A Satellite Fast Maneuvering Control Method Based on Parallel Systems

Abstract

The present invention discloses a satellite rapid maneuver control method based on a parallel system, constructs a satellite rapid maneuver parallel system, combines the on-orbit spacecraft telemetry data obtained in real time, uses a particle swarm algorithm to correct the thrust parameters of the artificial model, thereby establishing an artificial model that is kept consistent with the actual system in real time, and on this basis, through the artificial model simulation calculation test, optimally estimates the jet time of the maneuver control, obtains the jet time with the minimum error between the post-maneuver and the target position under actual conditions, and feeds it back to the actual system. The present invention can achieve long-term consistency between the artificial system and the actual system, predict and control the actual system, has high precision and robustness, and is applicable to satellite rapid maneuver control.

Description

Technical Field

[0001] The present invention relates to the field of spacecraft simulation and control, and particularly to a satellite rapid maneuvering control method based on a parallel system. Background Art

[0002] With the rapid growth of the number of spacecraft, the space orbital resources are becoming increasingly tense, and spacecraft need to frequently perform tasks such as orbital maneuvering and orbit change avoidance. High-precision spacecraft control is an important guarantee for realizing these maneuvering tasks. Therefore, it is particularly important to build a ground flight control and simulation system to simulate the operating state of the spacecraft and verify the rationality and accuracy of the control scheme design. Whether the control strategy is formulated appropriately depends on whether the real-time state of the on-orbit spacecraft can be accurately obtained. However, some states in the actual system are often unable to be directly obtained through observation, and due to the influence of uncertain factors, some parameters of the system are also in the process of dynamic change. Therefore, it is very difficult for ordinary offline simulation systems to accurately cope with the above problems.

[0003] Regarding the problem of rapid maneuvering of near-circular orbit satellites, the satellite thrust parameters are the key influencing factors because the thrust magnitude will directly affect the satellite maneuvering result. Secondly, during the process of the spacecraft performing maneuvering tasks on orbit, the jet thruster operates under a falling pressure. As the fuel consumption of the thruster and the gas pressure change, the actual output thrust of the thruster will also change accordingly. During the entire life cycle of the spacecraft, the change in the thrust magnitude is significant. Therefore, if the initial set thrust parameters are always used to formulate the control strategy, the deviation of the spacecraft orbit maneuvering result will become larger and larger. The existing thrust parameter estimation method is to model the thruster according to the thrust generation principle, estimate the actual thrust by combining the empirical formula obtained from the ground test of the thruster with the relevant on-orbit telemetry data, and calculate the second-order approximate estimation of the thrust using the pressure value measured by the pressure sensor in the thruster tank. The problem is that the relationship between the thrust and the maneuvering result is not established, and the unknown factors in the space environment will affect the accuracy of the empirical formula.

[0004] For the satellite rapid maneuvering strategy, most of the existing satellite maneuvering control strategies focus on the calculation of velocity increment, and there is no research on the relationship between the specific jet time and the satellite maneuvering error. For example, the literature (Wang Wei, Yuan Jianping, Luo Jianjun. Optimal single-pulse maneuver of spacecraft cluster formation [J]. Chinese Journal of Theoretical and Applied Mechanics, 2015, 47(05): 799-806.) believes that the satellite can instantaneously obtain the velocity increment, but in fact, the satellite thruster is a finite-thrust one, and it takes a certain amount of time to achieve acceleration; the literature (Yang Shengqing, Jia Yansheng, Cui Jia, Du Yaoke, Wang Wenyan, Wu Jingyu. A new method for calculating the jet duration of satellite gas cylinder thrusters [J]. Aerospace Shanghai, 2018, 35(04): 48-53. DOI: 10.19328 / j.cnki.1006-1630.2018.04.008.) models the thruster to optimize the jet time to calculate the accurate jet time required for different velocity increments. However, precisely because the satellite maneuvering is not instantaneous, during the maneuvering process, as the satellite position changes, the required velocity increment gradually changes. Secondly, the dynamic models used for calculating the velocity increment generally do not consider high-order term space disturbances, etc. These two problems will ultimately lead to the satellite being unable to accurately reach the target position. Summary of the Invention

[0005] Object of the Invention: The object of the present invention is to provide a satellite rapid maneuvering control method based on a parallel system that can reduce the terminal guidance pressure, achieve satellite-ground collaborative control, and improve the maneuvering control accuracy and efficiency at the same time.

