A control allocation optimization method for aircraft under fault based on weighted pseudo-inverse method

Abstract

The application discloses a control distribution optimization method under aircraft failure based on a weighted pseudo-inverse method, and relates to the field of aircraft fault-tolerant control technology. An aircraft weighted pseudo-inverse control distribution model is established; an expression of a control vector after control distribution is solved; an aircraft failure model is established and a constant matrix is solved; an improved positive definite symmetric weight matrix is constructed; the constant matrix is substituted into the expression of the control vector after failure, the improved positive definite symmetric weight matrix is used to replace the positive definite symmetric weight matrix, and a final control vector after control distribution is obtained; and the final control vector after control distribution is used to realize control distribution under aircraft actuator failure. The application can reduce the use of the failed actuator, prevent further damage of the failed actuator, avoid the event that the fragments of the failed actuator fly in disorder and cause failure of other actuators, and improve the safety and reliability of the flight control of the aircraft.

Description

Technical Field

[0001] This invention relates to the field of fault-tolerant control technology for aircraft, and in particular to a control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method. Background Technology

[0002] With the development of aviation technology, higher requirements have been placed on the safety and reliability of aircraft. A malfunction during flight can lead to serious accidents, causing loss of life and property damage. Therefore, researching ways to improve the reliability of flight control systems is of paramount importance. Fault-tolerant technology refers to the system's tolerance to faults. Depending on the fault-tolerant method, fault-tolerant control can be divided into passive fault-tolerant control and active fault-tolerant control. The starting point of passive fault-tolerant control design is to reduce the system's dependence on the operation of individual components; even in the event of a fault without corrective action, the system can still function. Active fault-tolerant control first needs to automatically and timely detect and diagnose system faults, and then adopt strategies to control or handle the faults. Therefore, active fault-tolerant technology typically includes fault diagnosis and reconfiguration design. Control allocation is an important aspect of active fault-tolerant control. When an actuator on an aircraft malfunctions, the control system can maintain the same control effectiveness by adjusting the use of other actuators. Currently, there is much research on control allocation methods, including pseudo-inverse methods, weighted pseudo-inverse methods, chained methods, direct allocation methods, and optimization-based allocation methods, among others. The pseudo-inverse method is simple and easy to use, and has been widely used in the aerospace field. However, it has certain limitations, namely, the stability of the system and actuator saturation problems after the inversion operation. To address these limitations, experts have improved the pseudo-inverse method, proposing the weighted pseudo-inverse method. The weighted pseudo-inverse method is one of the most widely used control allocation methods in engineering, attracting considerable attention and favor from experts and scholars.

[0003] Currently, the control allocation method for aircraft actuator failures based on the weighted pseudo-inverse method has two main drawbacks. First, the selection of weight parameters in the weight matrix often relies on the experience of experts, which results in weak operability and causes certain inconveniences in practical engineering applications. Second, the weight matrix in the existing weighted pseudo-inverse method is fixed. The fixed weight parameters prevent the aircraft from making specific and targeted control allocations when encountering different failure conditions, thus affecting the control reconfiguration effect. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention introduces actuator fault information to online modify the weight matrix in the weighted pseudo-inverse method, thereby reducing the use of actuators that have already failed, preventing further damage to faulty actuators, and enhancing the flight control safety and reliability of the aircraft.

[0005] A control assignment optimization method based on the weighted pseudo-inverse method for aircraft faults includes the following specific steps:

[0006] Step 1: Establish a weighted pseudo-inverse control allocation model for the aircraft;

[0007] Step 1.1: Establish the aircraft state equations when no operational failure occurs;

[0008] The state equation of the aircraft when no actuator failure occurs is:

[0009]

[0010] Where, x0∈R n Let u0 ∈ R be the state vector. p Let y0 ∈ R be the control vector. m Let A0 ∈ R be the observation vector. n×n B0∈R n ×p C0∈R m×n Let n be a constant matrix, and p, m be the vector dimensions.

