Optimization method of regional coverage satellite constellation based on dynamic fusion of genetic ant colony
Through dynamic fusion of genetic ant colony algorithm, satellite constellation parameters are optimized, and the problem of insufficient coverage and revisit accuracy in the design of specific regions is solved, and efficient coverage and revisiting of the target area is achieved.
Patent Information
- Application Number
- CN202110843279.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-07-26
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-07-26
AI Technical Summary
It is difficult for the prior art to effectively design satellite constellations with specific regions to meet the requirements of target area coverage and revisit accuracy, and traditional methods have problems with local iteration and accuracy influence.
The regional coverage satellite constellation optimization method based on dynamic fusion of genetic ant colony is adopted. By obtaining regional importance weights and sorting, dynamically fusion of genetic algorithms and ant colony algorithms, constellation parameters are optimized to improve coverage and revisit accuracy.
The accuracy and accuracy of constellation parameters are achieved, local iteration is avoided, regional target coverage and revisit requirements are met, and the accuracy of coverage time and revisit times is improved.
Smart Images

Figure CN113673148B_ABST
Abstract
Description
Technical Field
[0001] The invention discloses a method for optimizing a regional coverage satellite constellation based on dynamic fusion of genetic ant colonies, and belongs to the technical field of calculation, estimation or counting. Background Art
[0002] With the development of aerospace technology and demand, the mission requirements of satellite constellations have changed from global or latitudinal coverage to key regional coverage. Because the global population and resources are unevenly distributed and local natural disasters continue to occur frequently, the cost of building and maintaining satellite constellations with global or latitudinal coverage is relatively high, which makes the construction of regional satellite constellations a focus of aerospace missions. The geographical location of a specific area, the irregular boundary shape, and the complex space-time relationship caused by the rotation of the earth make the traditional typical constellation design method no longer applicable to regional coverage constellations. Regional coverage missions are mostly used to provide intensive communications and surveillance for key areas, so the constellation design for specific regional coverage needs to focus on the coverage characteristics of the target area.
[0003] The main method for designing a constellation covering a specific area is to design a common subsatellite trajectory constellation. The satellites that make up the common subsatellite trajectory constellation have overlapping subsatellite trajectories, their orbital altitudes and orbital inclinations are equal, but their ascending node right ascensions and phases are different. The design process of a satellite constellation is a typical multi-objective, multi-variable, and nonlinear problem. The paper "Design of a Constellation for Continuous Coverage of Irregular Areas Based on Particle Swarm" (Liu Ming, Yang Ming, Gao Xing, Yang Ding, Ge Yajie. Aerospace Control, 2020, 38(02): 31-37.) adopts a common subsatellite trajectory constellation design scheme based on regression orbits, and proposes a method of solving orbital parameters using a particle swarm algorithm and solving constellation parameters using an analytical method; "Optimal Design Method for Low-Earth Orbit Satellite Constellations for Intensive Observation in a Specific Area" (Ma Jian, Meng Yazhe, Zhu Xiaolong, He Shengmao, Gao Yang. Science China: Technological Sciences, 2018, 48(02) :170-184.) With the minimum number of satellites and revisit coverage time as the goal, the paper first established a "circular orbit-target point" database, and searched the regression orbits of all target points and the common subsatellite point trajectory constellation with a specific step size. Finally, the genetic algorithm was used to optimize the orbit parameter results of the global search to obtain the final constellation parameters. However, this would affect the accuracy of the final calculation results due to the accuracy of the search step size; "Optimization of low-orbit hybrid constellation design using improved non-dominated neighbor immune algorithm" (Jiang Xinglong, Jiang Quanjiang, Liu Huijie, Yu Jinpei. Acta Astronautics, 2014, 35(09):1007-1014.) uses an improved non-dominated nearest neighbor immune algorithm to optimize the constellation design problem with multiple constraints; “Optimization Design of Constellation Configuration for Irregular Area Imaging Coverage” (Yu Yanjun, Wang Feng, Miao Yue. Journal of Space Science, 2019, 39(04):494-501.) uses an improved simulated annealing algorithm combined with an improved equal-area grid point coverage method, and designs an imaging constellation for irregular areas with the goal of minimizing the number of satellites and meeting the revisit time. The paper "A method for designing satellite constellation configuration with quadruple coverage of the earth" (Meng Shaofei, Liu Xinxue, Yang Qi, Zhu Tao. Journal of Astronautics, 2017, 38(11): 1160-1167) combines genetic algorithm and ant colony algorithm to overcome the shortcomings of the two algorithms and proves the performance of the fusion algorithm through simulation calculation. However, the fusion algorithm is connected by setting a fixed iteration value, which has poor scalability and may cause the multi-constraint and multi-objective satellite constellation design problem to enter local iteration.