[0006] Technical Solution: The satellite rapid maneuvering control method of the present invention includes the following steps:

[0007] S1, select the J2000 inertial coordinate system as the basic coordinate system, and construct an artificial simulation model for the satellite rapid maneuvering system containing perturbation factors;

[0008] S2, on the basis of the artificial simulation model, by receiving the maneuvering strategy information and orbit position telemetry data of the actual satellite maneuvering system, use the particle swarm algorithm to correct the artificial simulation model to obtain the thrust value with the minimum error from the orbit position telemetry data as the optimal thrust estimate value;

[0009] S3, by setting a "virtual satellite" at the satellite maneuvering target position, convert the satellite maneuvering problem into a rendezvous problem between the satellite and the "virtual satellite", use the CW equation to describe the relative motion equation of the virtual satellite and the satellite, and solve the velocity increment required for satellite maneuvering;

[0010] S4, construct the relationship between velocity increment and jet time. Use the particle swarm algorithm to use the jet time as a reference to define the particle position range. Particles randomly generate jet times within the range. Substitute the randomly generated jet times into the modified artificial simulation model to conduct a rapid maneuver calculation experiment. The difference between the simulation result and the maneuvering target position is used as the objective function. Iterate and obtain the jet time with the smallest error with the maneuvering target position, which is the optimal estimate of the jet time.

[0011] Furthermore, in step S1, the expression of the artificial simulation model is as follows:

[0012]

[0013]

[0014] Where (x, y, z) is the coordinate point of the satellite in the J2000 coordinate system, μ is the Earth's gravitational constant, R is the satellite's radius vector, and R e is the earth's equatorial radius, J2 is the second-order harmonic coefficient, F x 、F y 、F z are the components of satellite thrust in the x-axis, y-axis, and z-axis directions, m is the current mass of the satellite, m0 is the mass of the satellite at the initial moment of jet ejection, is the total fuel consumption rate, t is the current time of injection, and t0 is the initial time of injection;

[0015] The components of the satellite thrust along the x-axis, y-axis, and z-axis in the J2000 coordinate system are:

[0016]

[0017] M vo is the transformation matrix from the body coordinate system to the orbital coordinate system:

[0018]

[0019] Among them, θ, and ψ are the pitch angle, roll angle and yaw angle of the satellite attitude angle respectively. If the satellite is oriented to the ground, then M vo is the identity matrix;

[0020] M Jv is the transformation matrix from the orbital coordinate system to the J2000 coordinate system:

[0021]

[0022] Where u, Ω and β are the argument of latitude, right ascension of ascending node and orbital inclination respectively;

[0023] f x 、fy and f z are the thrusts in the x-axis, y-axis, and z-axis directions under the satellite's own system, and their expressions are as follows:

[0024]

[0025] N is the number of thruster installations, F j is the output thrust of a single thruster, and α i is the installation angle of the corresponding thruster;

[0026] The calculation expression of the thrust of a single thruster is as follows:

[0027]

[0028] F is the thruster thrust, Isp is the specific impulse of the satellite fuel, g is the acceleration due to gravity, is the fuel consumption rate of the thruster.

[0029] Furthermore, in step S2, taking the thrust value as the particle position and the difference between the simulation result and the orbital position telemetry data as the objective function, the expression of the objective function J(F) is as follows:

[0030]

[0031] where X i , Y i , Z i are the actual system telemetry position data, x i , y i , z i are the simulation system calculated position data for the corresponding orbital epoch, and Num is the number of position data;

[0032] Taking the objective function J(F) as the population fitness, after iteration, the population position with the minimum population fitness is obtained. At this time, the error between the simulation result and the actual position result is the smallest, and the corresponding thrust value is the optimal thrust estimate value;

[0033] The particle update formula is as follows:

[0034]

[0035]

[0036]

[0037] where v i is the particle velocity, is the particle's next-generation velocity, rand() is a random number between (0, 1), F i is the current position of the particle, i.e., the thrust value, is the next position of the particle, pbest i is the historical optimal thrust value of the particle, gbest i is the optimal thrust value of the population, c1 and c2 are learning factors, ω (k) is the inertia factor, ω ini is the initial inertia weight, ω end is the inertia weight when iterating to the maximum number of generations, G is the maximum number of iterations, and k is the current number of iterations.

