Satellite component layout optimization method and system

Abstract

The application discloses a satellite assembly layout optimization method and system, and relates to the technical field of assembly layout. The method comprises the following steps: calculating a set of separation axes between cabin plates based on the edge vectors of the cabin plates; calculating the size and geometric center coordinates of each assembly after expansion according to the preset safety distance of each assembly; constructing a geometric constraint model including geometric non-interference constraints, assembly installation position constraints and assembly attitude uniqueness constraints based on the set of separation axes and the assemblies after expansion; and constructing and solving a mixed integer linear programming model with the geometric constraint model as the constraint condition to obtain an assembly layout scheme satisfying a preset objective function. The application can realize the optimization design of assembly layout when the cabin plates are in an orthogonal configuration or a non-orthogonal configuration, and can also consider the design requirement that the assemblies need to reserve space to meet the cable layout requirement, thereby significantly improving the optimization design effect and efficiency of assembly layout under a complex cabin plate configuration, and the design scheme has higher engineering feasibility.

Description

Technical Field

[0001] This invention relates to the field of component layout technology, and in particular to a method and system for optimizing satellite component layout. Background Technology

[0002] Satellite layout plays a crucial role in determining its on-orbit performance and functionality. The purpose of satellite layout design is to arrange electronic components or equipment in appropriate locations within the satellite to meet various system performance requirements, such as mass characteristics and thermal control. As a vital part of the overall satellite design, satellite layout design directly determines the overall performance, development cost, design cycle, and design level of the satellite system.

[0003] As satellite internal structures become increasingly complex and layout spaces become more compact, the component layout areas on satellites used to house components are evolving into three-dimensional layout areas, which may further be divided into multiple sub-regions by various panels. To address this, existing component layout optimization methods typically model the component layout optimization design problem as two coupled optimization sub-problems: component allocation and detailed component placement. Specifically, the allocation optimization between components and layout sub-regions is considered first, and then the specific location of the components within each layout sub-region is determined.

[0004] However, when using the above methods to optimize component layout, the solution efficiency is low due to the large number of discrete allocation variables involved in the optimization of components and layout sub-regions, and it is difficult to obtain the optimal layout result. Furthermore, existing component layout optimization methods assume that the cabins are orthogonally distributed, meaning that different cabins within the layout area are parallel or perpendicular to each other. In practical applications, multiple cabins within the layout area may also be non-orthogonally distributed, meaning that different cabins within the layout area are not parallel or perpendicular to each other. When a cabin is non-orthogonally distributed to other cabins, existing component layout optimization methods are not applicable.

[0005] Furthermore, since components and the cabling between them may need to be laid out, existing component layout optimization methods typically involve completing the component layout optimization design and then fine-tuning the component positions to meet wiring constraints, or directly expanding the component's outer envelope and then optimizing the component layout based on the expanded outer envelope. However, if the method of completing the component layout optimization design and then fine-tuning the component positions to meet wiring constraints is adopted, there may be situations where insufficient space is reserved between components for cable routing, resulting in cables not being able to connect to the components properly. Consequently, it is impossible to find a layout scheme that can further meet the wiring constraints by fine-tuning the component positions. If the method of directly expanding the component's outer envelope is adopted, since not every side of the component needs to reserve space, usually only the side with sockets needs to be considered for reserving space, directly expanding the component's outer envelope will simultaneously expand all sides of the component by a certain size, which will seriously waste the compact layout space.

[0006] Therefore, it is necessary to propose a satellite component layout optimization method and system that can achieve component layout optimization design when the cabin is orthogonal or non-orthogonal, and can also take into account the design requirements that the components need to reserve space to meet the cable laying requirements. Summary of the Invention

[0007] To address some or all of the technical problems existing in the prior art, the present invention provides a satellite component layout optimization method and system, which can realize component layout optimization design when the cabin is orthogonal, and can also realize component layout optimization design when the cabin is non-orthogonal, and can also consider the design requirements that the components need to reserve space to meet the cable laying requirements.

[0008] The technical solution of the present invention is as follows:

[0009] Firstly, a satellite component layout optimization method is provided, the method being used for component layout optimization when the layout area consists of multiple panels, including:

[0010] Based on the edge vectors of each compartment, the separation axis set between the compartments is calculated, the separation axis set including the axial directions defined by the edge vectors of the compartments and their cross product vectors;

[0011] Based on the preset safety distance of each component, calculate the dimensions and geometric center coordinates of each component after expansion;

[0012] Based on the set of separate axes and the expanded components, a geometric constraint model is constructed, including geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. The geometric non-interference constraints include non-interference constraints between components and non-interference constraints between components and the cabin plate. The component installation position constraints are used to limit the components to be completely located within the boundary of their respective cabin plates. The component attitude uniqueness constraints are used to limit the components to be installed on only one cabin plate and to have only one placement orientation. The geometric non-interference constraints are constructed in the following way: the corresponding inequality constraints are constructed using the geometric centers of two components, or the geometric centers of the components and the cabin plate on at least one separation axis, as constraints. The inequality constraints are linearized using the method of large numbers. The separation axis selection variables are coupled with the component-cabinet allocation variables to ensure that the components satisfy the corresponding inequality constraints when they are installed.

[0013] Using the geometric constraint model as the constraint condition, a mixed integer linear programming model is constructed and solved to obtain a component layout scheme that satisfies the preset objective function.

[0014] Furthermore, in some implementations, the dimensions and geometric center coordinates of the expanded component are calculated in the following manner:

[0015] The length of the expanded component is the initial length of the component plus a safety distance along the length direction;

[0016] The width of the expanded component is the component's initial width plus a safety distance along the width direction.

[0017] The height of the expanded component is the component's initial height plus a safety distance along the height direction.

[0018] The extended geometric center coordinates of the component are determined by superimposing the offset calculated from the safety distance difference onto the component's initial geometric center coordinates.