[0004] Therefore, the main difficulty in designing a constellation for covering a specific area is how to use intelligent algorithms to improve the coverage and revisit accuracy of the constellation to the target area to meet its mission requirements. The present invention aims to provide a regional coverage satellite constellation optimization method based on dynamic fusion of genetic ant colonies to solve the above problem. Summary of the invention
[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned background technology, provide a regional coverage satellite constellation optimization method based on dynamic fusion of genetic ant colonies, establish a regional key weight arrangement model to determine the priority of regional coverage, and adopt a dynamic judgment method to determine the fusion time and algebra of genetic algorithms and ant colony algorithms, so as to ensure the precision and accuracy of the constellation parameter optimization results, solve the technical problem that the existing constellation design scheme for specific regional coverage cannot meet the parameter accuracy requirements, and achieve the invention purpose of using a dynamic fusion intelligent algorithm to solve the constellation parameters that meet the regional target coverage and revisit requirements.
[0006] The present invention adopts the following technical solutions to achieve the above-mentioned invention object:
[0007] An optimization method for regional coverage satellite constellation based on dynamic fusion of genetic ant colony algorithm is proposed. The importance weight of the area to be covered and the ranking of the weights of each location in the area to be covered are obtained. Each satellite visits each location in turn according to the ranking of the weights of each location in the area to be covered. The goal is to maximize the coverage area of each satellite along the orbit during the overpass time. The satellite constellation is planned by dynamically fusion of genetic algorithm and ant colony algorithm.
[0008] Furthermore, the regional coverage satellite constellation optimization method based on the dynamic fusion of genetic ant colony algorithm is aimed at maximizing the coverage area of each satellite along the orbit during the overhead time. The specific method of planning the satellite constellation by the dynamic fusion method of genetic algorithm and ant colony algorithm is as follows:
[0009] The six Kepler orbit numbers of each satellite and the longitude and declination of the subsatellite track are used as chromosomes for genetic operation. When the difference between the maximum fitness value and the average fitness value of the population of n consecutive generations is less than the difference between the maximum fitness value and the average fitness value of the population of the n-1 generation, the genetic operation is stopped.
[0010] According to the six Kepler orbit numbers in the global optimal solution obtained by the nth genetic operation, pheromone is constructed, and the pheromone is updated until the maximum number of iterations is reached. The pheromone update values obtained by the maximum number of iterations are sorted, and the six Kepler orbit numbers represented by the maximum pheromone value are selected as the optimal solution.
[0011] Furthermore, based on the regional coverage satellite constellation optimization method of genetic ant colony dynamic fusion, the method for obtaining the importance weight of the area to be covered is: weighted integration of the weights of each location in the area to be covered to obtain the importance weight of the area to be covered.
[0012] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the specific method of genetic operation is to use the six elements of the Kepler orbit of each satellite and the longitude and declination of the sub-satellite trajectory as chromosomes: calculate the fitness value of each chromosome and then sort them, copy the chromosomes ranked within the preset range to the next generation population for further iterative calculation, and perform adaptive crossover operations on the unselected chromosomes.
[0013] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the expression for calculating the fitness value of each chromosome is: Among them, fitness_c is the fitness value of the cth satellite covering K locations, σ k is the importance weight of the area where the kth location is located, are the boundary functions of the left and right sides of the k-th location covered by the c-th satellite sub-satellite trajectory, is the operating mileage of the c-th satellite covering the area of the k-th location, t 0,k is the time when the cth satellite starts to cover the area of the kth location, t m,k is the time when the cth satellite completely moves out of the area where the kth location belongs, i c is the orbital inclination of the cth satellite, ω c is the coverage angle of the cth satellite, are the right ascension longitude and declination of the left boundary of the kth location covered by the subsatellite track of the cth satellite, are the right ascension longitude and declination of the right boundary of the kth location covered by the subsatellite track of the cth satellite, α m,c ,δ m,c are the right ascension and declination of the sub-satellite point trajectory of the cth satellite respectively.
