[0023] In formula (4) and Representing tasks r The lower and upper bounds of the corresponding decision variables.

[0024] Preferably, step S3 specifically includes:

[0025] S31. Generate an initial scheme population based on the preset satellite mission scheduling scheme coding rules and the set of available resources;

[0026] S32. Select the parent population according to the pairing selection strategy to generate offspring individuals;

[0027] S33. Under the constraints of ground station disabling time and non-overlapping ground station tasks, determine the start time of each task based on the decoding strategy and repair infeasible individuals.

[0028] S34. Construct a ring topology based on individual convergence and individual diversity distance, and update the population.

[0029] Preferably, step S31 specifically includes:

[0030] Solution for a single satellite mission scheduling scheme x = { x 1, x 2,…, x |R|}, its first r The coding rules for dimensional decision variables are as follows:

[0031] (5);

[0032] In formula (5) K For the task r The number of available time window resources; A value of 0 indicates that it is not in its order. kExecute within a time window, otherwise 1.

[0033] Preferably, the pairing selection strategy in step S32 specifically includes:

[0034] A restricted pairing strategy is adopted, generating a pairing pool based on the individual's neighborhood. Then, based on the binary tournament method, the parent is selected one by one from the pairing pool. The method for constructing the individual's neighborhood is to select the nearest neighbor in the decision space to the current individual. m One solution. m This is a preset domain size constant.

[0035] Preferably, step S33 specifically includes:

[0036] Set ground station disable time constraint c3:

[0037] (6);

[0038] In formula (6) est Indicates the start time of the task; prt Indicates the task preparation time;

[0039] Set ground station missions to have no overlap constraint c4:

[0040] (7);

[0041] The task execution time window is determined based on the coding values ​​of each gene position of an individual. The task start execution time is initialized to the start time of the time window. The gene position that violates the ground station time restriction constraint c3 and the ground station task non-overlap constraint c4 is detected, and the constraint can be eliminated by sliding the time window to the right. If it can be eliminated, the corresponding gene position coding value is adjusted to the current window number; otherwise, the gene position is zero.

[0042] Preferably, step S34 specifically includes:

[0043] The convergence degree Con(x) of an individual is described by neighborhood dominance information, which is the total number of neighborhood solutions dominated by the solution that dominates the current individual x in the individual's neighborhood solutions:

[0044] (8);

[0045] In equation (8) P x Let x represent the set of individuals in the neighborhood of individual x; y is a neighborhood solution of x and dominates x; z is a neighborhood solution of y and is dominated by y; |·| represents the number of elements in the set;

[0046] The individual diversity distance Div(x) is comprehensively described by the diversity of individual x in the goal space and decision space:

[0047] (9);

[0048] In equation (9), x f1 x f2 and x d1 x d2 These represent the closest and second-closest solutions to individual x in the target space and decision space, respectively. F max and F min Represents the maximum and minimum values ​​in the target space of the population; F (x) represents the individual objective function value; |·| represents the number of elements in the set; N r The dimension of the decision variables;

[0049] The population is sorted according to the individual convergence Con(x) and the individual diversity distance Div(x), and a ring topology is constructed based on the individual sorting index; the individual competition in the population update optimization is restricted to the previous and next neighborhoods in the topology, so as to realize the identification and selection of differentiated satellite mission scheduling schemes.

[0050] Compared with the prior art, the beneficial effects of the present invention are:

[0051] This invention provides a multimodal, multi-objective satellite mission scheduling method based on ring topology. The ring topology structure restricts information transfer and competition among individuals in the population, avoiding the loss of equivalent Pareto solutions due to global environmental selection pressure. Compared to traditional satellite mission scheduling methods, this invention enables multimodal, multi-objective optimization of the problem, thereby obtaining a set of satellite mission scheduling schemes that consider multiple optimization objectives and are differentiated from each other after a single run. Attached Figure Description

[0052] Figure 1 A flowchart of a multimodal, multi-objective satellite mission scheduling method based on ring topology provided by the present invention;

