Self-adaptive dynamic inverse fault-tolerant control method for heavy carrier rocket
By modeling the attitude dynamics and designing the fault model of heavy launch vehicles, combining dynamic inverse theory and Liyapunov stability theory, an adaptive dynamic inverse fault tolerance control method is proposed, which solves the attitude stability problem of heavy launch vehicles in the case of faults and achieves higher safety and robustness.
Patent Information
- Application Number
- CN202310652345.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-04
- Publication Date
- 2025-02-25
AI Technical Summary
During the flight of heavy carrier rockets, due to the complex flight environment and complex system design, actuator gain failure and bias failure are prone to occur, resulting in challenges in control system design. The existing fault-tolerant control methods are relatively complex, making it difficult to achieve effective fault-tolerant control.
An adaptive dynamic inverse fault tolerance control method is proposed. By modeling the attitude dynamics of heavy-duty launch vehicles, a model of actuator gain fault and bias fault is established, the control law under ideal conditions is designed, and the adaptive fault tolerance control law is designed in case of failures, and stability analysis is performed based on the Liyapunov stability theory.
It is realized that in the event of non-fatal failures, heavy launch vehicles can achieve stable attitude flight, enhance the safety and robustness of the flight, and improve the solution to fault-tolerant control problems.
Smart Images
Figure CN119511685A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of aerospace technology, and relates to an adaptive dynamic inverse fault-tolerant control method, and in particular to an adaptive dynamic inverse fault-tolerant control method for a heavy-duty launch vehicle. Background Art
[0002] Heavy-lift launch vehicles are the prerequisite for large-scale space exploration and development, and have become the research focus of many countries. Heavy-lift launch vehicles are responsible for manned spacecraft and other missions. At present, due to the immaturity of launch technology, launch missions still fail from time to time. Once a launch mission fails, it will not only bring huge economic losses. The research of robust and reliable flight control systems is essential.
[0003] The flight environment of heavy-lift launch vehicles is complex, and they are subject to large external disturbances during flight. At the same time, their own system design is becoming more and more complex. When the engine is involved in control, the number of servo mechanisms increases and the probability of failure increases, which brings great challenges to the control system design.
[0004] In addition, for the adaptive fault-tolerant control problem of heavy-lift launch vehicles, the actuator gain fault and bias fault must be considered in the control process, and the research on the fault-tolerant control problem of heavy-lift launch vehicles is relatively complicated. Therefore, it is urgent to study a more effective fault-tolerant control method for heavy-lift launch vehicles. Summary of the invention
[0005] Technical issues to be solved
[0006] In order to avoid the shortcomings of the prior art, the present invention proposes an adaptive dynamic inverse fault-tolerant control method for a heavy-lift launch vehicle, comprising: modeling the attitude dynamics of the heavy-lift launch vehicle, establishing models of actuator gain fault and bias fault; simplifying the attitude dynamics model to obtain a control-oriented system model, and designing a control law under ideal conditions based on dynamic inverse theory; designing an adaptive fault-tolerant control law for a heavy-lift launch vehicle with actuator gain fault and bias fault; completing a stability analysis of the designed adaptive fault-tolerant control law based on the Lyapunov stability theory, thereby achieving attitude-stable flight of the heavy-lift launch vehicle in the event of a non-fatal fault.