[0014] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the expression for constructing pheromone according to the six elements of Kepler orbit in the global optimal solution obtained by the nth genetic operation is: in, is the pheromone released by the cth ant crawling from node a to node b in the tth iteration, and the starting value of t is n+1.
[0015] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the expression of updating pheromone is: Among them, τ ab (t), τ ab (t+1) are the pheromones released by the ant colony when crawling from node a to node b in the tth and t+1th iterations, respectively. abis the change in pheromone concentration caused by the ant colony crawling from node a to node b, ρ is the volatility of pheromone, and N is the number of global optimal solutions passed into the ant algorithm.
[0016] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the expression of adaptive crossover operation on unselected chromosomes is: Among them, P z,b is the adaptive crossover probability of the bth chromosome, Δfitness n is the difference between the maximum fitness value in the nth generation genetic operation and the fitness value of the bth chromosome, is the maximum fitness value in the nth generation genetic operation, is the fitness value of the bth chromosome, k 2 is a constant c.
[0017] Furthermore, based on the regional coverage satellite constellation optimization method of dynamic fusion of genetic ant colonies, the right ascension α of the sub-satellite point trajectory of the cth satellite m,c , declination δ m,c According to the expression Calculate,Ω c ,Ω 0,c They are respectively the right ascension of the ascending node of the sub-satellite point trajectory of the cth satellite and the right ascension of the ascending node at the initial moment.
[0018] The present invention adopts the above technical scheme and has the following beneficial effects: combining the genetic algorithm and the ant colony algorithm, avoiding the problem of local convergence and slow initial information accumulation of a single optimization algorithm, ensuring that the satellite constellation meets the requirements of coverage time, regional coverage rate and revisit times, and at the same time, the key weights are prioritized according to the importance of the locations in the coverage area, and the constellation focuses on coverage and revisiting the set important targets in a targeted manner. The coverage time and revisit times within the full mission cycle and the unit cycle can meet the weight sorting of the locations, with prominent emphasis and clear levels. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 Schematic diagram of key weight locations.
[0020] Figure 2 Schematic diagram of the circular sensor coverage area of a satellite in orbit
[0021] Figure 3 Schematic diagram of the overall design plan for the constellation.
[0022] Figure 4 This is the schematic diagram of the common subsatellite trajectory constellation.
[0023] Figure 5 This is a model diagram of the satellite coverage area.
[0024] Figure 6 This is the flowchart of the genetic ant colony algorithm.
[0025] Figure 7 Schematic diagram of the area designed for the simulation experiment.
[0026] Figure 8 This is a three-dimensional distribution diagram of the constellations of common subsatellite point tracks.
[0027] Fig. 9 Schematic diagram of the satellite constellation's coverage of the target area in two-dimensional space.
[0028] Fig.10 This is a schematic diagram of the statistical results of the full cycle coverage time of key targets.
[0029] Fig.11 This is a diagram showing the statistical results of the revisit frequency of key target units. DETAILED DESCRIPTION
[0030] The technical solution of the invention is described in detail below with reference to the accompanying drawings.
[0031] (I) Ranking and coverage of key weighted areas
[0032] Traditional regional coverage design only considers full coverage and intensive revisits of satellite constellations within a specific mission cycle. The drawback of such design is that it cannot focus on one or more important locations, and non-important value locations may consume too much revisit time and density. Constellation optimization design for regional coverage using intelligent algorithms focuses more on reducing the size of the constellation configuration and the number of satellites to reduce costs, or increasing the overall coverage time or revisit times of the region. However, in actual missions, the importance of locations within a region is often unbalanced, and some locations within the region are relatively more important.
[0033] Therefore, for the area to be covered, a focus weight σ is set, and the locations and regions are sorted according to the focus weight σ when planning the constellation. The more important the location, the greater the focus weight σ, and the more intensive the satellite revisit will be when the constellation is fully covered. Figure 1 The key areas shown are marked as: Place 1, Place 2, Place 3, Place 4, Place 5, Place 6, Place 7.