[0038] Furthermore, in step S3, the implementation steps for solving the velocity increment required for satellite maneuver are as follows:

[0039] S31, Set a "virtual satellite" at the satellite maneuver target position, and describe the relative motion equation between the virtual satellite and the satellite using the CW equation:

[0040]

[0041] where, δx′, δy′, δz′ are the components of the relative position vector between the virtual satellite and the satellite on the x-axis, y-axis, and z-axis in the moving coordinate system of the virtual satellite, and n is the angular velocity of the virtual satellite;

[0042] Solve the above equation to obtain the relative position vector δr(t) and relative velocity vector δv(t) at time t as:

[0043]

[0044] where the CW matrix is:

[0045]

[0046]

[0047]

[0048]

[0049] S32, In the moving coordinate system of the virtual satellite at the initial moment, calculate the relative position vector and relative velocity vector between the virtual satellite and the satellite as follows:

[0050]

[0051] where, r, r 虚拟 are the position vectors of the satellite and the virtual satellite in the inertial system respectively, v, v 虚拟 are the velocity vectors of the satellite and the virtual satellite in the inertial system respectively, δr is the relative position vector between the two satellites in the inertial system, Ω 虚拟 is the angular velocity vector of the virtual satellite, M xJis the conversion matrix from the J2000 coordinate system to the virtual star motion coordinate system;

[0052] Then the satellite maneuver velocity increment is:

[0053]

[0054] At the rendezvous time t f the initial relative velocity vector is:

[0055]

[0056] S33, convert the satellite maneuver velocity increment to coordinates:

[0057]

[0058] where M Jx is the conversion matrix from the virtual star motion coordinate system to the J2000 coordinate system, M vJ is the conversion matrix from the J2000 coordinate system to the satellite orbit coordinate system, M ov is the conversion matrix from the satellite orbit coordinate system to the satellite body coordinate system.

[0059] Furthermore, in step S4, construct the relationship between the velocity increment Δv x and the jet time t x The relationship between the velocity increment on the x-axis and the thrust is:

[0060]

[0061] It can be obtained that:

[0062]

[0063] The relationship between the jet time and the velocity increment is solved as:

[0064]

[0065] Taking the jet time as the particle position, using the jet time solved by the CW equation as a reference, delimit the particle position range, and randomly generate specific maneuver strategies within a reasonable range through the particle swarm algorithm. Then the jet times of the satellite on the x-axis, y-axis, and z-axis are:

[0066]

[0067] Substitute the jet times of the satellite on the x-axis, y-axis, and z-axis into the satellite maneuver artificial model for calculation experiments, and take the difference between the simulation result and the maneuver target position as the objective function J(T):

[0068]

[0069] Among them, X p , Y p , Z p are the positions of the maneuvering target, and x p , y p , z p are the orbit recurrence simulation data at the corresponding epoch moments; the objective function J(T) is used as the population fitness. After iteration, the population position with the minimum population fitness is obtained, which is used as the maneuvering jet strategy corresponding to the minimum error between the simulation result and the maneuvering target position result.

[0070] Compared with the prior art, the remarkable effects of the present invention are as follows:

[0071] 1. By determining the state parameters that cannot be directly observed by the satellite, the present invention establishes an artificial model that is always consistent with the actual system in real time, thereby providing support for formulating the satellite maneuvering strategy, reducing the terminal guidance pressure, and realizing satellite-ground collaborative control; when the actual satellite performs the orbit maneuvering task, the present invention can realize high-efficiency and high-precision satellite on-orbit thrust estimation;

[0072] 2. According to the requirements of the maneuvering task, the present invention estimates the jet time required to minimize the error between the position after maneuvering and the target position, improves the maneuvering control accuracy and efficiency, and constructs a complete framework of the satellite maneuvering parallel system. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 is a schematic diagram of the satellite maneuvering parallel system of the present invention

[0074] Figure 2 is a schematic diagram of the satellite maneuvering artificial model of the present invention;

[0075] Figure 3 is a flow chart of the thrust estimation of the particle swarm algorithm of the present invention;