[0007] Technical Solution
[0008] An adaptive dynamic inverse fault-tolerant control method for a heavy-lift launch vehicle is characterized by the following steps:
[0009] Step 1: Build the attitude dynamics motion model and fault model of the heavy-lift launch vehicle:
[0010] The attitude dynamics motion model of the heavy-lift launch vehicle is:
[0011]
[0012] Said:
[0013]
[0014]
[0015]
[0016] in, represents the input of the controlled system, δ γ represents the roll channel rudder deflection, δ ψ Represents the rudder deviation of the yaw channel, Represents the rudder deviation of the pitch channel; Represent the state of the controlled system, γ T represents the roll angle, ψ T represents the yaw angle, represents the pitch angle, ω Tx represents the roll angular velocity, ω Ty represents the yaw angular velocity, ω Tz represents the pitch angular velocity; is the known nonlinear smooth function matrix; I x ,I y ,I z represents the moment of inertia of the heavy-lift launch vehicle about each axis of the coordinate system; represents the partial derivative of the yaw moment coefficient; represents the partial derivative of the pitch moment coefficient; q represents the dynamic pressure under the current flight conditions; s represents the reference area; L represents the reference length; α represents the angle of attack; β represents the sideslip angle; P1 represents the thrust of a single engine; L x Represents the distance between the engine nozzle swing center and the X-axis of the rocket body; L y ,L z Represents the distance between the projection of the engine thrust point on the X-axis and the projection of the center of mass on the X-axis; represents the output of the controlled system; Q(u) represents the fault model input of u(t);
[0017] The heavy-lift launch vehicle failure model is:
[0018] Q(u)=Λ(t)u+Δ(t) (2)
[0019] in, is the control efficiency matrix of the actuator, Λ i,i (t)∈(0,1], i=1,2,3; Bias failure of the actuator;
[0020] Step 2: According to the attitude dynamics motion model of the heavy-lift launch vehicle, design the control law under ideal conditions, which includes the inner loop virtual control law and the outer loop control law;
[0021] The inner loop virtual control law:
[0022] Where: k1 is the inner loop virtual control law parameter; is the expected angular differential value of the three channels; e1 is the dynamic equation of the tracking error of the inner loop subsystem;
[0023] The outer loop control law:
[0024] Where: k2 is the outer loop control law parameter, is the differential value of the inner loop virtual control law; e2 is the tracking error of the outer loop subsystem;
[0025] Step 3: According to the attitude dynamics motion model and fault model of the heavy-lift launch vehicle, the actuator gain fault model is established, and the adaptive fault-tolerant control law under the actuator gain fault is designed:
[0026] The actuator gain fault model is: Q gain (u)=Λu (6)
[0027] Where: Λ is the actuator gain fault size;
[0028] The adaptive fault-tolerant control law under the actuator gain failure is: gain =u+u a (7)
[0029] Among them: u is the outer loop control law, u a is the control law compensation term;
[0030] The control law compensation term is:
[0031]
[0032] in: is an estimate of W; definition Where W = Λ -1 (Λ-I), I is the unit matrix;
[0033]
[0034] is the adaptive law, η is the adaptive law gain, and P is a positive definite symmetric matrix;
[0035] Step 4: According to the attitude dynamics motion model and fault model of the heavy-lift launch vehicle, establish the actuator bias fault model and design the adaptive fault-tolerant control law under the actuator bias fault:
[0036] The actuator bias fault model is:
[0037] Q bias (u)=u+Δ (10)
[0038] Where: Δ is the actuator bias fault size;
[0039] The actuator bias fault adaptive fault-tolerant control law is:
[0040] u bias =u+u d (11)
[0041] Where: u is the ideal control law designed in step 2, u d is the control law compensation term;
[0042] The control law compensation term
[0043]
[0044] in: is an estimate of Δ, and is defined as
[0045]
[0046] is the adaptive law expression, G is the adaptive law gain, and P is a positive definite symmetric matrix;
[0047] Step 5: Apply the adaptive fault-tolerant control law u designed in step 3 under the actuator gain fault gain , the actuator bias fault adaptive fault-tolerant control law u designed in step 4 bias , respectively, as the controller input of the heavy-duty launch vehicle with gain fault and bias fault in the actuator. When the inner loop tracking error tends to zero and the outer loop tracking error tends to zero, the attitude of the heavy-duty launch vehicle tends to be stable.
[0048] The dynamic equation of the inner loop subsystem tracking error in step 2 is: e1 = x1-y d , where: y d is the expected angle of three channels.
[0049] The tracking error dynamic equation of the outer loop subsystem in step 2 is: e2 = x2-x 2c , where: e2 is the tracking error of the outer loop subsystem; x 2c is the inner loop virtual control law.