[0034] Among them, the key weights of Place 1, Place 2, Place 3, Place 4, Place 5, Place 6, and Place 7 are σ respectively. c , where σ c As c increases, the key weights of other unspecified locations are all σ a , and σ ais a constant. The region is a location within a certain range. The weighted integral method can be used to solve the key weights of a small area. Regional focus weight By sorting, the priority of satellite constellation coverage and revisit can be determined.
[0035] The size of the coverage area of a circular sensor on an orbiting satellite is affected by the orbital altitude h and the sensor half angle γ, as Figure 2 As shown in the figure, the projection of the satellite circular sensor on the earth's surface can be approximately regarded as a two-dimensional circle. Although the accuracy of the projection area is not high, it can be used as a rough estimate and verification of the sub-satellite point track coverage.
[0036] The half angle γ and orbital position h, coverage radius r and two-dimensional coverage area S of the circular sensor loaded on the satellite are known. cover As shown in formula (1):
[0037] S cover =π·r 2 =π(tanγ·h) 2 (1)
[0038] In formula (1), γ is the coverage angle of the satellite orbit.
[0039] In view of this method, Figure 3 The overall constellation design plan shown is as follows: first, analyze the distribution of target areas, and determine the weights of key areas based on the mission coverage and revisit requirements; second, limit the satellite orbit to a near-Earth circular recurrent orbit, reduce the initial population of satellite orbits, and increase the calculation rate; then use the common subsatellite point trajectory constellation configuration to design the orbit model; then use intelligent algorithms to calculate the optimization processing, and obtain the results after optimization and local adjustments; finally, after experimental verification, the final design results can be output.
[0040] (II) Constellation Model of Common Subsatellite Point Trajectory of Regression Orbit
[0041] The regression orbit refers to the orbit in which the subsatellite point trajectory of the satellite is repeated periodically. The angular velocity of the earth's rotation is ω E , the satellite angular velocity is ω, the orbital period is T, and its orbit satisfies:
[0042] NT(ω E -ω)=πD (2),
[0043] In formula (2), NT is the regression period and D is the diameter of the Earth.
[0044] The common subsatellite point trajectory constellation is a constellation configuration in which the satellite orbit and the subsatellite point trajectory completely overlap. Its principle diagram is as follows Figure 4 As shown, here T represents a return period, the satellite passes through m locations in sequence and then returns again.
[0045] The common subsatellite point trajectory constellation configuration design using regression orbits first requires modeling the satellite constellation operation, and solving the satellite's position and velocity in the J2000 coordinate system using analytical methods. Then, the satellite subsatellite point trajectory and coverage area are solved using geometric methods. Among them, the non-spherical perturbation of the earth's gravity will have a greater impact on the low-Earth orbit, including J2, J3, and J4 perturbations. The satellite coverage area models from two perspectives are as follows: Figure 5 shown.
[0046] J 2 The orbit recursion model under gravitational perturbation is formula (3):
[0047]
[0048] In formula (3), h x 、h y 、h z They represent the three components of the satellite position vector h in the J2000 coordinate system, and the modulus length is: J 2 is the second-order band co-term coefficient of the earth, r E is the radius of the Earth, are the three components of the second differential of the position vector, and μ is the orbital constant.
[0049] The position and velocity of the satellite in the J2000 coordinate system at any time are calculated using equation (3), and the obtained position and velocity are converted into Kepler's six elements. The declination and right ascension of the subsatellite point trajectory can be obtained from spherical trigonometric functions:
[0050]
[0051] In formula (4), δ, α, i, Ω, Ω 0 They are the declination, right ascension, orbital inclination, right ascension of the ascending node and the right ascension of the ascending node at the initial moment of the sub-satellite point trajectory.
[0052] (III) Regional coverage constellation design based on dynamic fusion of genetic ant colonies
[0053] Genetic algorithms have fast global search capabilities, but they do not use the feedback information in the system, resulting in a large number of useless iterative operations and low efficiency in accurate solutions. Ant colony algorithms have the ability to distribute, parallelize, and converge globally, but the lack of pheromone accumulation in the early stages leads to a long time for pheromone accumulation in the initial search and a slow solution speed.