[0019] Furthermore, in some implementations, the non-interference constraints between the components are constructed as follows:

[0020] ;

[0021] in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the j-th component after expansion. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the j-th component on the k-th separation axis after expansion. This represents a preset positive integer. and Choose a variable for the first separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between components. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between components. and Assign variables to the component-cabinet. , , This indicates that the i-th component is installed in the i-th position. On the deck. This indicates that the j-th component is installed on the h-th compartment panel. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

[0022] Furthermore, in some embodiments, the non-interference constraint between the component and the cabin panel is constructed as follows:

[0023] ;

[0024] in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the nth deck. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the nth compartment on the kth separation axis. This represents a preset positive integer. and Choose a variable for the second separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between the component and the cabin plate. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between the component and the cabin plate. Assign variables to the component-cabinet. , This indicates that the i-th component is installed in the i-th position. On the deck. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

[0025] Furthermore, in some implementations, the component installation location constraints are constructed in the following manner:

[0026] The dimensional differences between the extended component and its corresponding compartment in the length and width directions, and the sum of their dimensions in the height direction, shall not be less than the projected distance between their geometric centers in the corresponding directions.

[0027] Furthermore, in some embodiments, the component installation location constraints are constructed as follows:

[0028] ;

[0029] in, , , Let represent the three edge vectors corresponding to the nth cabin panel. This represents the geometric center coordinates of the nth deck. This represents the geometric center coordinates of the i-th component after expansion. This represents the length of the nth cabin panel. This represents the width of the nth deck. This represents the height of the nth deck. This represents the dimension of the i-th expanded component along the length of the hatch when it is installed on the hatch. This represents the dimension of the i-th expanded component in the width direction of the compartment when it is installed on the compartment panel. This represents the dimension of the i-th expanded component in the height direction of the compartment when it is installed on the compartment panel. This represents a preset positive integer. Assign variables to the component-cabinet. , This indicates that the i-th component is installed on the n-th compartment.

[0030] Furthermore, in some embodiments, the objective function is a multi-objective composite function, which balances at least two of the following optimization objectives through a weighted approach:

[0031] The centroid optimization objective is used to minimize the positional deviation between the overall centroid of the component and the preset target centroid.

[0032] The goal of heat dissipation optimization is to prioritize the allocation of higher power consumption components to compartments with lower heat accumulation coefficients.

[0033] Furthermore, in some embodiments, the method further includes:

[0034] The set of separated axes is filtered to remove duplicate separated axes.

[0035] Furthermore, in some embodiments, the mixed-integer linear programming model is solved using an integer programming algorithm, or by using the following methods:

[0036] For mixed-integer linear programming models, an initial population containing multiple individuals is randomly generated, where each individual represents a component layout scheme, and one individual corresponds to one component layout scheme.

[0037] For each generation of the population, the population is divided into multiple species using an improved niche algorithm. Differential evolution algorithm is used to update the species. Based on the updated population, information-guided local search strategy is used to perform local search operations on individuals in the population, and individuals are updated according to the results of the local search operations. Based on the updated population again, information-guided reinforcement evolution mechanism is used to perform reinforcement evolution operations on the inferior individuals in the population, and inferior individuals are updated according to the results of the reinforcement evolution operations. The updated population is used as the next generation population to continue the population iteration update until the set number of population iteration updates is completed.

[0038] Based on the updated population, obtain multiple component layout schemes.

[0039] Secondly, a satellite component layout optimization system is also provided, the system being used for component layout optimization when the layout area consists of multiple panels, including:

[0040] The separation axis pre-calculation module is used to calculate the separation axis set between the compartments based on the edge vectors of each compartment. The separation axis set includes the axis directions defined by the edge vectors of the compartments and their cross product vectors.

[0041] The safety distance processing module is used to calculate the dimensions and geometric center coordinates of each component after expansion, based on the preset safety distance of each component.

[0042] The constraint construction module is used to construct a geometric constraint model based on the set of separation axes and the expanded components. This model includes geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. The geometric non-interference constraints include non-interference constraints between components and non-interference constraints between components and the cabin. The component installation position constraints are used to limit the components to be completely located within the boundaries of their respective cabins. The component attitude uniqueness constraints are used to limit the components to be installed on only one cabin and to have only one placement orientation. The geometric non-interference constraints are constructed as follows: the corresponding inequality constraints are constructed using the geometric centers of two components, or the geometric centers of the components and the cabin on at least one separation axis, as constraints. The inequality constraints are linearized using the method of large numbers. The separation axis selection variables are coupled with the component-cabinet allocation variables to ensure that the components satisfy the corresponding inequality constraints when they are installed.

[0043] The optimization and solution module is used to construct and solve a mixed-integer linear programming model using the geometric constraint model as a constraint condition, so as to obtain a component layout scheme that satisfies the preset objective function.

[0044] The main advantages of the technical solution of this invention are as follows:

[0045] The satellite component layout optimization method and system of this invention introduces a pre-calculation mechanism based on the separation axis set of the edge vectors of the cabin, treating the cabin as a "virtual component" with a fixed position. By coupling the component-cabinet allocation variables, the complex non-orthogonal three-dimensional geometric interference constraint problem can be transformed into a computable projection relationship and linearized. At the same time, combined with the geometric extension modeling of the component's safety distance, a geometric constraint model is constructed, including geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. Based on the geometric constraint model, a mixed integer linear programming model is constructed and solved. This enables the component layout optimization design when the cabin is orthogonal or non-orthogonal. It can also consider the design requirements that the component needs to reserve space to meet the cable laying requirements. This significantly improves the effect and efficiency of component layout optimization design under complex cabin configurations, and the design scheme has higher engineering feasibility. Attached Figure Description

[0046] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and constitute a part of this invention, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0047] Figure 1 A flowchart illustrating a satellite component layout optimization method provided in an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of a layout area provided in an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of another layout area provided in an embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of a satellite component layout optimization system provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] The technical solutions provided by the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] refer to Figure 1-3 In a first aspect, embodiments of the present invention provide a satellite component layout optimization method. This method is used for component layout optimization when the layout area consists of multiple panels, and includes the following steps:

[0054] Step 1: Based on the edge vectors of each compartment, calculate the separation axis set between the compartments. The separation axis set includes the axial directions defined by the edge vectors of the compartments and their cross product vectors.