[0076] FIG. 4(a) is a diagram of the thrust estimation result of the particle swarm algorithm of the present invention,

[0077] FIG. 4(b) is a diagram of the change of the population fitness (i.e., the objective function) with the number of iterations during the thrust estimation process;

[0078] Figure 5 is a flow chart of the jet strategy formulation in the present invention;

[0079] FIG. 6(a) is a diagram of the x-axis jet time estimation result,

[0080] FIG. 6(b) is a diagram of the y-axis jet time estimation result,

[0081] FIG. 6(c) is a diagram of the change of the population fitness (i.e., the objective function) with the number of iterations during the jet time estimation process. DETAILED DESCRIPTION OF THE INVENTION

[0082] The present invention will be further described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0083] To solve the control problem under the state of missing spacecraft parameter information and inaccurate physical models, the parallel system method is proposed. This method integrates artificial systems, computational experiments, and parallel execution, uses quantitative and repeatable computational experiments to harden the "virtual" and "soft" parts, combines actual system information to correct the artificial system, so that it has the same internal operation rules as the actual system, and can obtain the same output under the same input. Furthermore, during the parallel execution process, through the mutual interaction and real-time feedback between the actual system and the artificial system, the control and management of the actual system, the experiment and evaluation of decision-making schemes, and the evolution and correction of the artificial system are completed. As Figure 1 shown, in this embodiment, by constructing a satellite maneuver parallel system, the telemetry data of the on-orbit spacecraft is obtained in real time, and the artificial model is corrected by using the particle swarm optimization algorithm (PSO) to make its thrust value consistent with the actual satellite, realizing the dynamic evolution of the artificial model and establishing an artificial model that is always consistent with the actual system in real time.

[0084] Based on the correction of the artificial model, the optimal estimation of the jet time for maneuver control is carried out. Taking the specific jet time as the optimization parameter, the jet time with the minimum error between the actual situation after maneuver and the target position is obtained. Based on this, a support strategy is given and fed back to the actual system.

[0085] The satellite rapid maneuver control method of the present invention includes the following steps:

[0086] Step 1, constructing an artificial simulation model for the satellite rapid maneuver system;

[0087] Select the J2000 inertial coordinate system (i.e., the geocentric inertial coordinate system) as the basic coordinate system, and establish a satellite orbit maneuver dynamics model with perturbation factors in this coordinate system as the artificial simulation model. The expression is as follows:

[0088]

[0089]

[0090] where (x, y, z) are the coordinate points of the satellite in the J2000 coordinate system, μ = 3.986004418×10 14 (m 3 / s 2 ) is the gravitational constant of the earth, R is the magnitude of the satellite radius vector, R e = 6378136.3 is the equatorial radius of the earth, J2 = 1.082629989052×10 -3 is the second-order zonal harmonic coefficient, F x 、F y, F z are the components of the satellite thrust in the x-axis, y-axis, and z-axis directions respectively, m is the current mass of the satellite, and m0 is the mass of the satellite at the initial moment of jetting. is the total fuel consumption rate, t is the current moment of jetting, and t0 is the initial moment of jetting.

[0091] The non-powered recursive model of the spacecraft adopts the high-precision HPOP model.

[0092] The three-axis components of the satellite thrust in the J2000 coordinate system are:

[0093]

[0094] M vo is the transformation matrix from the body coordinate to the orbital coordinate system, as shown below:

[0095]

[0096] where θ, and ψ are the pitch angle, roll angle, and yaw angle in the satellite attitude angles respectively. If the satellite is earth-oriented, this transformation matrix is the identity matrix;

[0097] M Jv is the transformation matrix from the orbital coordinate system to the J2000 coordinate system, as follows:

[0098]

[0099] where u, Ω, and β are the argument of latitude, right ascension of the ascending node, and orbital inclination respectively;

[0100] f[[ID=..]] x 、f y and f z are the thrusts in the three-axis directions in the satellite body coordinate system. The specific calculation methods for the thrusts in the three-axis directions are all:

[0101]

[0102] N is the number of thruster installations, F j is the output thrust of a single thruster, and α j is the installation angle of the corresponding thruster;

[0103] The thrust calculation formula for a single thruster is as follows:

[0104]

[0105] F is the thruster thrust, Isp is the specific impulse of the satellite fuel, g = 9.80665 (m / s 2 ) is the acceleration due to gravity, Let \(\dot{m}\) be the fuel consumption rate of the thruster. During a single maneuver, the fuel consumption rate is approximately constant. Therefore, the average satellite mass change rate during the maneuver can be used as the fuel consumption rate.