[0050] Beneficial Effects
[0051] The present invention proposes an adaptive dynamic inverse fault-tolerant control method for a heavy-duty launch vehicle, which includes modeling the attitude dynamics of the heavy-duty launch vehicle, establishing a model of actuator gain fault and bias fault; simplifying the attitude dynamics model to obtain a control-oriented system model, and designing an ideal control law based on the dynamic inverse theory; designing an adaptive fault-tolerant control law for a heavy-duty launch vehicle with actuator gain fault and bias fault; completing the stability analysis of the designed adaptive fault-tolerant control law based on the Lyapunov stability theory, and realizing that the heavy-duty launch vehicle can achieve attitude-stable flight in the event of a non-fatal fault. The present invention has a good improvement effect on solving the fault-tolerant control problem of heavy-duty launch vehicles, enhances the safety and robustness of heavy-duty launch vehicle flight, and has a good application prospect in the field of aerospace.
[0052] The present invention is aimed at heavy-lift launch vehicles with gain faults and bias faults in actuators, and designs an adaptive fault-tolerant attitude controller for heavy-lift launch vehicles based on dynamic inversion. The main reason why the controller can achieve accurate tracking is that the actuator adopts an adaptive control method to deal with possible faults: first, the uncertainties Λ and Δ caused by gain faults and bias faults are adaptively estimated and compensated, and then an additional control law is used to Force the nonlinear terms B2(Λ-I)u and B2Δ to converge to 0, and let Reach its expected value W = Λ -1 (Λ-I), Δ. Therefore, it is possible to achieve attitude stability for heavy-duty launch vehicles in the event of a failure. Finally, simulation analysis is performed, and the results show that the controller designed by the present invention has excellent performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 It is a schematic diagram of the overall flow of the adaptive fault-tolerant control method for a heavy-lift launch vehicle provided by the present invention;
[0054] Figure 2 is a roll angle tracking effect diagram under Example 1;
[0055] Figure 3 This is a diagram showing the yaw angle tracking effect under Example 1;
[0056] Figure 4 is a pitch angle tracking effect diagram under Example 1;
[0057] Figure 5 is a roll angle tracking effect diagram under Example 2;
[0058] Figure 6 This is a diagram showing the yaw angle tracking effect under Example 2;
[0059] Figure 7This is a pitch angle tracking effect diagram under Example 2;
[0060] Figure 8 is a roll angle tracking effect diagram under Example 3;
[0061] Fig. 9 This is a diagram showing the yaw angle tracking effect under Example 3;
[0062] Fig.10 This is a pitch angle tracking effect diagram under Example 3; DETAILED DESCRIPTION
[0063] The present invention will now be further described with reference to the embodiments and the accompanying drawings:
[0064] like Figure 1 As shown, an adaptive dynamic inverse fault-tolerant control method comprises the following steps:
[0065] 1. Build the attitude dynamics motion model and fault model of heavy-lift launch vehicles;
[0066] Step ①: The attitude dynamics motion model of the heavy-lift launch vehicle is expressed as:
[0067]
[0068] in:
[0069]
[0070]
[0071]
[0072] in, represents the input of the controlled system, δ γ represents the roll channel rudder deflection, δ ψ Represents the rudder deviation of the yaw channel, Represents the rudder deviation of the pitch channel; Represent the state of the controlled system, γ T represents the roll angle, ψ T represents the yaw angle, represents the pitch angle, ω Tx represents the roll angular velocity, ω Ty represents the yaw angular velocity, ω Tz represents the pitch angular velocity; is the known nonlinear smooth function matrix; I x ,I y ,I z represents the moment of inertia of the heavy-lift launch vehicle about each axis of the coordinate system; represents the partial derivative of the yaw moment coefficient; represents the partial derivative of the pitch moment coefficient; q represents the dynamic pressure under the current flight conditions; s represents the reference area; L represents the reference length; α represents the angle of attack; β represents the sideslip angle; P1 represents the thrust of a single engine; L x Represents the distance between the engine nozzle swing center and the X-axis of the rocket body; L y ,L z Represents the distance between the projection of the engine thrust point on the X-axis and the projection of the center of mass on the X-axis; Represents the output of the controlled system; Q(u) represents the fault model input of u(t).