[0055] Step 2: Calculate the dimensions and geometric center coordinates of each expanded component based on the preset safety distance of each component;

[0056] Step 3: Based on the separated axis set and the expanded components, construct a geometric constraint model including geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. Geometric non-interference constraints include non-interference constraints between components and non-interference constraints between components and the cabin plate. Component installation position constraints are used to limit the components to be completely located within the boundaries of their respective cabin plates. Component attitude uniqueness constraints are used to limit the components to be installed on only one cabin plate and to have only one placement orientation. Geometric non-interference constraints are constructed in the following way: the corresponding inequality constraints are constructed using the geometric centers of two components, or the geometric centers of the components and the cabin plate on at least one separation axis, as constraints. The inequality constraints are linearized using the method of large numbers, and coupled with the component-cabinet allocation variables through separation axis selection variables, so that the components satisfy the corresponding inequality constraints when they are installed.

[0057] Step 4: Using the geometric constraint model as the constraint condition, construct and solve the mixed integer linear programming model to obtain the component layout scheme that satisfies the preset objective function.

[0058] In this embodiment of the invention, the multiple panels constituting the layout area can be orthogonally distributed, or non-orthogonally distributed, or partly orthogonally distributed and partly non-orthogonally distributed.

[0059] When the cabin panels are non-orthogonally distributed, their tilt angles differ, requiring components to be installed to match the panel tilt attitude. This results in arbitrary oblique angles in the geometric relationships between components and between components and cabin panels. In this embodiment of the invention, by utilizing the aforementioned method of constructing geometrically non-interference constraints, the complex non-orthogonal three-dimensional geometric interference constraint problem can be transformed into a computable projection relationship and linearized.

[0060] In this embodiment of the invention, the separated axis set does not contain zero vectors; that is, if the cross product vector of the edge vectors of the cabin plate is a zero vector, then the cross product vector is removed.

[0061] In this embodiment of the invention, the edge variables of the cabin plate represent a set of basic direction vectors used to mathematically describe the orientation and attitude of the cabin plate in three-dimensional space. For example, for a cabin plate with a cuboid or hexahedral shape, the cabin plate includes three mutually perpendicular edge vectors, which correspond to the length direction, width direction, and height direction of the cabin plate, respectively.

[0062] In this embodiment of the invention, the edge vector of the cabin is predetermined based on the structural dimensions of the cabin, its position in the layout area, and a preset global coordinate system.

[0063] The global coordinate system is set according to the specific circumstances, for example... Figure 2-3 In this system, different global coordinate systems o-xyz are set for different types of layout areas.

[0064] In this embodiment of the invention, the safety distance of the components is predetermined according to actual needs. Specifically, when no cables need to be laid on a certain side of the component, the safety distance corresponding to that side can be set to 0.

[0065] In this embodiment of the invention, the objective function is set according to the optimization objective required by actual needs, and the geometric constraint model is used as the constraint condition. Based on the constraint condition and the objective function, a mixed integer linear programming model is constructed and solved to obtain a component layout scheme that satisfies the preset objective function.

[0066] The satellite component layout optimization method provided in this invention introduces a pre-calculation mechanism based on the separation axis set of the edge vectors of the cabin, treating the cabin as a "virtual component" with a fixed position. By coupling the component-cabinet allocation variables, the complex non-orthogonal three-dimensional geometric interference constraint problem can be transformed into a computable projection relationship and linearized. At the same time, combined with the geometric extension modeling of the component's safety distance, a geometric constraint model is constructed, including geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. Based on the geometric constraint model, a mixed integer linear programming model is constructed and solved. This method can realize the component layout optimization design when the cabin is orthogonal or non-orthogonal. It can also consider the design requirements that the component needs to reserve space to meet the cable laying requirements. This method can significantly improve the effect and efficiency of component layout optimization design under complex cabin configurations, and the design scheme has higher engineering feasibility.

[0067] Furthermore, in this embodiment of the invention, considering that the set of separation axes is determined by the edge vectors of the cabin and their cross product vectors, the corresponding set of separation axes may include duplicate separation axes. Therefore, the set of separation axes is filtered to remove duplicate separation axes, thereby reducing the amount of computation in the subsequent optimization process.

[0068] Furthermore, in this embodiment of the invention, in order to facilitate calculation and modeling, improve optimization efficiency, and ensure the feasibility of the obtained component layout scheme, the components and cabins are simplified into square bodies with uniform mass distribution.

[0069] The components and panels can be described using the outer envelope cube of the components and panels.

[0070] Furthermore, in this embodiment of the invention, the dimensions and geometric center coordinates of the extended component are calculated in the following manner:

[0071] The length of the expanded component is the initial length of the component plus a safety distance along the length direction;

[0072] The width of the expanded component is the component's initial width plus a safety distance along the width direction.

[0073] The height of the expanded component is the component's initial height plus a safety distance along the height direction.

[0074] The extended geometric center coordinates of the component are determined by superimposing the offset calculated from the safety distance difference onto the component's initial geometric center coordinates.

[0075] In this embodiment of the invention, the correspondence between the size and geometric center coordinates of the component and the size and geometric center coordinates of the expanded component can be determined by the above method.

[0076] Furthermore, in this embodiment of the invention, based on the above-defined construction method of geometric non-interference constraints, taking the i-th component and the j-th component as examples, when the two components satisfy the non-interference constraint, the two components satisfy the following inequality on at least one separating axis:

[0077] ;

[0078] By linearizing the above inequality constraints using the method of large numbers, and coupling the separation axis selection variable with the component-cabinet allocation variable, the following non-interference constraints between components can be obtained:

[0079] ;

[0080] in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the j-th component after expansion. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the j-th component on the k-th separation axis after expansion. This represents a preset positive integer. and Choose a variable for the first separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between components. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between components. and Assign variables to the component-cabinet. , , This indicates that the i-th component is installed in the i-th position. On the deck. This indicates that the j-th component is installed on the h-th compartment panel. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

[0081] In the aforementioned non-interference constraints between components, by setting This ensures that when the i-th component and the j-th component are installed on the cabin plate, the above-mentioned inequality constraints are satisfied on at least one separation axis, that is, the components satisfy the non-interference constraint.