[0054] The integration of genetic algorithm and ant colony algorithm can overcome the defects of the two algorithms and form complementary advantages of the two algorithms. Figure 6As shown in the figure, the algorithm first performs the calculation iteration of the genetic algorithm part. After satisfying the dynamic judgment conditions, it enters the iterative optimization of the ant colony algorithm part. The genetic algorithm part adopts the sorting coding method, and encodes the six Kepler numbers and the corresponding right ascension and longitude of the subsatellite point trajectory, that is, δ, α, i, Ω, e, a, ω, ξ as a group of chromosomes. Because the constellation satellite is designed to be in a near-earth circular orbit, e = 0, and e is the eccentricity of the orbital ellipse.
[0055] After completing the chromosome encoding, a preliminary population is generated as the starting population using a random heuristic factor s, and the population size ranges from 100 to 300.
[0056] The key to the genetic algorithm is the determination of the fitness function. When the specific coverage area and the coverage mission execution time are determined, the basic conditions for coverage are: the orbital area of each satellite in the coverage area is the area of the specific area, and the coverage time reaches the mission execution time. The coverage rate is calculated using the grid point method, and the time when satellite c starts to cover the specific area k is recorded as t 0,k , the time to completely run out of a specific area k is recorded as t m,k The fitness function is a function of the area covered by the satellite along the orbit during the time of passing the top. In general genetic algorithms, the fitness function seeks the minimum or minimum value. Therefore, the fitness function fitness_c calculated based on the cth group of chromosomes for the cth satellite covering K locations is:
[0057]
[0058]
[0059]
[0060] In formula (5) to formula (7), σ k is the importance weight of the area where the kth location is located, are the boundary functions of the left and right sides of the k-th location covered by the c-th satellite sub-satellite trajectory, is the operating mileage of the c-th satellite covering the area of the k-th location, i c is the orbital inclination of the cth satellite, ω c is the coverage angle of the cth satellite; are the right ascension longitude and declination latitude of the left boundary of the kth location covered by the sub-satellite track of the cth satellite; are the right ascension longitude and declination latitude of the right boundary of the kth location covered by the subsatellite track of the cth satellite; α m,c ,δ m,c are the right ascension and declination of the subsatellite point trajectory of the cth satellite, respectively, calculated according to formula (4).
[0061] After calculating the fitness function, the population is sorted according to the size of the fitness function, and the top 40% of chromosomes are selected and copied to the next generation population for further iterative calculation.
[0062] Perform adaptive crossover operation on chromosomes that have not been copied to the next generation population. First determine the execution generation of the current population. If it is less than G min , performing a fixed crossover probability p g =0.88; otherwise, the adaptive crossover probability is executed, and the adaptive crossover probability P of the bth chromosome is z,b As shown in formula (8):
[0063]
[0064] In formula (8), Δfitness n is the maximum fitness value in the nth generation genetic operation and the fitness value of the bth chromosome The difference, k 2 is a constant c, and c ranges from 0.01 to 0.09.
[0065] The genetic ant colony algorithm establishes a dynamic judgment control function, which controls the iteration of the genetic algorithm by calculating the evolution rate of the offspring to ensure the dynamic cross-fusion of the genetic algorithm and the ant colony algorithm.
[0066] The steps of dynamic determination are as follows:
[0067] 1) In genetic operations, in order to determine the iteration range, the minimum number of iterations G is set min and the maximum number of iterations G max .
[0068] 2) If within the range of the number of iterations, the following conditions are satisfied for n consecutive generations:
[0069]
[0070] Δfitness n <Δfitness n-1 (10)
[0071] In formula (9) and formula (10), are the maximum fitness value and average fitness value of the nth generation population, Δfitness n-1 It is the difference between the maximum fitness value and the average fitness value in the n-1th generation genetic operation.
[0072] If the above dynamic judgment conditions are met, the genetic operation is stopped, and the global optimal solution of the six satellite orbit numbers is passed into the ant colony algorithm as pheromone and enters the ant colony algorithm part. The pheromone that originally needed to be accumulated slowly has been solved by the genetic part.
[0073] The parameter settings of the ant colony algorithm are as follows:
[0074] 1) Pheromone function setting:
[0075]
[0076] In formula (11), is the pheromone released by the cth ant crawling from node a to node b in the tth iteration.