[0053] Figure 2 This is a flowchart illustrating the population update strategy based on a ring topology structure, as described in this invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Figure 1A flowchart illustrating a multimodal, multi-objective satellite mission scheduling method based on ring topology provided by this invention. Figure 1 As shown, an embodiment of the present invention provides a multimodal, multi-objective satellite mission scheduling method based on ring topology, comprising the following steps:

[0056] S1: With the optimization objectives of minimizing the mission request failure rate and the ground station load imbalance rate, a multimodal and multi-objective optimization scheduling model for satellite mission scheduling is constructed to determine the problem objective space;

[0057] S2: Based on the user task request, the ground station set, the satellite-to-ground visible time window set, and the problem constraint set, determine the available time window set for the task and construct the problem decision space;

[0058] S3: Iterative optimization using a knowledge evolution algorithm is employed to explore and maintain more equivalent Pareto optimal solution sets, thereby obtaining a high-quality scheduling scheme set. The knowledge evolution algorithm is based on a ring topology decision population update strategy.

[0059] This invention provides a multimodal, multi-objective satellite mission scheduling method based on ring topology. The ring topology structure restricts information transfer and competition among individuals in the population, avoiding the loss of equivalent Pareto solutions due to global environmental selection pressure. Compared to traditional satellite mission scheduling methods, this invention enables multimodal, multi-objective optimization of the problem, thereby obtaining a set of satellite mission scheduling schemes that consider multiple optimization objectives and are differentiated from each other after a single run.

[0060] In one embodiment of the present invention, the multimodal multi-objective optimization scheduling model in step 1 is as follows:

[0061] (1);

[0062] In formula (1): f 1(x) and f 2(x) represents the mission request failure rate and ground station load imbalance rate of satellite mission scheduling scheme x, respectively; R and A They represent sizes of | R | and | A |The task request set and the ground station set; L (x a ) represents the ground station in scheme x. a The workload; Let x be the average load of each ground station in scheme x.

[0063] The multi-objective nature of the multimodal multi-objective optimization scheduling model is reflected in the following: f 1(x) and f2(x) constructs a multi-dimensional target space; multimodality is reflected in the location and maintenance of multiple equivalent high-quality solutions in the scheduling process.

[0064] In one embodiment of the present invention, step S2 specifically includes:

[0065] The user task request set R = { r , s , st , et , dut , prt , swt , θ max}, where r Indicates the subtask number; s For the sub-mission target satellite; st The earliest start time of the subtask; et The latest end time for the subtask; dut Duration of the subtask; prt and swt These are the subtask preparation time and switching time, respectively. θ `max` represents the maximum elevation angle requirement for the subtask;

[0066] The ground station collection A = { a , fbst , fbet},in a Indicates the ground station number; fbst and fbet These are the set of start times and end times for disabling the corresponding ground stations, respectively.

[0067] The set of time windows visible between the satellite and the ground VTW = { vst , vet , θ},in vst , vet and θ These represent the start time, end time, and highest elevation angle of the visible time window, respectively.

[0068] Set the maximum elevation angle constraint c1:

[0069] (2);

[0070] Set time window constraint c2:

[0071] (3);

[0072] The problem decision space Ω is described as follows:

[0073] (4);

[0074] In formula (4) and Representing tasks r The lower and upper bounds of the corresponding decision variables.

[0075] Figure 2 This is a flowchart illustrating the population update strategy based on a ring topology, as described in this invention. Further, as... Figure 2 As shown, in one embodiment of the present invention, step S3 specifically includes:

[0076] S31. Generate an initial scheme population based on the preset satellite mission scheduling scheme coding rules and the set of available resources;

[0077] S32. Select the parent population according to the pairing selection strategy to generate offspring individuals;

[0078] S33. Under the constraints of ground station disabling time and non-overlapping ground station tasks, determine the start time of each task based on the decoding strategy and repair infeasible individuals.

[0079] S34. Construct a ring topology based on individual convergence and individual diversity distance, and update the population.

[0080] Based on the aforementioned multimodal and multi-objective optimization scheduling model for satellite mission scheduling, and by constructing a ring topology structure according to individual convergence and individual diversity distance, an equivalent Pareto solution set is discovered and maintained, thereby obtaining differentiated high-quality satellite mission scheduling schemes.

