A method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system

Abstract

The present invention discloses a method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system. The method belongs to the field of attitude control of gliding high-speed aircraft. The method is oriented to high-speed aircraft with a reaction control system. An incremental model with equivalent constraints is constructed based on its attitude second-order dynamic system. While ensuring the tracking performance constraints, it reduces the dependence on the accuracy of the aerodynamic model. On the basis of incremental modeling, an event-triggered reaction control system is designed. The event triggering conditions are related to the performance constraints and the control costs, so that compensation is performed at the appropriate time through the reaction control system. Finally, the estimation error, input cost and constraint conditions are simultaneously considered, and an adaptive optimization control of a single evaluation network is constructed based on the adaptive dynamic programming method, thereby achieving optimal control of the gliding attitude of the high-speed aircraft.

Description

Technical Field

[0001] The present invention belongs to the field of attitude control of gliding high-speed aircraft, and more specifically, relates to a method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system. Background Art

[0002] To address the issue of decreased efficiency of aerodynamic actuators caused by the thin atmosphere of high-altitude gliding flight, high-speed gliding aircraft equipped with reaction control systems can effectively enhance the aircraft's attitude control capabilities through boosters. However, ensuring high-precision tracking control is a complex and challenging task. First, weak aerodynamic conditions often lead to saturation of aerodynamic actuators, resulting in a short-term inability to respond to control commands and a loss of system stability. Second, the actual flight environment of high-speed aircraft is complex and variable, and unknown external disturbances can easily cause the aircraft to deviate from its intended attitude. Furthermore, strict control accuracy is required during gliding to ensure gliding conditions and maneuverability. Existing methods generally use preset performance control to ensure control performance constraints. However, the application of preset performance methods to high-speed aircraft can be vulnerable. First, due to the nonlinear mapping of the transformation function, uncertain perturbations are also nonlinearly amplified, further impacting the stability of the closed-loop system. Second, due to the forced nature of the preset performance method, the input cost increases sharply at the performance boundary. When aerodynamic control is insufficient, the reaction control system must be triggered to supplement it. Therefore, in order to efficiently realize a large range of maneuvers during the gliding process of the aircraft, it is necessary to develop a robust optimization control method for the preset performance of high-speed aircraft based on a reaction control system. Summary of the Invention

[0003] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides a preset performance optimization control method for a gliding high-speed aircraft based on a reaction control system for tracking angle of attack instructions, and adopts a reaction control system to compensate for the saturation phenomenon of the pneumatic actuator to achieve optimal control of the gliding attitude of the high-speed aircraft.

[0004] To achieve the above objectives, according to a first aspect of the present invention, a method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system is provided. The gliding high-speed aircraft is equipped with a reaction control system, comprising:

[0005] S1, constructing a second-order attitude tracking error system for the gliding high-speed aircraft;

[0006] S2, establishing a tracking error constraint, and dynamically transforming the tracking error, and determining a sliding surface and sliding surface dynamics of the transformed tracking error based on the transformed tracking error and the second-order attitude tracking error system;

[0007] S3, dynamically constructing a neural network controller based on the sliding surface and the sliding surface to control the attitude of the gliding high-speed aircraft;

[0008] Wherein, the neural network controller is are the activation function and weight of the evaluation neural network, which is used to observe the sliding surface. The relevant value function is used to obtain The approximate optimal solution, R>0, is the gain coefficient of the control increment of the reaction control system, is the sliding surface at the last trigger moment, and the trigger condition is the event error Greater than the threshold.

[0009] According to a second aspect of the present invention, there is provided an electronic device comprising: a computer-readable storage medium and a processor;

[0010] The computer-readable storage medium is used to store executable instructions;

[0011] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method according to the first aspect.

[0012] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method described in the first aspect.

[0013] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0014] The method provided by this invention constructs an incremental model with equivalent constraints based on the second-order dynamics of the aircraft attitude system. This model ensures tracking performance constraints while reducing reliance on the accuracy of the aerodynamic model. Based on this incremental modeling, an event-triggered reaction control system is designed. The event triggering conditions are related to performance constraints and control costs, allowing for timely compensation through the reaction control system. Finally, by simultaneously considering estimation errors, input costs, and constraints, an adaptive optimization control system with a single evaluation network is constructed based on an adaptive dynamic programming method, thereby achieving optimal control of the high-speed aircraft's gliding attitude. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Flowchart of a method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system provided in an embodiment of the present invention.

[0016] Figure 2A schematic diagram of the angle of attack tracking control effect curve provided by an embodiment of the present invention.

[0017] Figure 3 A schematic diagram of a pitch angular velocity control response effect curve provided by an embodiment of the present invention.