[0077] 2) Pheromone update model:
[0078]
[0079] In formula (10), τ ab (t), τ ab (t+1) are the pheromones released by the ant colony when crawling from node a to node b in the tth and t+1th iterations, respectively. ab is the change in pheromone concentration caused by the ant colony crawling from node a to node b; ρ is the volatility of pheromone, and N is the number of global optimal solutions passed into the ant algorithm.
[0080] 3) The pheromone is updated iteratively through the pheromone update model. The starting value of t is the number of genetic operations before the start of the ant colony algorithm plus 1. After a limited number of pheromone updates, the maximum number of iterations G is reached. max When the maximum number of iterations G max Obtained pheromone update value Sort, that is, for the G max In the -1 iteration, the pheromone concentrations of each ant are sorted from a to b, and the six satellite orbit numbers represented by the maximum pheromone concentration are selected as the optimal solution. When the maximum number of iterations is not reached, the path list recording the six satellite orbit numbers is cleared and the pheromone is continued to be updated.
[0081] (IV) Simulation Experiment and Coverage Analysis
[0082] Simulation experiment design: To achieve intensive observation of the area from 3° to 19° north latitude and 108° to 120° east longitude (divided into two longitude and latitude units), the observation time is from 00:00 on March 20, 2025 (UTCG) to 00:00 on March 27, 2025 (UTCG). It is required to achieve full coverage of the entire target area.
[0083] Target location Latitude and longitude Key weight Place1 16.02;113.34 5.23 Place2 18.67;109.56 4.17 Place3 4.02;108.99 9.53 Place4 4.01;116.03 9.09 Place5 9.98;112.99 8.52 Place6 10.03;119.03 8.04 Place6 5.81;117.33 8.86
[0084] Table 1. Design of weight ranking of important locations
[0085] As shown in Table 1, the weight ranking of important locations achieves intensive revisit of less than 1 hour for the target area, and focuses on revisiting and tilting mission resources for key locations. The half angle of the circular sensors carried by all satellites in the constellation is set to 45°, and the satellites operate in a low orbit of 350 to 1500 km.
[0086] Based on the initial orbital parameters, the six Kepler numbers of all satellites are calculated using the common subsatellite point trajectory constellation solution algorithm described above. The orbital radius of all satellites is 6932.46km, the eccentricity is 0, and the perigee angle is 0. Based on this, the corresponding constellation is established. Figure 7 is the three-dimensional model of the 48 satellites of the constellation in the target area. Figure 7 It can be seen that its sub-satellite point trajectory can completely cover the specific target area, and the coverage of the locations with the highest importance weight is more dense, which meets the requirements of the designed experiment.
[0087] Based on the construction of the common subsatellite point trajectory constellation, through STK simulation data, Figure 8 The maximum revisit time of all satellites in the constellation for a specific target during the entire mission execution period is analyzed. The maximum revisit time is less than 3600s, and the revisit time of some satellites is less than 2800s, which fully meets the design requirement of a revisit time of less than 1 hour. The schematic diagram of the target area covered by the satellite constellation in two-dimensional space is shown in the figure below. Fig. 9 shown.
[0088] Next, for the key weight σ set above i The full cycle revisit time and revisit frequency density analysis of the locations are as follows Fig.10 and Fig.11 As shown in the figure, Place 3 and Place 4, which have the highest corresponding key weights, have the longest full-cycle coverage time, and are significantly ahead of Place 1 and Place 2 by about 40%, and ahead of Place 5, Place 6, and Place 7 by about 13%, which well reflects that the set key weight ranking has a good effect on the focused revisit of key locations.
[0089] Then count the daily revisit frequency within the task execution cycle, such as Fig.11 As shown in Figure 2, Place 3 and Place 4, which have the highest corresponding focus weights, have maintained a relatively high revisit frequency, surpassing other locations.
[0090] It can be concluded that the constellation design guided by the focus weights can fully meet the needs of access emphasis in general.