[0073] Step ②: Considering the gain fault and bias fault of the actuator, the fault model Q(u) is expressed as:
[0074] Q(u)=Λ(t)u+Δ(t) (15)
[0075] in, is the control efficiency matrix of the actuator, Λ i,i (t)∈(0,1], i=1,2,3; It is the bias fault of the actuator.
[0076] 2. Based on the attitude dynamics model established in step 1-①, design the control law under ideal conditions;
[0077] Step ①: Establish the dynamic equation of the inner loop subsystem tracking error and the dynamic equation of the outer loop subsystem tracking error;
[0078] The dynamic equations of the inner loop subsystem tracking error and the outer loop subsystem tracking error are expressed as follows:
[0079]
[0080] Where e1 is the tracking error of the inner loop subsystem; e2 is the tracking error of the outer loop subsystem; y d is the expected angle of the three channels; x 2c is the inner loop virtual control law.
[0081] Step ②: Based on the two dynamic equations in step 2-①, design the inner loop virtual control law and the outer loop control law;
[0082] The inner loop virtual control law is expressed as:
[0083]
[0084] Among them, k1 is the inner loop virtual control law parameter; is the expected angular differential value of the three channels.
[0085] The outer loop control law is expressed as:
[0086]
[0087] Among them, k2 is the outer loop control law parameter, is the derivative value of the inner loop virtual control law.
[0088] Step ③: Perform stability analysis on the control law designed in step 2-②;
[0089] According to formula (17) and formula (18), formula (16) is derived and the expression is as follows:
[0090]
[0091] Construct the Lyapunov function, the expression is as follows:
[0092]
[0093] By taking the derivative of formula (20), we can get:
[0094]
[0095] Substituting formula (19) into formula (21), we can obtain:
[0096]
[0097] Therefore, it can be seen that the closed-loop system converges asymptotically and the tracking error converges asymptotically to 0.
[0098] 3. Based on the attitude dynamics model established in step 1-① and the fault model established in step 1-②, establish an actuator gain fault model and design an adaptive fault-tolerant control law under actuator gain fault;
[0099] Step ①: Establish the actuator gain fault model;
[0100] The fault model is expressed as:
[0101] Q gain (u)=Λu (23)
[0102] Where Λ is the actuator gain fault size.
[0103] Step ②: Consider the fault model of step 3-① and design an adaptive fault-tolerant control law;
[0104] The adaptive fault-tolerant control law is expressed as:
[0105] u gain =u+u a (twenty four)
[0106] Where u is the ideal control law designed in step 2-②, u a is the control law compensation term.
[0107] The control law compensation term is expressed as:
[0108]
[0109] in, is an estimate of W; definition Where W = Λ -1 (Λ-I), I is the unit matrix.
[0110] The adaptive law expression is as follows:
[0111]
[0112] Wherein, η is the adaptive law gain, and P is a positive definite symmetric matrix.
[0113] Step ③: Perform stability analysis on the control law designed in step 3-②;
[0114] Derivative of formula (16) yields the following expression:
[0115]
[0116] Substituting formula (23) into formula (27), the expression is as follows:
[0117]
[0118] Substitute the designed adaptive fault tolerance law (24) into (28), that is, u = u gain , the expression is as follows:
[0119]
[0120] Substituting formula (25) into formula (29), the expression is as follows:
[0121]
[0122] Construct the Lyapunov function as
[0123]
[0124] Taking the derivative we get:
[0125]
[0126] Substituting formula (26) into formula (32), we can get the following expression:
[0127]
[0128] Therefore, it can be seen that the closed-loop system converges asymptotically and the tracking error converges asymptotically to 0.
[0129] 4. Based on the attitude dynamics model established in step 1-① and the fault model established in step 1-②, establish an actuator bias fault model and design an adaptive fault-tolerant control law under actuator bias fault;
[0130] Step ①: Establish the actuator bias fault model;
[0131] The fault model is expressed as:
[0132] Q bias (u)=u+Δ (34)
[0133] Where Δ is the actuator bias fault size.