[0082] It should be noted that, This indicates that the positive direction of the k-th separation axis is not selected to satisfy the non-interference constraint between components. This indicates that the opposite direction of the k-th separation axis is not chosen to satisfy the non-interference constraint between components. This indicates that the i-th component is not installed in the i-th position. On the deck. This indicates that the j-th component is not installed on the h-th compartment.

[0083] Furthermore, in this embodiment of the invention, based on the above-defined geometric non-interference constraint construction method, taking the i-th component and the n-th compartment as examples, when the non-interference constraint is satisfied between the component and the compartment, the component and the compartment satisfy the following inequality on at least one separation axis:

[0084] ;

[0085] By linearizing the above inequality constraints using the method of large numbers, and coupling the separation axis selection variable with the component-cabinet allocation variable, the following non-interference constraints between the component and the cabin can be obtained:

[0086] ;

[0087] in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the nth deck. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the nth compartment on the kth separation axis. This represents a preset positive integer. and Choose a variable for the second separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between the component and the cabin plate. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between the component and the cabin plate. Assign variables to the component-cabinet. , This indicates that the i-th component is installed in the i-th position. On the deck. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

[0088] In the aforementioned non-interference constraints between components and cabin panels, by setting This ensures that when the i-th component is installed on the compartment plate, the above-mentioned inequality constraint is satisfied on at least one separation axis, that is, the component and the compartment plate satisfy the non-interference constraint.

[0089] It should be noted that, This indicates that the positive direction of the k-th separation axis is not selected to satisfy the non-interference constraint between the component and the cabin plate. This indicates that the opposite direction of the k-th separation axis is not chosen to satisfy the non-interference constraint between the component and the cabin plate. This indicates that the i-th component is not installed in the i-th position. On the cabin deck.

[0090] Furthermore, in this embodiment of the invention, the component installation position constraints are constructed in the following manner:

[0091] The dimensional differences between the extended component and its corresponding compartment in the length and width directions, and the sum of their dimensions in the height direction, shall not be less than the projected distance between their geometric centers in the corresponding directions.

[0092] Based on the above-mentioned construction method with constraints on component installation positions, taking the i-th component and the n-th compartment as examples, when the i-th component is installed on the n-th compartment, the component and the compartment must satisfy the following inequality:

[0093] ;

[0094] By linearizing the above inequality constraints using the method of large numbers and coupling them through component-cabinet allocation variables, the following component installation position constraints can be obtained:

[0095] ;

[0096] in, , , Let represent the three edge vectors corresponding to the nth cabin panel. This represents the geometric center coordinates of the nth deck. This represents the geometric center coordinates of the i-th component after expansion. This represents the length of the nth cabin panel. This represents the width of the nth deck. This represents the height of the nth deck. This represents the dimension of the i-th expanded component along the length of the hatch when it is installed on the hatch. This represents the dimension of the i-th expanded component in the width direction of the compartment when it is installed on the compartment panel. This represents the dimension of the i-th expanded component in the height direction of the compartment when it is installed on the compartment panel. This represents a preset positive integer. Assign variables to the component-cabinet. , This indicates that the i-th component is installed on the n-th compartment panel. This indicates that the i-th component is not installed on the n-th compartment.

[0097] Furthermore, in this embodiment of the invention, the component orientation uniqueness constraint includes: the component can only be installed on one panel, and the component can only be installed in one orientation.

[0098] In this embodiment of the invention, the constraint that a component can only be installed on one compartment is limited by the component-compartment allocation variable set above, which can be specifically expressed as follows: .

[0099] In actual component layout optimization, the mounting surface of the component is fixed, but the component can be placed horizontally or vertically. At the same time, considering that each surface of the component has a corresponding safety distance, for components simplified into a square, the component has four placement directions on the cabin plate.

[0100] Therefore, in this embodiment of the invention, the constraint that a component can only be installed in one placement direction is limited by setting a placement direction selection variable, specifically expressed as follows: , Choose a variable for the placement direction of the i-th component. , , This indicates that the i-th component is placed in the m-th orientation. This indicates that the i-th component does not choose the m-th placement orientation.

[0101] Furthermore, in this embodiment of the invention, based on the placement direction selection variable set above, the relationship between the dimensions of the component in each direction of the cabin plate and the dimensions of the component when the component is installed on the cabin plate can be expressed as:

[0102] ;

[0103] in, This represents the length of the i-th component after expansion. This represents the width of the i-th component after expansion. This represents the height of the i-th component after expansion.

[0104] Furthermore, in this embodiment of the invention, the objective function is a single-objective function or a multi-objective composite function.

[0105] Specifically, when the objective function is a single objective function, it can be one of the centroid optimization objective function or the heat dissipation performance optimization objective function.

[0106] When the objective function is a multi-objective composite function, the multi-objective composite function balances at least two of the following optimization objectives through a weighted approach: centroid optimization objective and heat dissipation performance optimization objective.

[0107] In this embodiment of the invention, the centroid optimization target is used to minimize the positional deviation between the overall centroid of the component and the preset target centroid; the heat dissipation performance optimization target is used to prioritize the allocation of components with higher power consumption to the compartment with a lower heat accumulation coefficient.

[0108] It should be noted that the better the heat dissipation performance of the compartment, the lower the corresponding heat accumulation coefficient of the compartment.

[0109] Furthermore, in this embodiment of the invention, the centroid optimization objective function is constructed as follows:

[0110] ;

[0111] in, This represents the centroid optimization objective function. Indicates the overall centroid coordinates of the component. This indicates the coordinates of the preset target centroid.

[0112] Among them, the overall centroid coordinates of the component Determined in the following ways:

[0113] ;

[0114] in, Indicates the quality of the i-th component. This represents the geometric center coordinates of the i-th component, i.e., the geometric center coordinates of the i-th component before expansion.

[0115] In this embodiment of the invention, the relationship between the geometric center coordinates of the components before and after the expansion can be expressed as:

[0116] ;

[0117] in, This represents the coordinate offset corresponding to the i-th component.

[0118] The coordinate offset is determined based on the component's safety distance, the component's mounting plate, and the component's placement orientation.