[0011] Step 1.2: Establish the aircraft state equations when an actuator failure occurs;

[0012]

[0013] Where, x f ∈R n Let u be the state vector after the fault. f ∈R p The control vector after the fault, y f ∈R m B is the observation vector after the fault. f ∈R n×p C f ∈R m×n It is a constant matrix.

[0014] Step 1.3: Based on the aircraft state equations when there is no actuator failure and when there is an actuator failure, establish a weighted pseudo-inverse control allocation model for the aircraft;

[0015]

[0016] Where W∈R p×p Let J be a positive definite symmetric weight matrix; J is the objective function.

[0017] Step 2: Solve for the expression of the control vector after control allocation based on the weighted pseudo-inverse control allocation model of the aircraft;

[0018] The expression for the control vector after control allocation is:

[0019]

[0020] Among them, superscript Represents the Moore-Penrose inverse.

[0021] Step 3: Establish an aircraft fault model based on the four types of aircraft actuator failures and solve for the constant matrix B. f ;

[0022] The four types of aircraft actuator failures include efficiency loss, jamming, saturation, and looseness.

[0023] The aircraft fault model is as follows:

[0024]

[0025] Where F is the efficiency residual matrix, F∈R p×p F = diag(f1, f2, ..., f p ), where diag represents a diagonal matrix, f i f is the residual efficiency value of the i-th actuator and 0 ≤ f i ≤1, G is the jamming indicator matrix, G∈(0,1), ε is an auxiliary parameter and 0<ε<<1. For the stuck fault value matrix, This represents the fault jam value of the i-th actuator;

[0026] The constant matrix B f for:

[0027]

[0028] Step 4: Construct the improved positive definite symmetric weight matrix W f ;

[0029] The improved positive definite symmetric weight matrix is:

[0030]

[0031] Where k is the weight of the impact of fault information, 0 <k<1;

[0032] Step 5: Convert the constant matrix B obtained in Step 3 into... f Substitute these values ​​into the expression for the control vector after the fault, and use the improved positive definite symmetric weight matrix obtained in step 4 to replace the positive definite symmetric weight matrix W, to solve for the final control vector after control allocation.

[0033] Step 6: Use the control vector after final control allocation to implement control allocation under aircraft actuator failure.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] 1. After an aircraft actuator fails, the traditional weighted pseudo-inverse method can avoid actuator saturation. Based on this, the present invention can reduce the use of the already failed actuator, prevent further damage to the failed actuator, and even avoid the event that the excessive use of the failed actuator may cause debris to fly around and lead to the failure of other actuators, thereby improving the safety and reliability of the aircraft's flight control.

[0036] 2. The impact weight of fault information can be adjusted according to the type of fault, making it more operable, the controller more flexible, and the application range wider. Attached Figure Description

[0037] Figure 1 This is a flowchart of a control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method, as described in an embodiment of the present invention.

[0038] Figure 2 The elevator deflection angle reconstructed using the ordinary weighted pseudo-inverse method in this embodiment of the invention;

[0039] Figure 3 The elevator deflection angle reconstructed using the improved weighted pseudo-inverse method in this embodiment of the invention;

[0040] Figure 4 This is a comparison diagram of angles of attack in an embodiment of the present invention;

[0041] Figure 5 This is a comparison diagram of sideslip angles in an embodiment of the present invention;

[0042] Figure 6 This is a comparison diagram of roll angles in an embodiment of the present invention. Detailed Implementation

[0043] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0044] The key point of this invention is:

[0045] 1. Introduce fault information into the weight matrix so that the weight matrix can be adjusted online according to the fault situation.

[0046] 2. The new weight matrix designed in this invention can change the influence weight of the introduced fault information by adjusting the k value, which greatly improves the adaptability of the control rate.