[0134] Step ②: Consider the fault model of step 4-① and design an adaptive fault-tolerant control law;
[0135] The adaptive fault-tolerant control law is expressed as:
[0136] u bias =u+u d (35)
[0137] Where u is the ideal control law designed in step 2-②, u d is the control law compensation term.
[0138] The control law compensation term is expressed as:
[0139]
[0140] in, is an estimate of Δ, and is defined as
[0141] The adaptive law expression is as follows:
[0142]
[0143] Wherein, G is the adaptive law gain, and P is a positive definite symmetric matrix.
[0144] Step ③: Perform stability analysis on the control law designed in step 4-②;
[0145] Derivative of formula (16) yields the following expression:
[0146]
[0147] Substituting formula (34) into formula (38), the expression is as follows:
[0148]
[0149] Substitute the designed adaptive fault-tolerant law formula (35) into formula (39), that is, u = u bias , the expression is as follows:
[0150]
[0151] Substituting formula (36) into formula (40), the expression is as follows:
[0152]
[0153] Construct the Lyapunov function as
[0154]
[0155] Taking the derivative we get:
[0156]
[0157] Substituting formula (37) into formula (43), we can get the following expression:
[0158]
[0159] Therefore, it can be seen that the closed-loop system converges asymptotically and the tracking error converges asymptotically to 0.
[0160] 5. Apply the designed control law to the heavy-lift launch vehicle. When the inner-loop tracking error and the outer-loop tracking error approach zero, the attitude of the heavy-lift launch vehicle approaches stability.
[0161] Specific embodiments The adaptive dynamic inverse fault-tolerant control method proposed in the present invention is verified in the MATLAB environment and simulated in three different situations. The simulation results of the three different situations are as follows:
[0163] Example 1: The heavy-lift launch vehicle flies under ideal conditions, and no actuator gain failure or bias failure occurs;
[0164] Example 2: A heavy-lift launch vehicle flies when a gain failure occurs in the actuator, and the actuator gain failure matrix is selected as follows:
[0165]
[0166] Example 3: A heavy-lift launch vehicle flies when an actuator bias fault occurs. The actuator bias fault matrix is selected as follows:
[0167]
[0168] Figure 2-Figure 4 is the rolling angle, yaw angle, and pitch angle tracking effect curve in Example 1, Figure 5-Figure 7 is the rolling angle, yaw angle, and pitch angle tracking effect curve in Example 2, Figure 8-Figure 10 : is the tracking effect curve of the roll angle, yaw angle and pitch angle in Example 3. It can be seen that the heavy-lift launch vehicle actuator can achieve attitude stable flight and converge to the expected value in the event of a non-fatal failure.
[0169] Figure 2-Figure 4 In order to make the heavy-lift launch vehicle fly under ideal conditions, the dynamic inverse control method can make the actual output values of the three channels stabilize to the expected values in about 10s, with a fast convergence speed. Figure 5-Figure 7 In order to fly a heavy-lift launch vehicle under the condition of gain failure of the actuator, the adaptive dynamic inverse fault-tolerant control method designed in this paper can make the actual output value of the three channels stabilize to the expected value in about 15s. The overshoot is smaller than that in the ideal case, and the convergence time becomes longer, but the heavy-lift launch vehicle can still fly stably. Figure 8-Figure 10 For the heavy-lift launch vehicle flying under the condition of bias failure of the actuator, the actual output value of the three channels stabilizes to the expected value in about 8s, and the overshoot and convergence time are smaller than those in the above two cases. Through the simulation of the three cases, it can be seen that when the heavy-lift launch vehicle actuator has a gain failure or a bias failure, the method designed in this paper can still ensure its stable flight and converge to the expected value.