[0119] Furthermore, in this embodiment of the invention, the objective function for optimizing heat dissipation performance is constructed as follows:

[0120] ;

[0121] in, This represents the objective function for optimizing heat dissipation performance. This represents the power consumption of the i-th component. Indicates the first The thermal accumulation coefficient of each compartment plate.

[0122] The power consumption of the components and the thermal accumulation coefficient of the cabin are predetermined based on the actual situation.

[0123] Furthermore, in this embodiment of the invention, the multi-objective composite function is constructed as follows:

[0124] ;

[0125] in, Represents a multi-objective composite function. and This refers to the weighting coefficient, and the value of the weighting coefficient is set according to actual needs.

[0126] Furthermore, based on the above analysis, taking a multi-objective composite function as an example, the mixed-integer linear programming model is constructed as follows:

[0127] ;

[0128] in, This indicates the component layout scheme.

[0129] Furthermore, in this embodiment of the invention, an integer programming algorithm is used to solve the mixed integer linear programming model.

[0130] Specifically, in this embodiment of the invention, an existing mature mixed-integer programming solver is used to solve the constructed mixed-integer linear programming model to obtain the corresponding component layout scheme and determine the mounting plate, placement direction, and position of each component. Examples of mixed-integer programming solvers include the SCIP optimization solver and the CPLEX optimization solver.

[0131] Furthermore, when solving mixed-integer linear programming models using integer programming algorithms—that is, using existing mature mixed-integer programming solvers—only one component layout scheme can generally be obtained. However, satellite component layout optimization problems may have multiple optimal or engineering-satisfactory solutions. To achieve a "good, fast, and cost-effective" satellite component layout design, shorten the design cycle, reduce development costs, and improve design reliability, it is necessary to search for multiple optimal or engineering-satisfactory solutions to meet practical engineering requirements.

[0132] In order to quickly obtain multiple feasible and optimal component layout schemes to meet actual engineering needs, in another embodiment of the present invention, the mixed-integer linear programming model is solved using the following method:

[0133] Step 41: For the mixed integer linear programming model, randomly generate an initial population containing multiple individuals, where each individual represents a component layout scheme, and one individual corresponds to one component layout scheme.

[0134] Step 42: For each generation of the population, the population is divided into multiple species using the improved niche algorithm, and the species are updated using the differential evolution algorithm. Based on the updated population, the information-guided local search strategy is used to perform local search operations on the individuals in the population, and the individuals are updated according to the results of the local search operations. Based on the updated population again, the information-guided reinforcement evolution mechanism is used to perform reinforcement evolution operations on the inferior individuals in the population, and the inferior individuals are updated according to the results of the reinforcement evolution operations. The updated population is used as the next generation population to continue the population iteration update until the set number of population iteration updates is completed.

[0135] Step 43: Obtain multiple component layout schemes based on the updated population.

[0136] In this embodiment of the invention, the initial population is updated three times as described above to obtain the updated population as the next generation population. Then, the next generation population is updated three times as described above to obtain the updated population as the next generation population, and the iterative updates continue until the set number of population iterations is completed to obtain the final population.

[0137] In this embodiment of the invention, the number of population iterations is specifically set according to actual needs.

[0138] In this embodiment of the invention, the population size is the same across different generations.

[0139] In this embodiment of the invention, based on the population obtained from the final update, several superior individuals in the population are selected as the final determined multiple component layout schemes.

[0140] In this embodiment of the invention, by using the component layout scheme as an individual to generate a population based on the constructed mixed integer linear programming model, and by using the improved niche algorithm, the information-guided local search strategy and the information-guided enhanced evolution mechanism to iteratively update the population, multiple feasible and better component layout schemes can be solved quickly to meet different practical engineering needs.

[0141] Furthermore, in this embodiment of the invention, the population is divided into multiple species using an improved niche algorithm, including the following steps:

[0142] Step 4201: Sort all individuals in the population according to feasibility rules;

[0143] Step 4202: For individuals other than the optimal individual, connect each individual to the nearest better individual to form a spanning tree rooted at the globally optimal individual;

[0144] Step 4203: Delete connected edges whose length exceeds the average margin, and treat a connected subgraph as a species;

[0145] Step 4204: Species with fewer individuals than the preset minimum species size are merged into the nearest neighbor species to obtain the final division of multiple species.

[0146] In this embodiment of the invention, the minimum species size is set according to actual needs, for example, it is set to 5% of the population size.

[0147] In this embodiment of the invention, the feasibility rule is defined as:

[0148] When constraint violation values ​​are different, individuals with lower constraint violation values ​​are better; when constraint violation values ​​are the same, individuals with lower objective function values ​​are better.

[0149] In this embodiment of the invention, the constraint violation value represents the degree to which the component layout scheme does not meet the constraint conditions, and is defined as:

[0150] ;

[0151] in, Indicates a constraint violation value. This represents the weight corresponding to the k-th constraint. This represents the violation value corresponding to the k-th constraint. This represents the given number of constraints.

[0152] In this embodiment of the invention, the constraints used when calculating the constraint violation value include at least one of the following: non-interference constraints between components, non-interference constraints between components and cabins, component installation position constraints, and component attitude uniqueness constraints.

[0153] The calculation method for the violation value corresponding to different constraints is set according to actual needs.

[0154] Specifically, taking the non-interference constraint between components as an example, the violation value corresponding to the non-interference constraint between components can be expressed as:

[0155] .

[0156] In this embodiment of the invention, the objective function value corresponding to an individual is calculated according to the objective function set above.

[0157] In this embodiment of the invention, by using the improved niche algorithm described above for species division, the minimum size of each species can be guaranteed, and the species division and clustering effect can be improved. It can better adapt to the diverse individual distribution in different problems, avoid the strong sensitivity of performance to parameters, has better robustness, and does not require prior knowledge of the optimization problem (such as the number of peaks), which is very suitable for the multimodal and unknown peak number characteristics of component layout optimization problems.