[0018] Figure 4 This is a schematic diagram of the pneumatic rudder angle control response effect curve provided by an embodiment of the present invention.

[0019] Figure 5 A schematic diagram of the event error and trigger threshold response effect curve provided by an embodiment of the present invention.

[0020] Figure 6 A schematic diagram of event triggering time and sampling period provided by an embodiment of the present invention.

[0021] Figure 7 A schematic diagram of the weight update and control response curve of the evaluation neural network provided by an embodiment of the present invention.

[0022] Figure 8 A schematic diagram of a control torque response curve provided by a reaction control system (RCS) provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0024] The embodiment of the present invention provides a method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system. Figure 1 Shown, including:

[0025] S1, construct the second-order attitude tracking error system of the gliding high-speed aircraft with a reaction control system, that is, the high-speed aircraft attitude control model.

[0026] Considering the longitudinal second-order attitude dynamics of a gliding high-speed aircraft, the model is constructed as follows

[0027]

[0028] Among them, the system tracking error e = α - α d α, α d are the system state angle of attack and the desired angle of attack command respectively. The system control input δ e is the rudder deflection angle, u=M ris the torque input of the reaction control system. V is the system velocity, ρ is the atmospheric density, is the dynamic pressure, S and c represent the wing reference area and the average aerodynamic chord respectively, I yy is the moment of inertia. are the second-order derivatives of the track inclination and angle of attack commands respectively. m1 、C m2 、 are aerodynamic parameters.

[0029] S2, establishing a tracking error constraint, and performing a dynamic transformation on the tracking error, and determining a sliding mode surface and sliding mode surface dynamics of the tracking error after the transformation based on the tracking error after the transformation and the second-order attitude tracking error system.

[0030] In S2, an incremental control model is designed and constructed according to the performance constraints of the aircraft (i.e., tracking error constraints).

[0031] The tracking error constraint preset by the system is

[0032]

[0033] Where 0<κ<1, the performance constraint function of tracking error P(t)=(P0-P ∞ )exp(-vt)+P ∞ , P0>P ∞ >0, exp(-vt) is an exponential function and v>0, the initial error e(0) of angle of attack tracking satisfies |e(0)|<P0.

[0034] In order to meet the tracking error constraint, we construct The error conversion mechanism is designed as:

[0035]

[0036] Where ln is the natural logarithm function.

[0037] Constructing sliding surface based on error transformation

[0038]

[0039] By performing incremental transformations using Taylor linear expansion between adjacent sampling moments, the sliding surface dynamics are obtained as:

[0040]

[0041] in, is the rate of change of the sliding surface at the last sampling time, function

[0042]

[0043] Δδe is the pneumatic control increment. Generally, the pneumatic control increment is designed to be The control gain parameter k>0. Δu is the control increment of the reaction system, which is designed by the subsequent adaptive dynamic programming controller. are the gain coefficients of the pneumatic control increment and the reaction system control increment respectively.

[0044] It is understandable that the tracking error constraint may also adopt other constraint equations, such as a fixed time constraint equation. The embodiment of the present invention does not impose a unique limitation on this, and the above tracking error constraint equation is only used as an example.

[0045] In addition, in addition to the above-mentioned incremental transformation using Taylor linear expansion between adjacent sampling moments, other incremental transformation methods may also be used, such as a small deviation linearization method. The embodiment of the present invention does not impose a unique limitation on this, and the above-mentioned incremental transformation method is only used as an example.

[0046] S3, constructing a neural network controller according to the sliding surface and the sliding surface dynamics to control the attitude of the gliding high-speed aircraft.

[0047] In S3, an adaptive dynamic programming controller based on a single-criteria network is proposed.

[0048] Based on the incremental model, the sliding surface dynamics is assumed to be

[0049]

[0050] in The input Δu of the reaction control system is event-triggered update. Let the sequence of event triggering time be j is a natural number 0, 1, 2, ..., used to distinguish different triggering moments), then at t∈[s j ,s j+1 ) The expression of the sliding surface continuous system is

[0051]

[0052] in The feedback value of the last trigger moment is recorded, which will be kept until the next moment t=s j+1 To set the trigger condition, let the event error be

[0053]

[0054] When the event error ∈ j (t) When the threshold ∈ T is exceeded, the event will be triggered and the controller Δu will be updated. T Related to controller design, it will be given after the controller design is completed.

[0055] Design an adaptive dynamic programming controller Δu, taking into account the estimation error, input cost and constraints, and construct a cost function

[0056]

[0057] in, is the cost function of the control process, is an admissible control law. Quadratic utility function U(χ,Δu)=Qχ 2 +RΔu 2 , optimization parameters Q, R>0.