Claims
1. An adaptive dynamic inverse fault-tolerant control method for a heavy-lift launch vehicle, characterized in that Here are the steps: Step 1: Build the attitude dynamics motion model and fault model of the heavy-lift launch vehicle: The attitude dynamics motion model of the heavy-lift launch vehicle is: y(t)=x1(t) Said: in, represents the input of the controlled system, δ γ represents the roll channel rudder deflection, δ ψ Represents the rudder deviation of the yaw channel, Represents the rudder deviation of the pitch channel; Represent the state of the controlled system, γ T represents the roll angle, ψ T represents the yaw angle, represents the pitch angle, ω Tx represents the roll angular velocity, ω Ty represents the yaw angular velocity, ω Tz represents the pitch angular velocity; is the known nonlinear smooth function matrix; I x ,I y ,I z represents the moment of inertia of the heavy-lift launch vehicle about each axis of the coordinate system; represents the partial derivative of the yaw moment coefficient; represents the partial derivative of the pitch moment coefficient; q represents the dynamic pressure under the current flight conditions; s represents the reference area; L represents the reference length; α represents the angle of attack; β represents the sideslip angle; P1 represents the thrust of a single engine; L x Represents the distance between the engine nozzle swing center and the X-axis of the rocket body; L y ,L z Represents the distance between the projection of the engine thrust point on the X-axis and the projection of the center of mass on the X-axis; represents the output of the controlled system; Q(u) represents the fault model input of u(t); The heavy-lift launch vehicle failure model is: Q(u)=Λ(t)u+Δ(t) in, is the control efficiency matrix of the actuator, Λ i,i (t)∈(0,1], i=1,2,3; Bias failure of the actuator; Step 2: According to the attitude dynamics motion model of the heavy-lift launch vehicle, design the control law under ideal conditions, which includes the inner loop virtual control law and the outer loop control law; The inner loop virtual control law: Where: k1 is the inner loop virtual control law parameter; is the expected angular differential value of the three channels; e1 is the dynamic equation of the tracking error of the inner loop subsystem; The outer loop control law: Among them: k2 is the outer loop control law parameter, is the differential value of the inner loop virtual control law; e2 is the tracking error of the outer loop subsystem; Step 3: According to the attitude dynamics motion model and fault model of the heavy-lift launch vehicle, the actuator gain fault model is established, and the adaptive fault-tolerant control law under the actuator gain fault is designed: The actuator gain fault model is: Q gain (u)=Λu Where: ΛΛ is the actuator gain fault size; The adaptive fault-tolerant control law under the actuator gain failure is: gain =u+u a Among them: u is the outer loop control law, u a is the control law compensation term; The control law compensation term is: in: is an estimate of W; definition Where W = Λ -1 (Λ-I), I is the unit matrix; is the adaptive law, η is the adaptive law gain, and P is a positive definite symmetric matrix; Step 4: According to the attitude dynamics motion model and fault model of the heavy-lift launch vehicle, establish the actuator bias fault model and design the adaptive fault-tolerant control law under the actuator bias fault: The actuator bias fault model is: Q bias (u)=u+Δ Where: Δ is the actuator bias fault size; The actuator bias fault adaptive fault-tolerant control law is: in bias =u+u d Where: u is the ideal control law designed in step 2, u d is the control law compensation term; The control law compensation term in: is an estimate of Δ, and is defined as is the adaptive law expression, G is the adaptive law gain, and P is a positive definite symmetric matrix; Step 5: Apply the adaptive fault-tolerant control law u designed in step 3 under the actuator gain fault gain , the actuator bias fault adaptive fault-tolerant control law u designed in step 4 bias , respectively, as the controller input of the heavy-duty launch vehicle with gain fault and bias fault in the actuator. When the inner loop tracking error tends to zero and the outer loop tracking error tends to zero, the attitude of the heavy-duty launch vehicle tends to be stable.
2. The adaptive dynamic inverse fault-tolerant control method for a heavy-lift launch vehicle according to claim 1, characterized in that: The dynamic equation of the inner loop subsystem tracking error in step 2 is: e1 = x1-y d , where: y d is the expected angle of three channels.
3. The adaptive dynamic inverse fault-tolerant control method for a heavy-lift launch vehicle according to claim 1, characterized in that: The tracking error dynamic equation of the outer loop subsystem in step 2 is: e2 = x2-x 2c , where: e2 is the tracking error of the outer loop subsystem; x 2c is the inner loop virtual control law.