[0081] In one embodiment of the present invention, step S31 specifically includes:

[0082] Solution for a single satellite mission scheduling scheme x = { x 1, x 2,…, x |R|}, its first r The coding rules for dimensional decision variables are as follows:

[0083] (5);

[0084] In formula (5) K The number of available time window resources for task r; A value of 0 indicates that it is not in its order. k Execute within a time window, otherwise 1.

[0085] In one embodiment of the present invention, step S33 specifically includes:

[0086] Set ground station disable time constraint c3:

[0087] (6);

[0088] In formula (6) est Indicates the start time of the task; prt Indicates the task preparation time;

[0089] Set ground station missions to have no overlap constraint c4:

[0090] (7);

[0091] The task execution time window is determined based on the coding values ​​of each gene position of an individual. The task start execution time is initialized to the start time of the time window. The gene position that violates the ground station time restriction constraint c3 and the ground station task non-overlap constraint c4 is detected, and the constraint can be eliminated by sliding the time window to the right. If it can be eliminated, the corresponding gene position coding value is adjusted to the current window number; otherwise, the gene position is zero.

[0092] In one embodiment of the present invention, step S34 specifically includes:

[0093] The convergence degree Con(x) of an individual is described by neighborhood dominance information, which is the total number of neighborhood solutions dominated by the solution that dominates the current individual x in the individual's neighborhood solutions:

[0094] (8);

[0095] In equation (8) P x Let x represent the set of individuals in the neighborhood of individual x; y is a neighborhood solution of x and dominates x; z is a neighborhood solution of y and is dominated by y; |·| represents the number of elements in the set;

[0096] The individual diversity distance Div(x) is comprehensively described by the diversity of individual x in the goal space and decision space:

[0097] (9);

[0098] In equation (9), x f1 x f2 and x d1 x d2 These represent the closest and second-closest solutions to individual x in the target space and decision space, respectively. F max and F min Represents the maximum and minimum values ​​in the target space of the population; F (x) represents the individual objective function value; |·| represents the number of elements in the set; Nr The dimension of the decision variables;

[0099] The population is sorted according to the individual convergence Con(x) and the individual diversity distance Div(x), and a ring topology is constructed based on the individual sorting index; the individual competition in the population update optimization is restricted to the previous and next neighborhoods in the topology, so as to realize the identification and selection of differentiated satellite mission scheduling schemes.

[0100] Through iterative processes of offspring generation, attribute value calculation, and population update, a diverse and high-quality satellite mission scheduling scheme is generated.

[0101] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multimodal, multi-objective satellite mission scheduling method based on ring topology, characterized in that, Includes the following steps: S1: With the optimization objectives of minimizing the mission request failure rate and the ground station load imbalance rate, a multimodal and multi-objective optimization scheduling model for satellite mission scheduling is constructed to determine the problem objective space; S2: Based on the user task request set, ground station set, satellite-to-ground visible time window set, and problem constraint set, determine the available time window set for the task and construct the problem decision space; S3: Iterative optimization using a knowledge evolution algorithm is employed to explore and maintain more equivalent Pareto optimal solution sets and obtain a high-quality scheduling scheme set, wherein the knowledge evolution algorithm is based on a ring topology decision population update strategy. Step S2 specifically includes: The user task request set R = { r , s , st , et , dut , prt , swt , θ max}, where r Indicates the subtask number; s For the sub-mission target satellite; st The earliest start time of the subtask; et The latest end time for the subtask; dut Duration of the subtask; prt and swt These are the subtask preparation time and switching time, respectively. θ `max` represents the maximum elevation angle requirement for the subtask; The ground station collection A = { a , fbst , fbet },in a Indicates the ground station number; fbst and fbet These are the set of start times and end times for disabling the corresponding ground stations, respectively. The set of time windows visible between the satellite and the ground VTW = { vst , vet , θ },in vst , vet and θ These represent the start time, end time, and highest elevation angle of the visible time window, respectively. Set the maximum elevation angle constraint c1: (2)； Set time window constraint c2: (3)； The problem decision space Ω is described as follows: (4)； In formula (4) and Representing tasks r The lower and upper bounds of the corresponding decision variables; Step S3 specifically includes: S31. Generate an initial scheme population based on the preset satellite mission scheduling scheme coding rules and the set of available resources; S32. Select the parent population according to the pairing selection strategy to generate offspring individuals; S33. Under the constraints of ground station disabling time and non-overlapping ground station tasks, determine the start time of each task based on the decoding strategy and repair infeasible individuals. S34. Construct a circular topology based on individual convergence and individual diversity distance, update the population, and discover and maintain an equivalent Pareto solution set.