[0158] Furthermore, in this embodiment of the invention, updating the species using the differential evolution algorithm includes the following steps:

[0159] Step 4211: Select an individual from the current species for mutation, randomly select two other individuals from the current species to calculate the weighted vector difference, and combine the calculated weighted vector difference with the individual for mutation to obtain the corresponding mutated individual. Repeat the mutation operation multiple times to obtain multiple mutated individuals.

[0160] Step 4212: Perform crossover operation on each mutated individual to obtain the corresponding crossover individual;

[0161] Step 4213: For individuals in the current species and their corresponding crossover individuals, calculate the fitness function value, select the individual with the smaller fitness function value as the updated individual, and obtain the updated species.

[0162] In this embodiment of the invention, based on the above-described mutation operation, the mutated individual can be represented as:

[0163] ;

[0164] in, Indicates the first A mutated individual, Indicates the individual used for mutation. and Indicates the selection of two other individuals. Indicates the scaling factor. .

[0165] Among them, the scaling factor Configure according to actual needs.

[0166] In this embodiment of the invention, the crossover operation on the mutated individuals is performed using the following formula:

[0167] ;

[0168] in, Indicates the first The first cross-individual dimensional elements, Indicates the first The mutated individual dimensional elements, Indicates the first The individual dimensional elements, The dimension index of the vector corresponding to the mutated individual. This represents a randomly generated dimension index. A random number in [0,1] This represents the crossover rate.

[0169] Among them, cross rate Configure according to actual needs.

[0170] It should be noted that one element in an individual corresponds to the layout parameters of a component. The layout parameters include the geometric center coordinates of the component, the component-cabinet assignment variable, and the placement direction selection variable.

[0171] In this embodiment of the invention, the selection operation is performed using the following formula:

[0172] ;

[0173] in, Indicates the first The selected individual, i.e., the updated individual. Indicates the first A crossover individual, Indicates the first Individual, Represents the fitness function;

[0174] The fitness function is expressed as:

[0175] ;

[0176] in, Represents variables, Indicates the penalty coefficient. This represents the adjustment coefficient, when the variable... When the corresponding component layout scheme is feasible, When the variable When the corresponding component layout scheme is not feasible, , Indicates based on variables The objective function value obtained by solving the corresponding component layout scheme.

[0177] In this embodiment of the invention, the penalty coefficient The specific values ​​are set according to actual needs.

[0178] In this embodiment of the invention, by using the differential evolution algorithm described above to update each species independently, different species can evolve independently, avoiding mode loss caused by global competition, while maintaining multiple local optimal solutions.

[0179] Furthermore, in this embodiment of the invention, an information-guided local search strategy is used to perform local search operations on individuals in the population, and the individuals are updated based on the results of the local search operations, including the following steps:

[0180] Step 4221: For the best individual in the population, generate at least two new individuals based on Gaussian distribution, compare the best individual and its corresponding at least two new individuals according to feasibility rules, and select the best individual as the best individual in the population.

[0181] Step 4222: For each individual in the population other than the optimal individual, calculate the objective function search probability, constraint violation search probability, and distance search probability for each individual.

[0182] Step 4223: For each individual in the population other than the optimal individual, generate a random number for each individual. Determine whether the random number for each individual is less than the objective function search probability, constraint violation search probability, and distance search probability for that individual. If so, generate a new individual based on a Gaussian distribution. Compare the current individual and its corresponding new individual according to the feasibility rules and replace the current individual with the better individual. If not, retain the individual.

[0183] In this embodiment of the invention, since the objective function is minimized, individuals with lower objective function values ​​or constraint violation values ​​are more likely to approach the optimal value. Therefore, compared to other individuals, these superior individuals should be assigned a higher probability to perform local search operations. To this end, in this embodiment of the invention, local search operations are always performed on the best individual in the population, while for other individuals in the population besides the best individual, the decision to perform a local search operation is based on three local search probabilities to improve the efficiency and convergence speed of the algorithm.

[0184] In this embodiment of the invention, the objective function search probability is calculated using the following formula:

[0185] ;

[0186] in, Indicates the first The objective function search probability for each individual. Indicates the first The objective function value corresponding to each individual. This represents the maximum value of the objective function for all individuals. This represents the minimum value of the objective function among all individuals. It is a positive number.

[0187] in, It is a small positive number to avoid the case where the denominator is 0 in the formula. The value should be set according to the actual situation, for example, set to 0.0001.

[0188] The probability of constraint violation is calculated using the following formula:

[0189] ;

[0190] in, Indicates the first The constraint violation search probability corresponds to each individual. Indicates the first The constraint violation value corresponding to each individual, This represents the maximum value among all constraint violation values ​​for all individuals. This represents the minimum constraint violation value among all individuals.

[0191] Furthermore, when only two types of information, objective function value or constraint violation value, are used to guide the search direction, the algorithm may perform repeated and meaningless searches on a large number of similar (i.e., closer) excellent individuals. To address this, distance information is introduced in this embodiment of the invention. Higher search probabilities are assigned to individuals located in different promising regions, while lower search probabilities are assigned to individuals located in similar regions, in order to ensure the diversity of the population and enhance the optimization ability.

[0192] In this embodiment of the invention, the distance search probability is calculated using the following formula:

[0193] ;

[0194] in, Indicates the first The distance search probability corresponding to each individual Indicates the first The distance between each individual and the optimal individual This represents the distance between the optimal individual and its nearest neighbor. This represents the distance between the best individual and the individual furthest from it.

[0195] In this embodiment of the invention, the new individual generated based on the Gaussian distribution is represented as follows:

[0196] ;

[0197] in, This represents a new individual generated based on a Gaussian distribution. Represents the original individual. Indicates a Gaussian distribution. Indicates standard deviation, This indicates that the mean is 0 and the standard deviation is 0. A Gaussian distributed random variable.

[0198] In this embodiment of the invention, the standard deviation is adaptively adjusted according to the design space range, specifically in the following ways:

[0199] ;

[0200] in, and These represent the upper and lower bounds of the design variable, respectively. This represents the scaling factor, which can be set according to actual needs, for example, 0.01~0.1.

[0201] If the design space is multidimensional, a Gaussian distribution operation is performed on each dimension.