[0106] Step 2: Based on the artificial simulation model constructed in Step 1, by receiving the maneuver strategy information and orbital position telemetry data of the actual satellite maneuvering system, using the Particle Swarm Optimization (PSO) algorithm, with the thrust value as the particle position, and taking the difference between the simulation result and the orbital position telemetry data as the objective function, obtain the thrust value with the smallest error from the orbital position telemetry data to achieve the correction of the artificial model.

[0107] Satellite thrust value estimation is carried out based on the artificial simulation model. The specific process of thrust estimation is as follows: Take the thrust value as the particle position. The particle randomly generates thrust parameters within a certain range. Input the thrust parameters into the artificial simulation model and perform orbital recursive simulation calculations on the maneuvering process according to the actual control strategy to obtain a series of simulation position data corresponding to the actual telemetry position data at the orbital epoch moments. Take the difference between the actual telemetry position data and the simulation position data at the corresponding orbital epoch moments as the objective function \(J(F)\), and the expression is as follows:

[0108]

[0109] where \(X\) i 、\(Y\) i 、\(Z\) i are the actual system telemetry position data, \(x\) i 、\(y\) i 、\(z\) i are the simulation system calculation position data corresponding to the orbital epoch, and Num is the number of position data. Taking this objective function as the population fitness, after iteration, obtain the population position with the smallest population fitness, that is, when the error between the simulation result and the actual position result is the smallest, the corresponding thrust value is the optimal thrust estimation value.

[0110] The particle update formula is as follows:

[0111]

[0112]

[0113]

[0114] where \(v\) i is the particle velocity, is the particle's next-generation velocity, rand() is a random number between (0, 1), \(F\) i is the particle's current position, that is, the thrust value, is the next position of the particle, pbest i is the historical optimal thrust value of the particle, gbest i is the optimal thrust value of the population, c1 and c2 are learning factors, ω (k) is the inertia factor, ω ini is the initial inertia weight, ω end is the inertia weight at the maximum number of iterations. G is the maximum number of iterations, and k is the current number of iterations. The dynamic inertia factor enables the algorithm to have better optimization performance, ω (k) When it is larger, it has better global optimization ability, ω (k) When it is smaller, the local optimization ability is stronger.

[0115] Step 3: For the requirements of fast maneuvering tasks, use the method of setting a virtual satellite, that is, by setting a "virtual satellite" at the satellite maneuver target position, convert the satellite maneuver problem into a rendezvous problem between the satellite and the "virtual satellite", and use the CW equation (Clohessy-Wiltshire equation) to describe the relative motion equation between the virtual satellite and the satellite to solve the velocity increment required for satellite maneuvering.

[0116] Set a "virtual satellite" at the satellite maneuver target position and use the CW equation to describe the relative motion equation between the virtual satellite and the satellite:

[0117]

[0118] where δx′, δy′, and δz′ are the components of the relative position vector of the two on the x-axis, y-axis, and z-axis in the moving coordinate system of the virtual satellite, and n is the angular velocity of the virtual satellite.

[0119] Continue to solve Equation (12), and the relative position vector δr(t) and relative velocity vector δv(t) at time t can be obtained as:

[0120]

[0121] where the CW matrix is:

[0122]

[0123]

[0124]

[0125]

[0126] The calculation methods of the relative position vector and relative velocity vector of the two satellites in the moving coordinate system of the virtual satellite at the initial moment are as follows:

[0127]

[0128] where r and r 虚拟 are the position vectors of the satellite and the virtual satellite in the inertial frame, v and v 虚拟 are the velocity vectors of the satellite and the virtual satellite in the inertial frame, δr is the relative position vector between the two satellites in the inertial frame, Ω 虚拟 is the angular velocity vector of the virtual satellite, and M xJ is the transformation matrix from the J2000 coordinate system to the moving coordinate system of the virtual satellite. Therefore, as long as the initial relative position vector and the rendezvous time with the virtual satellite are given, the initial relative velocity vector required for the rendezvous orbit can be obtained, and the satellite maneuver velocity increment is:

[0129]

[0130] The initial relative velocity vector calculated by Equation (13) at the rendezvous time t f is:

[0131]

[0132] When actually formulating the satellite maneuver control strategy, it is necessary to know the velocity increment in the satellite's body frame. Therefore, it is also necessary to perform coordinate transformation on the satellite maneuver velocity increment:

[0133]

[0134] where M Jx is the transformation matrix from the moving coordinate system of the virtual satellite to the J2000 coordinate system, M vJ is the transformation matrix from the J2000 coordinate system to the satellite orbit coordinate system, and M ov is the transformation matrix from the satellite orbit coordinate system to the satellite's body frame.

[0135] Step 4: Construct the relationship between the velocity increment and the jet time, convert the velocity increment obtained in Step 3 into the jet time, use the particle swarm algorithm to take the calculated jet time as a reference, delimit the particle position range, and randomly generate the jet time within the range for the particles, that is, randomly generate a specific maneuver strategy within a reasonable range through the particle swarm algorithm, substitute it into the corrected artificial simulation model in Step 2 for rapid maneuver calculation experiments, take the difference between the simulation result and the jet time of the maneuver target position as the objective function, and perform iteration to obtain the jet time with the smallest error from the maneuver target position, which is the optimal estimated value of the jet time.

[0136] The control strategy of the actual satellite is the jet time. Construct the relationship between the velocity increment Δv x and the jet time t x . The relationship between the velocity increment on the x-axis and the thrust is:

[0137]

[0138] It can be obtained from formula (19) that:

[0139]

[0140] The relationship between the jet time and the velocity increment is solved as:

[0141]

[0142] The jet time required for maneuvering can be calculated through the above formula (21).

[0143] Taking the jet time as the particle position, and using the jet time solved by the CW equation as a reference, the particle position range is delimited. The jet time is randomly generated within the range, that is, a specific maneuvering strategy is randomly generated within a reasonable range through the particle swarm algorithm. That is, the jet times of the satellite on the x-axis, y-axis, and z-axis are:

[0144]

[0145] Substitute formula (22) into the artificial model of satellite maneuvering for calculation experiments, and take the difference between the simulation result and the maneuvering target position as the objective function:

[0146]

[0147] Among them, X p , Y p , Z p are the maneuvering target positions, x p , y p , z p are the orbit recurrence simulation data at the corresponding epoch moments. The objective function J(T) is used as the population fitness. After iteration, the population position with the minimum population fitness is obtained, that is, the maneuvering jet strategy corresponding to the minimum error between the simulation result and the maneuvering target position result.

[0148] For the satellite rapid maneuvering control method of this embodiment, the following application background is set: Assume a geostationary orbit satellite (total mass m 质量 = 1500 kg), the satellite always maintains earth orientation, and performs orbit maneuvering under the action of a constant thrust F = 25 N in the -z axis direction of the satellite orbit coordinate system from the initial orbit (orbit epoch T is 2022-1-1-000000, A = 42166.3 km, e = 0, i = 0.12°, Ω = 90.45°, ω = 10.75°, θ = 204.0°). The jet time is set to t = 300 s, and the fuel consumption rate The telemetry data noise is 100 m. The second maneuver is performed after twelve hours. The target orbit position is T = 2022-1-1-120500, A = 42166.3 km, e = 0, i = 0.12°, Ω = 90.09°, ω = 250.92°, θ = 146.42°. The implementation steps of the satellite rapid maneuver control method are as follows:

[0149] Step A1, as Figure 2 shown, construct an artificial simulation model for the satellite rapid maneuver.

[0150] Step A2, as Figure 3 shown, based on the constructed artificial simulation model of satellite rapid maneuver, use the orbit telemetry data, and adopt the particle swarm algorithm to estimate the thrust. Set the learning factors c1 = 2, c2 = 2, the initial inertia weight ω ini = 0.5, the final inertia weight ω end = 1, the population size popsize = 200, the number of iterations N = 100, and the particle position value range is [10, 30].