[0047] A control assignment optimization method based on weighted pseudo-inverse method under aircraft faults, such as Figure 1 As shown, the specific steps include the following:

[0048] Step 1: Establish a weighted pseudo-inverse control allocation model for the aircraft;

[0049] Step 1.1: Establish the aircraft state equations when no operational failure occurs;

[0050] The state equation of the aircraft when no actuator failure occurs is:

[0051]

[0052] Where, x0∈R n Let u0 ∈ R be the state vector. p Let y0 ∈ R be the control vector. m Let A0 ∈ R be the observation vector. n×n B0∈R n ×p C0∈R m×n Let n be a constant matrix, and p, m be the vector dimensions.

[0053] Step 1.2: Establish the aircraft state equations when an actuator failure occurs;

[0054] When an actuator failure occurs in an aircraft, it affects the control efficiency matrix, and its state equation becomes:

[0055]

[0056] Where, x f ∈R n Let u be the state vector after the fault. f ∈R p The control vector after the fault, y f ∈R m B is the observation vector after the fault. f ∈R n×p C f ∈R m×n It is a constant matrix.

[0057] Step 1.3: Based on the aircraft state equations when there is no actuator failure and when there is an actuator failure, establish a weighted pseudo-inverse control allocation model for the aircraft;

[0058] To ensure that the control effect is the same after an actuator failure as under normal conditions, the following weighted pseudo-inverse control allocation model for the aircraft is obtained:

[0059]

[0060] Where W∈R p×p Let J be a positive definite symmetric weight matrix; J is the objective function.

[0061] Step 2: Solve for the expression of the control vector after control allocation based on the weighted pseudo-inverse control allocation model of the aircraft;

[0062] The expression for the control vector after control allocation is:

[0063]

[0064] Among them, superscript Represents the Moore-Penrose inverse.

[0065] Step 3: Establish an aircraft fault model based on the four types of aircraft actuator failures and solve for the constant matrix B. f ;

[0066] There are four types of aircraft actuator failures: efficiency loss, jamming, saturation, and slack. Actuator efficiency loss mainly refers to the reduction in control efficiency caused by partial damage to the aircraft's control surfaces; actuator jamming means that the aircraft's control surfaces are fixed in a certain position and cannot respond normally to control commands; when the control surfaces are jammed at the maximum or minimum position of their actuation range, it is called actuator saturation; slack means that a certain aircraft control surface is not affected by command control and swings arbitrarily with the wind direction, having no effect on the control input.

[0067] The aircraft fault model is as follows:

[0068]

[0069] Where F is the efficiency residual matrix, F∈R p×p F = diag(f1, f2, ..., f p ), where diag represents a diagonal matrix, f i f is the residual efficiency value of the i-th actuator and 0 ≤ f i ≤1, G is the jamming indicator matrix, G∈(0,1), ε is an auxiliary parameter and 0<ε<<1. For the stuck fault value matrix, This represents the fault jam value of the i-th actuator;

[0070] Formula (5) can be used to represent the four types of actuator faults using a single parameter, and further, the constant matrix B in formula (4) can be obtained. f ;

[0071]

[0072] Step 4: Construct the improved positive definite symmetric weight matrix W f ;

[0073] The selection of the weight matrix W has a very significant impact on the effect of control allocation. The larger the value of the diagonal element of the W matrix is selected, the lower the amplitude of the corresponding actuator deflection tends to be, and vice versa. The currently commonly used method for selecting weight parameters is to select the reciprocal of the maximum limit amplitude, as shown in formula (7).

[0074] The weight parameters in the weight matrix W can be selected as:

[0075]

[0076] where is the saturation value of the i-th actuator.

[0077] Based on this, the present invention makes improvements by introducing fault information to obtain an improved positive definite symmetric weight matrix:

[0078]

[0079] where k is the weight of the influence of fault information, 0 < k < 1. The larger k is, the higher the weight value of the fault information, and vice versa. Therefore, the weight matrix in the control law can be changed online, and the use of the already faulty actuator is considered to be reduced to prevent further damage to the faulty actuator. The safety and reliability of the flight control of the aircraft are improved.