Claims
1. A method for optimizing regional coverage satellite constellations based on dynamic fusion of genetic ant colonies, characterized in that: Obtain the importance weight of the area to be covered and the ranking of the weights of each location in the area to be covered. Each satellite visits each location in turn according to the ranking of the weights of each location in the area to be covered. The goal is to maximize the coverage area of each satellite along the orbit within the time of passing the top. The satellite constellation is planned by using a dynamic fusion method of genetic algorithm and ant colony algorithm. Among them, With the goal of maximizing the coverage area of each satellite along the orbit during the overhead time, the specific method of planning the satellite constellation by dynamically integrating the genetic algorithm and the ant colony algorithm is as follows: The six Kepler orbit numbers of each satellite and the longitude and declination of the subsatellite track are used as chromosomes for genetic operations. When the difference between the maximum fitness value and the average fitness value of the n-th generation population is less than the difference between the maximum fitness value and the average fitness value of the n-1 generation population, the genetic operation is stopped. The expression for calculating the fitness value of each chromosome is: Among them, fitness_c is the fitness value of the cth satellite covering K locations, σ k is the importance weight of the area where the kth location is located, are the boundary functions of the left and right sides of the k-th location covered by the c-th satellite sub-satellite trajectory, is the operating mileage of the c-th satellite covering the area of the k-th location, t 0,k is the time when the cth satellite starts to cover the area of the kth location, t m,k is the time when the cth satellite completely moves out of the area where the kth location belongs, i c is the orbital inclination of the cth satellite, ω c is the coverage angle of the cth satellite, are the right ascension longitude and declination of the left boundary of the kth location covered by the subsatellite track of the cth satellite, are the right ascension longitude and declination of the right boundary of the kth location covered by the subsatellite track of the cth satellite, α m,c , δ m,c are the right ascension and declination of the subsatellite point trajectory of the cth satellite, According to the six Kepler orbit numbers in the global optimal solution obtained by the nth genetic operation, pheromone is constructed, and the pheromone is updated until the maximum number of iterations is reached. The pheromone update values obtained by the maximum number of iterations are sorted, and the six Kepler orbit numbers represented by the maximum pheromone value are selected as the optimal solution. The expression for constructing pheromone according to the six Kepler orbit numbers in the global optimal solution obtained by the nth genetic operation is: The expression for updating pheromone is: in, is the pheromone released by the cth ant crawling from node a to node b in the tth iteration. The starting value of t is n+1, τ ab (t), τ ab (t+1) are the pheromones released by the ant colony when crawling from node a to node b in the tth and t+1th iterations, respectively. ab is the change in pheromone concentration caused by the ant colony crawling from node a to node b, ρ is the volatility of pheromone, and N is the number of global optimal solutions passed into the ant algorithm.
2. According to claim 1, the method for optimizing regional coverage satellite constellation based on dynamic fusion of genetic ant colonies is characterized in that: The method for obtaining the importance weight of the area to be covered is: performing weighted integration on the weights of various points in the area to be covered to obtain the importance weight of the area to be covered.
3. According to claim 1, the method for optimizing regional coverage satellite constellation based on dynamic fusion of genetic ant colonies is characterized in that: The specific method of performing genetic operations using the six Kepler orbit numbers of each satellite and the right ascension and longitude of the subsatellite track as chromosomes is as follows: calculating the fitness value of each chromosome and sorting it, copying the chromosomes ranked within a preset range to the next generation population for continued iterative calculation, and performing adaptive crossover operations on the unselected chromosomes.
4. According to claim 3, the method for optimizing regional coverage satellite constellation based on dynamic fusion of genetic ant colonies is characterized in that: The expression for performing adaptive crossover operation on unselected chromosomes is: Among them, P z,b is the adaptive crossover probability of the bth chromosome, Δfitness n is the difference between the maximum fitness value in the nth generation genetic operation and the fitness value of the bth chromosome, is the maximum fitness value in the nth generation genetic operation, is the fitness value of the bth chromosome, and k2 is the constant c.
5. According to claim 1, the method for optimizing regional coverage satellite constellation based on dynamic fusion of genetic ant colonies is characterized in that: The right ascension α of the sub-satellite trajectory of the cth satellite m,c , declination δ m,c According to the expression Calculate,Ω c ,Ω 0,c They are respectively the right ascension of the ascending node of the sub-satellite point trajectory of the cth satellite and the right ascension of the ascending node at the initial moment.