[0151] As Figure 4(a) , 4(b) shown, obtain the thrust estimation result. The time used for calculation is 78.364 s, the optimal thrust estimation value is 24.9689 N, and the error from the actual thrust value is 0.1244%. At this thrust value, the error between the recursive orbit position of the simulation model and the telemetry data is 2.0564 m.

[0152] Step A3, set a virtual satellite in the target orbit, and calculate the relative position vector and relative velocity vector of the two satellites in the virtual satellite coordinate system at the initial moment:

[0153]

[0154]

[0155] Calculate the required velocity increment in the satellite orbit coordinate system as:

[0156]

[0157] Step A4, based on the velocity increment obtained in Step A3, calculate the jet time as tx x = 394.1066 s in the x-axis direction and ty y = 23.9984 s in the y-axis direction. As Figure 5 shown, based on the modification of the artificial model, perform an optimal estimation of the jet time for the maneuver control. Set the particle range with the above values as a reference. The jet time range in the x-axis direction is [350, 450], and the jet time range in the y-axis direction is [0, 100]. Use the particle swarm algorithm for estimation.

[0158] like Figure 6(a) 、 6(b) 6(c), the final estimated result is the jet time t in the x-axis direction x =407.1501s, y-axis jet time t y =0.2897s. Twelve hours after the maneuver, the error with the target position is 9.4318m.

[0159] Examples demonstrate that this invention, through the interaction between artificial simulation models and actual systems, can efficiently and accurately achieve thrust estimation, with an estimation accuracy better than 0.1%, ensuring consistency between the artificial and actual systems. The maneuvering strategy developed by this invention enables a satellite to achieve an accuracy of 9 meters from its target position over a movement distance of several hundred kilometers, significantly reducing the error in a single maneuver.

[0160] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A satellite rapid maneuvering control method based on a parallel system, characterized in that, It includes the following steps: S1. Select the J2000 inertial coordinate system as the basic coordinate system, and construct an artificial simulation model for the satellite rapid maneuvering system with perturbation factors; S2. On the basis of the artificial simulation model, by receiving the maneuvering strategy information and orbit position telemetry data of the actual satellite maneuvering system, and using the particle swarm algorithm, correct the artificial simulation model to obtain the thrust value with the minimum error from the orbit position telemetry data as the optimal thrust estimate value; S3. By setting a "virtual satellite" at the satellite maneuvering target position, convert the satellite maneuvering problem into a rendezvous problem between the satellite and the "virtual satellite", use the CW equation to describe the relative motion equation between the virtual satellite and the satellite, and solve the velocity increment required for satellite maneuvering; S4. Construct a relationship between the velocity increment and the jet time. Using the particle swarm algorithm with the jet time as a reference, delimit the particle position range. The particle randomly generates the jet time within the range, substitute the randomly generated jet time into the corrected artificial simulation model for rapid maneuvering calculation experiments, and use the difference between the simulation result and the maneuvering target position as the objective function for iteration to obtain the jet time with the minimum error from the maneuvering target position as the optimal jet time estimate value; In step S1, the expression of the artificial simulation model is as follows: where (x, y, z) are the coordinates of the satellite in the J2000 coordinate system, μ is the Earth's gravitational constant, R is the magnitude of the satellite's radius vector, R e is the Earth's equatorial radius, J2 is the second-order zonal harmonic coefficient, F x 、F y 、F z are the components of the satellite's thrust in the x-axis, y-axis, and z-axis directions respectively, m is the current mass of the satellite, m0 is the mass of the satellite at the initial moment of jetting, is the total fuel consumption rate, t is the current moment of jetting, and t0 is the initial moment of jetting; The components of the satellite thrust on the x-axis, y-axis, and z-axis in the J2000 coordinate system are: M vo is the transformation matrix from the body coordinates to the orbital coordinate system: where θ, and ψ are the pitch angle, roll angle, and yaw angle in the satellite attitude angles respectively; if the satellite is earth-oriented, then M vo is the identity matrix; M Jv is the transformation matrix from the orbital coordinate system to the J2000 coordinate system: where u, Ω, and β are the argument of latitude, right ascension of the ascending node, and orbital inclination respectively; f x 、f y and f z are the thrusts in the x-axis, y-axis, and z-axis directions under the satellite's own system, and their expressions are all as follows: N is the number of thruster installations, F j is the output thrust of a single thruster, α i is the installation angle of the corresponding thruster; The calculation expression of the thrust of a single thruster is as follows: F is the thrust of the thruster, Isp is the specific impulse of the satellite fuel, and g is the acceleration due to gravity. is the fuel consumption rate of the thruster.