[0080] Step 5: Substitute the constant matrix B obtained in Step 3 f into the expression of the control vector after the fault, and at the same time use the improved positive definite symmetric weight matrix obtained in Step 4 to replace the positive definite symmetric weight matrix W, and solve to obtain the final control vector after control allocation;

[0081] Step 6: Use the final control vector after control allocation to implement control allocation under the actuator fault of the aircraft.

[0082] In this embodiment, the final control vector after control allocation is input into the aircraft dynamics model, and fault-tolerant control of the aircraft after a fault can be completed without changing the structure of the original system controller.

[0083] When a 40% efficiency loss fault of the right elevator is introduced at 15 s, the simulation results show that compared with the ordinary weighted pseudoinverse method, the improved weighted pseudoinverse method can reduce the use of the faulty actuator, as Figure 2 shown in Figure 3 ; and the improved weighted pseudoinverse method can well complete the fault-tolerant control task, that is, restore the angle of attack, sideslip angle, and roll angle to the normal level, as Figures 4 to 6 shown in.

Claims

1. A control allocation optimization method based on weighted pseudo-inverse method under aircraft faults, characterized in that, Includes the following steps: Step 1: Establish a weighted pseudo-inverse control allocation model for the aircraft; Step 2: Solve for the expression of the control vector after control allocation based on the weighted pseudo-inverse control allocation model of the aircraft; Step 3: Establish aircraft fault models based on the four types of aircraft actuator failures and solve for the constant matrix. ; The four types of aircraft actuator failures include efficiency loss, jamming, saturation, and looseness. The constant matrix for: （6）； in, It is a constant matrix. For control vectors, Let the dimension be the vector. For the efficiency residual matrix, For the efficiency residual matrix, , ,in Represents a diagonal matrix. For the first The remaining efficiency value of each actuator and , For the stuck indicator matrix, , , For auxiliary parameters and , For the stuck fault value matrix, , For the first Fault jamming value of an actuator; superscript " ” represents the Moore-Penrose inverse; Step 4: Construct the improved positive definite symmetric weight matrix ; The improved positive definite symmetric weight matrix is: （8） in, The weighting of fault information ; For the first Saturation value of each actuator; Step 5: Convert the constant matrix obtained in Step 3 into a single matrix. Substitute these values ​​into the expression for the control vector after the fault, and simultaneously use the improved positive definite symmetric weight matrix obtained in step 4 to replace the positive definite symmetric weight matrix. Solving for the final control vector after control allocation yields the solution. Step 6: Use the control vector after final control allocation to implement control allocation under aircraft actuator failure.

2. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 1, characterized in that, Step 1 specifically includes: Step 1.1: Establish the aircraft state equations when no operational failure occurs; Step 1.2: Establish the aircraft state equations when an actuator failure occurs; Step 1.3: Based on the aircraft state equations when the aircraft does not experience actuator failure and when the aircraft experiences actuator failure, establish a weighted pseudo-inverse control allocation model for the aircraft.

3. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 2, characterized in that, The state equation of the aircraft when no actuator failure occurs is: （1） in, For state vectors, For the observation vector, , It is a constant matrix. Let be the dimension of the vector.

4. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 3, characterized in that, The state equation of the aircraft when an actuator failure occurs is as follows: （2） in, This is the state vector after the fault. This is the control vector after the fault. This is the observation vector after the fault. , It is a constant matrix.

5. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 4, characterized in that, The weighted pseudo-inverse control allocation model for the aircraft is as follows: （3） in, It is a positive definite symmetric weight matrix; The objective function is denoted as .

6. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 5, characterized in that, The expression for the control vector after control allocation is: （4）。 7. The control allocation optimization method for aircraft faults based on the weighted pseudo-inverse method according to claim 6, characterized in that, The aircraft fault model is as follows: （5）。

Images