[0202] In this embodiment of the invention, the above-mentioned information-guided local search strategy introduces objective function information, constraint violation information, and distance information, and sets three local search probabilities to guide and balance the search direction together. This ensures that the search region has potential (guided by objective function information), feasibility (guided by constraint violation information), and diversity (guided by distance information). It can enhance the ability to explore, discover, and locate multiple optimal solutions, accelerate the convergence speed, and improve the solution accuracy.

[0203] Furthermore, in this embodiment of the invention, an information-guided reinforcement evolution mechanism is used to perform reinforcement evolution operations on inferior individuals in the population, and the inferior individuals are updated based on the results of the reinforcement evolution operations, including the following steps:

[0204] Step 4231: Sort all individuals in the population according to the feasibility rule and select several inferior individuals;

[0205] Step 4232: For each selected inferior individual, determine the species key point closest to it and the two neighboring individuals that are closest to it and better. Based on the determined species key point and neighboring individuals, perform mutation evolution of the individual to obtain a mutated individual. Perform cross-evolution on the mutated individual to obtain a cross-evolution individual. Use the cross-evolution individual as the next generation individual to continue the mutation evolution and cross-evolution of the individual until the set number of individual iterations is completed.

[0206] Step 4233: According to the feasibility rule, compare the updated individual with its corresponding original inferior individual. If the updated individual is better than the original inferior individual, replace the original inferior individual with the updated individual. If the updated individual is worse than the original inferior individual, retain the original inferior individual.

[0207] In this embodiment of the invention, the number of inferior individuals selected is specifically set according to actual needs, for example, set to 10% of the population size.

[0208] In this embodiment of the invention, the species key point is defined as the optimal individual in each divided species.

[0209] In this embodiment of the invention, the number of individual iterations is set according to actual needs.

[0210] In this embodiment of the invention, the mutational evolution of an individual is represented as follows:

[0211] ;

[0212] in, Indicates the first A mutated, evolving individual. Indicates the first A less desirable individual, and This represents the two neighboring individuals corresponding to the inferior individual. This indicates the species key point corresponding to the inferior individual. This represents the scaling factor.

[0213] Among them, the scaling factor Configure according to actual needs.

[0214] In this embodiment of the invention, cross-evolution of mutated individuals is performed using the following formula:

[0215] ;

[0216] in, Indicates the first The first cross-evolutionary individual dimensional elements, Indicates the first The mutated evolutionary individual dimensional elements, Indicates the first The inferior individual dimensional elements, The dimension index of the vector corresponding to the mutated evolutionary individual. A random number in [0,1] Indicates the crossover rate. Let represent a random integer uniformly distributed in the interval [1, D], where D represents the total dimension of the vector corresponding to the individual.

[0217] Among them, cross rate Configure according to actual needs.

[0218] In this embodiment of the invention, the above-mentioned information-guided reinforcement evolution mechanism ranks individuals in the population by using objective function values ​​and constraint violation values. Based on the ranking information and distance information, it performs multiple reinforcement evolutions on inferior individuals without evaluation, which can accelerate the evolution of individuals to feasible regions. It can enhance constraint handling capabilities by overcoming the severe evolutionary stagnation problem in a small number of discrete feasible regions in compact layout problems.

[0219] refer to Figure 4 Secondly, embodiments of the present invention also provide a satellite component layout optimization system, which is used for component layout optimization when the layout area is composed of multiple panels, including:

[0220] The separation axis pre-calculation module 100 is used to calculate the separation axis set between the compartments based on the edge vectors of each compartment. The separation axis set includes the axis directions defined by the edge vectors of the compartments and their cross product vectors.

[0221] The safety distance processing module 200 is used to calculate the dimensions and geometric center coordinates of each component after expansion based on the preset safety distance of each component.

[0222] The constraint construction module 300 is used to construct a geometric constraint model based on the set of separation axes and the extended components. This model includes geometric non-interference constraints, component installation position constraints, and component attitude uniqueness constraints. Geometric non-interference constraints include non-interference constraints between components and non-interference constraints between components and the cabin. Component installation position constraints are used to limit the components to be completely located within the boundaries of their respective cabins. Component attitude uniqueness constraints are used to limit the components to be installed on only one cabin and to have only one placement orientation. Geometric non-interference constraints are constructed in the following way: the corresponding inequality constraints are constructed using the geometric centers of two components, or the geometric centers of the components and the cabin on at least one separation axis, as constraints. The inequality constraints are linearized using the method of large numbers and coupled with the component-cabinet allocation variables through separation axis selection variables so that the components satisfy the corresponding inequality constraints when they are installed.

[0223] The optimization and solution module 400 is used to construct and solve a mixed-integer linear programming model with geometric constraint model as constraint conditions, so as to obtain a component layout scheme that satisfies the preset objective function.

[0224] In this embodiment of the invention, each of the above modules is a device corresponding to the above method steps. The specific working principle and beneficial effects of each module can be found in the above method, and will not be repeated here.

[0225] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Additionally, the terms "front," "back," "left," "right," "upper," and "lower" in this document refer to the placement shown in the accompanying drawings.

[0226] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the layout of satellite components, characterized in that, The method is used for component layout optimization when the layout area is composed of multiple panels, and comprises the following steps: Based on the edge vectors of the panels, a set of separation axes between the panels is calculated, the set of separation axes comprising axis directions defined by the edge vectors of the panels and their cross product vectors; According to the preset safety distance of each component, the size and geometric center coordinates of each extended component are calculated; Based on the set of separation axes and the extended components, a geometric constraint model is constructed, which comprises geometric non-interference constraints, component mounting position constraints and component attitude uniqueness constraints, the geometric non-interference constraints comprising non-interference constraints between components and non-interference constraints between components and panels, the component mounting position constraints being used to limit the components to be completely located within the boundary of the panel to which the components belong, and the component attitude uniqueness constraints being used to limit the components to be mounted on only one panel and to have only one placement direction, the geometric non-interference constraints being constructed by taking the projection distance of the geometric centers of two components or the geometric centers of the components and the panels on at least one separation axis to be not less than half of the sum of the projection lengths of the two on the current separation axis as a constraint condition, linearizing the inequality constraint by using the large number method, and coupling the separation axis selection variable and the component-panel distribution variable to make the components meet the corresponding inequality constraint when being mounted; The geometric constraint model is taken as a constraint condition to construct and solve a mixed integer linear programming model, so as to obtain a component layout scheme meeting a preset objective function.