2. The multimodal, multi-objective satellite mission scheduling method based on ring topology as described in claim 1, characterized in that, The multimodal, multi-objective optimization scheduling model described in step 1 is as follows: (1)； In formula (1): f 1(x) and f 2(x) represents the mission request failure rate and ground station load imbalance rate of satellite mission scheduling scheme x, respectively; R and A These represent the task request set and the ground station set, respectively, with sizes of |R| and |A|. L (x a ) represents the ground station in scheme x. a The workload; Let x be the average load of each ground station in scheme x.

3. The multimodal, multi-objective satellite mission scheduling method based on ring topology as described in claim 2, characterized in that, Step S31 specifically includes: Solution for a single satellite mission scheduling scheme x = { x 1, x 2,…, x |R| }, its first r The coding rules for dimensional decision variables are as follows: （5）； In formula (5) K For the task r The number of available time window resources; A value of 0 indicates that it is not in its order. k Execute within a time window, otherwise 1.

4. The multimodal, multi-objective satellite mission scheduling method based on ring topology as described in claim 3, characterized in that, The pairing selection strategy described in step S32 specifically includes: A restricted pairing strategy is adopted, generating a pairing pool based on the individual's neighborhood. Then, based on the binary tournament method, the parent is selected one by one from the pairing pool. The method for constructing the individual's neighborhood is to select the nearest neighbor in the decision space to the current individual. m One solution. m This is a preset domain size constant.

5. The multimodal, multi-objective satellite mission scheduling method based on ring topology as described in claim 4, characterized in that, Step S33 specifically includes: Set ground station disable time constraint c3: （6）； In formula (6) est Indicates the start time of the task; prt Indicates the task preparation time; Set ground station missions to have no overlap constraint c4: （7）； The task execution time window is determined based on the coding values ​​of each gene position of an individual. The task start execution time is initialized to the start time of the time window. The gene position that violates the ground station time restriction constraint c3 and the ground station task non-overlap constraint c4 is detected, and the constraint can be eliminated by sliding the time window to the right. If it can be eliminated, the corresponding gene position coding value is adjusted to the current window number; otherwise, the gene position is zero.

6. The multimodal, multi-objective satellite mission scheduling method based on ring topology as described in claim 5, characterized in that, Step S34 specifically includes: The convergence degree Con(x) of an individual is described by neighborhood dominance information, which is the total number of neighborhood solutions dominated by the solution that dominates the current individual x in the individual's neighborhood solutions: （8）； In equation (8) P x Let x represent the set of individuals in the neighborhood of individual x; y is a neighborhood solution of x and dominates x; z is a neighborhood solution of y and is dominated by y; |·| represents the number of elements in the set; The individual diversity distance Div(x) is comprehensively described by the diversity of individual x in the goal space and decision space: （9）； In equation (9), x f1 x f2 and x d1 x d2 These represent the closest and second-closest solutions to individual x in the target space and decision space, respectively. F max and F min Represents the maximum and minimum values ​​in the target space of the population; F (x) represents the individual objective function value; |·| represents the number of elements in the set; N r The dimension of the decision variables; The population is sorted according to the individual convergence Con(x) and the individual diversity distance Div(x), and a ring topology is constructed based on the individual sorting index; the individual competition in the population update optimization is restricted to the previous and next neighborhoods in the topology, so as to realize the identification and selection of differentiated satellite mission scheduling schemes.

Images