[0058] To estimate the cost function for infinite horizon Building an evaluation network:

[0059]

[0060] in, is the network weight, φ c (χ) is the neural network activation function, Built for evaluation network The partial derivative of the observation value with respect to the comprehensive error variable is

[0061]

[0062] in,

[0063] The adaptive dynamic programming controller constructed by the single evaluation network based on event triggering is:

[0064]

[0065] Furthermore, to optimize the weights of the neural network control law According to the Hamiltonian equation of the robust optimization problem:

[0066]

[0067] To minimize e c , let the objective function Get the weight of the evaluation neural network, that is, the weight of the neural network controller The update law is

[0068]

[0069] According to the design of the adaptive neural network controller, the event trigger threshold is set to

[0070]

[0071] Where 0 < η ∈ <1 is the adaptive adjustment factor. They are the neural network activation gradient functions and system parameters So far, the specific implementation steps of the controller have been explained.

[0072] In summary, the method provided by the embodiment of the present invention is aimed at high-speed aircraft with a reaction control system, designs a pitch channel attitude tracking control scheme, and optimizes the controller parameters online based on the adaptive dynamic programming method to ensure the preset performance index of the tracking error. Figure 2-8 As shown, the method provided by the embodiment of the present invention has the characteristics of good tracking error convergence and fast reaction control trigger response.

[0073] An embodiment of the present invention provides an electronic device, comprising: a computer-readable storage medium and a processor;

[0074] The computer-readable storage medium is used to store executable instructions;

[0075] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method described in any one of the above embodiments.

[0076] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method described in any of the above embodiments.

[0077] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements 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 method for optimizing the preset performance of a gliding high-speed aircraft based on a reaction control system, wherein the gliding high-speed aircraft is equipped with a reaction control system, characterized in that: include: S1, constructing a second-order attitude tracking error system for the gliding high-speed aircraft; S2, establishing a tracking error constraint, and dynamically transforming the tracking error, and determining a sliding surface and sliding surface dynamics of the transformed tracking error based on the transformed tracking error and the second-order attitude tracking error system; S3, dynamically constructing a neural network controller based on the sliding surface and the sliding surface to control the attitude of the gliding high-speed aircraft; Wherein, the neural network controller is are the activation function and weight of the evaluation neural network, which is used to observe the sliding surface. The relevant value function is used to obtain The approximate optimal solution, R>0, is the gain coefficient of the control increment of the reaction control system, is the sliding surface at the last trigger moment, and the trigger condition is the event error is greater than the threshold, j is a natural number used to distinguish different trigger moments; The sliding surface is: in, is the first-order derivative of ε(t), λ is the sliding surface parameter; The sliding surface dynamics are: in, is the first-order derivative of χ0(t), χ0(t) is the sliding surface at the previous moment, Δδ e is the pneumatic control increment, Δu is the reaction system control increment, are the gain coefficients of the pneumatic control increment and the reaction control system control increment respectively.

2. The method according to claim 1, wherein In step S1, the second-order attitude tracking error system is: in, is the second-order derivative of the angle of attack tracking error e, e = α - α d , α and α d are the system state angle of attack and the desired angle of attack command, respectively. ρ is the atmospheric density, V is the velocity, and c are the airfoil reference area and mean aerodynamic chord, respectively. yy is the moment of inertia, are the second-order derivatives of the track inclination and angle of attack commands, C m1 、C m2 、 is the aerodynamic parameter, u=M r is the torque input of the reaction control system, and the system control input δ e is the rudder deflection angle.

3. The method according to claim 1 or 2, wherein: In step S2, the tracking error constraint is: Where 0<κ<1, P(t) is the performance constraint function of the tracking error, and e(0) is the initial error of the angle of attack tracking.

4. The method according to claim 1, wherein In step S3, constructing a controller based on the sliding surface and its dynamics includes: A1, define the input Δu of the reaction control system as an event-triggered update. If the event error at the current moment is greater than the threshold, the event is triggered, the current moment is the next trigger moment, and Δu is updated; A2, constructing the cost function of the dynamic control process in, is the cost function of the control process, is an admissible control law, and the quadratic utility function U(χ,Δu)=Qχ 2 +RΔu 2 , optimization parameters Q, R>0; A3, building an evaluation neural network Used to obtain the observed value of the cost function To build a neural network controller 5. The method according to claim 1, wherein In step S3, the threshold is: Where 0 < η ∈ <1 is the adaptive adjustment factor, They are the neural network activation gradient functions and The maximum value of .

6. The method according to claim 1, wherein The weights of the evaluation neural network The update law To make the objective function Minimize to get; in, η c To evaluate the learning rate of the neural network, 7. An electronic device, characterized in that: include: Computer-readable storage media and processor; The computer-readable storage medium is used to store executable instructions; The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method according to any one of claims 1 to 6.

Images