2. The satellite assembly layout optimization method of claim 1, wherein, The size and geometric center coordinates of the extended components are calculated by the following method: The length of the extended component is the initial length of the component plus the safety distance of the component in the length direction; The width of the extended component is the initial width of the component plus the safety distance of the component in the width direction; The height of the extended component is the initial height of the component plus the safety distance of the component in the height direction; The geometric center coordinates of the extended component are determined by superimposing the offset calculated from the safety distance difference on the initial geometric center coordinates of the component.

3. The satellite assembly layout optimization method of claim 1, wherein, The non-interference constraints between the components are constructed by the following method: ； in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the j-th component after expansion. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the j-th component on the k-th separation axis after expansion. This represents a preset positive integer. and Choose a variable for the first separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between components. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between components. and Assign variables to the component-cabinet. , , This indicates that the i-th component is installed in the i-th position. On the deck. This indicates that the j-th component is installed on the h-th compartment panel. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

4. The satellite assembly layout optimization method of claim 1, wherein, The non-interference constraints between the components and the panels are constructed by the following method: ； in, Let k be the unit vector representing the k-th separation axis. This represents the geometric center coordinates of the nth deck. This represents the geometric center coordinates of the i-th component after expansion. This represents the maximum projected length of the i-th component on the k-th separation axis after expansion. This represents the maximum projected length of the nth compartment on the kth separation axis. This represents a preset positive integer. and Choose a variable for the second separation axis. , , This indicates that the positive direction of the k-th separation axis is selected to satisfy the non-interference constraint between the component and the cabin plate. This indicates that the opposite direction of the k-th separation axis is chosen to satisfy the non-interference constraint between the component and the cabin plate. Assign variables to the component-cabinet. , This indicates that the i-th component is installed in the i-th position. On the deck. Indicates the number of separation shafts. Indicates the number of decks. Indicates the number of components.

5. The satellite assembly layout optimization method of claim 1, wherein, The component mounting position constraints are constructed by the following method: The size difference of the extended component and the panel to which the component belongs in the length direction and the width direction of the panel, and the size sum in the height direction of the panel, are respectively not less than the projection distance of the geometric centers of the two in the corresponding direction.

6. The satellite assembly layout optimization method of claim 5, wherein, The component mounting position constraints are constructed by the following method: ； wherein, , , represents three edge vectors corresponding to the nth panel, represents the geometric center coordinate of the nth panel, represents the geometric center coordinate of the ith extended component, represents the length of the nth panel, represents the width of the nth panel, represents the height of the nth panel, represents the dimension of the ith extended component in the length direction of the panel when the ith extended component is installed on the panel, represents the dimension of the ith extended component in the width direction of the panel when the ith extended component is installed on the panel, represents the dimension of the ith extended component in the height direction of the panel when the ith extended component is installed on the panel, represents a preset normal number, assigns variables to the component-panel, , represents that the ith component is installed on the nth panel.

7. The satellite assembly layout optimization method of claim 1, wherein, The objective function is a multi-objective composite function, and the multi-objective composite function balances at least two optimization objectives by a weighting method, the two optimization objectives being: A centroid optimization objective for minimizing the position deviation of the overall centroid of the components from a preset target centroid; A heat dissipation performance optimization objective for preferentially allocating components with higher power consumption to panels with lower heat accumulation coefficients.

8. The satellite assembly layout optimization method of claim 1, wherein, The method further comprises the following steps: The set of separation axes is filtered to remove duplicate separation axes.

9. The satellite assembly layout optimization method of claim 1, wherein, The mixed integer linear programming model is solved by using an integer programming solving algorithm, or the mixed integer linear programming model is solved by the following method: For the mixed integer linear programming model, an initial population containing multiple individuals is randomly generated, the individuals representing component layout schemes, one individual corresponding to one component layout scheme; For each generation population, the population is divided into multiple species by using the improved niche algorithm, the species are updated by using the differential evolution algorithm, based on the updated population, the individuals in the population are subjected to local search operation by using the information-guided local search strategy, and the individuals are updated according to the local search operation result, based on the population updated again, the inferior individuals in the population are subjected to the strengthening evolution operation by using the information-guided strengthening evolution mechanism, and the inferior individuals are updated according to the strengthening evolution operation result, the updated population is taken as the next generation population to continue the population iteration update until the population iteration update is completed for a set number of times; According to the updated population, a plurality of component layout schemes are obtained.

10. A satellite assembly layout optimization system, characterized by, The system is used for component layout optimization when the layout area is composed of multiple deck plates, and comprises: A separation axis pre-computation module is configured to compute a separation axis set between the deck plates based on edge vectors of the deck plates, the separation axis set including axis directions defined by the edge vectors of the deck plates and cross product vectors thereof; A safety distance processing module is configured to compute sizes and geometric center coordinates of the expanded components according to preset safety distances of the components; A constraint construction module is configured to construct a geometric constraint model including geometric non-interference constraints, component installation position constraints and component attitude uniqueness constraints based on the separation axis set and the expanded components, the geometric non-interference constraints including non-interference constraints between the components and non-interference constraints between the components and the deck plates, the component installation position constraints being configured to limit the components to be completely located within boundaries of the deck plates, and the component attitude uniqueness constraints being configured to limit the components to be installed on only one deck plate and to have only one placement direction, the geometric non-interference constraints being constructed by taking a projection distance of geometric centers of two components or the components and the deck plates on at least one separation axis to be not less than half of a sum of projection lengths of the two components or the components and the deck plates on the current separation axis as a constraint condition to construct corresponding inequality constraints, linearizing the inequality constraints by using a large number method, and coupling separation axis selection variables and component-deck plate assignment variables to make the components meet the corresponding inequality constraints when being installed; An optimization solving module is configured to construct and solve a mixed integer linear programming model with the geometric constraint model as a constraint condition to obtain a component layout scheme meeting a preset objective function.

Images