2. The satellite rapid maneuvering control method for a parallel system according to claim 1, characterized in that, In step S2, taking the thrust value as the particle position and the difference between the simulation result and the orbit position telemetry data as the objective function, the expression of the objective function J(F) is as follows: Among them, X i , Y i , Z i are the telemetry position data of the actual system, and x i , y i , z i are the calculated position data of the simulation system corresponding to the orbit epoch, and Num is the number of position data; Taking the objective function J(F) as the population fitness, after iteration, obtain the population position with the minimum population fitness. At this time, the error between the simulation result and the actual position result is the smallest, and the corresponding thrust value is the optimal thrust estimate value; The particle update formula is as follows: where v i is the particle velocity, is the particle's next-generation velocity, rand() is a random number between (0, 1), F i is the current position of the particle, i.e., the thrust value, is the particle's next-generation position, pbest i is the particle's historical optimal thrust value, gbest i is the population-optimal thrust value, c1 and c2 are learning factors, ω (k) is the inertia factor, ω ini is the initial inertia weight, ω end is the inertia weight at the maximum number of iterations, G is the maximum number of iterations, and k is the current iteration number.

3. The satellite rapid maneuver control method in the parallel system according to claim 1, wherein In step S3, the implementation steps for solving the velocity increment required for satellite maneuvering are as follows: S31. Set a "virtual satellite" at the satellite maneuvering target position, and use the CW equation to describe the relative motion equation between the virtual satellite and the satellite: where δx′, δy′, and δz′ are the components of the relative position vector between the virtual satellite and the satellite on the x-axis, y-axis, and z-axis in the moving coordinate system of the virtual satellite, and n is the angular velocity of the virtual satellite; Solve the above formula to obtain the relative position vector δr(t) and relative velocity vector δv(t) at time t as: where the CW matrix is: S32. In the moving coordinate system of the virtual satellite at the initial moment, calculate the relative position vector and relative velocity vector between the virtual satellite and the satellite as follows: where r and r 虚拟 are the position vectors of the satellite and the virtual satellite in the inertial frame, v and v 虚拟 are the velocity vectors of the satellite and the virtual satellite in the inertial frame, δr is the relative position vector of the two satellites in the inertial frame, Ω 虚拟 is the angular velocity vector of the virtual satellite, and M xJ is the transformation matrix from the J2000 coordinate system to the moving coordinate system of the virtual satellite; Then the satellite maneuvering velocity increment is: At the rendezvous time t f The initial relative velocity vector is: S33. Perform coordinate transformation on the satellite maneuvering velocity increment: Among them, M Jx is the transformation matrix from the virtual star motion coordinate system to the J2000 coordinate system, and M vJ is the transformation matrix from the J2000 coordinate system to the satellite orbit coordinate system, and M ov is the transformation matrix from the satellite orbit coordinate system to the satellite body coordinate system.

4. The satellite rapid maneuvering control method in the parallel system according to claim 3, wherein In step S4, construct the speed increment Δv x and the jet time t x The relationship between the speed increment on the x-axis and the thrust is: It can be obtained that: Solve to obtain the relationship between the jet time and the velocity increment as: Taking the jet time as the particle position, using the jet time solved by the CW equation as a reference, delimit the particle position range, and randomly generate specific maneuvering strategies within a reasonable range through the particle swarm algorithm. Then the jet times of the satellite on the x-axis, y-axis, and z-axis are: Substitute the jet times of the satellite on the x-axis, y-axis, and z-axis into the artificial satellite maneuver model for calculation experiments, and use the difference between the simulation result and the maneuver target position as the objective function J(T): Among them, X p , Y p , Z p are the positions of maneuvering targets, and x p , y p , z p are the orbit recurrence simulation data at the corresponding epoch moments; the objective function J(T) is used as the population fitness. After iteration, the population position with the minimum population fitness is obtained as the maneuvering jet strategy corresponding to the minimum error between the simulation result and the maneuvering target position result.

Images