Active fault-tolerant control and optimization method for aviation fuel adjusting pump test equipment

Abstract

The invention particularly relates to an active fault-tolerant control and optimization method for aviation fuel adjusting pump test equipment, and the method comprises the steps: carrying out the design of a linear variable parameter active fault-tolerant controller through employing a self-adaptive sliding mode control technology and a self-adaptive radial basis function neural network based on fault estimation information obtained by a linear variable parameter self-adaptive sliding mode observer; a self-adaptive gain change law and a self-adaptive RBF algorithm weight self-adaptive updating law are designed, accurate fitting of unknown interference is achieved, the control performance of a sliding-mode fault-tolerant controller is improved, and optimization and improvement of the performance of an active fault-tolerant controller of aviation fuel adjusting pump test equipment are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of aviation fuel regulating pump test equipment control, and in particular to an active fault-tolerant control and optimization method for aviation fuel regulating pump test equipment. Background Art

[0002] In related technologies, aviation fuel conditioning pump test equipment is a precision device used for testing the performance of fuel regulators and fuel pumps. It is a key component of aircraft engine fuel supply, and the quality of the fuel conditioning pump affects the overall performance of aircraft engines. By integrating multiple advanced technologies, this equipment enables full-function testing of fuel regulators and fuel pumps, including servo motor drive, test circuit switching, PLC (Programmable Logic Controller) control system, safety protection settings, and displacement servo control system, enabling precise control of flow parameters. However, because the fuel conditioning pump test equipment is a complex thermodynamically coupled system operating under high pressure and high speed conditions, it has very high safety requirements. Any equipment failure will have a significant impact on operational stability. Therefore, to improve the safety and reliability of the equipment, effective measures must be taken to accurately diagnose and isolate the fuel conditioning pump test equipment to ensure its normal operation.

[0003] In the prior art, fault-tolerant control methods for aviation fuel conditioning pump test equipment can be divided into two categories: active fault-tolerant control (ATFC) and passive fault-tolerant control (PFTC). Among them, the passive fault-tolerant control method is a scheme that uses robust control theory to consider faults as a disturbance in control design. This method does not require a fault diagnosis scheme or reconfiguration of the controller; however, its fault tolerance is limited and will cause partial performance loss of the controller. Active fault-tolerant control uses the online information of the system after the fault to synchronously reconstruct the controller design to maintain system stability and achieve the desired dynamic and steady-state performance. This method can reconfigure the controller using fault information provided by the online fault diagnosis system, and can use the available configurations in the system to make structural changes to achieve optimal fault-tolerant control performance, with greater adaptability. Among them, the sliding mode variable structure control method is a method that has been widely studied and applied to control systems.

[0004] However, there are certain defects in the existing technical solutions. For example: 1) The aviation fuel regulating pump test equipment has complex working conditions. The parameters of the system model under multiple working conditions will change with the scheduling parameters at all times. The single fixed sliding surface generally designed in the existing technology is difficult to ensure the effect of global fault-tolerant control; 2) Sensor failures are not considered in the existing solutions; 3) The aviation fuel regulating pump test equipment may be under complex interference and fault shocks. The use of robust theory to suppress observer errors and system interference has certain limitations.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0006] The present invention provides an active fault-tolerant control and optimization method for aviation fuel regulating pump test equipment, a storage medium, a computer program product, and an electronic device, which solve the potential problems of sensor and actuator failure faced by aviation fuel regulating pump test equipment under multiple working conditions in the prior art, construct a dynamic active fault-tolerant control system, and thus can overcome the defects in the prior art to a certain extent.

[0007] Other features and advantages of the present invention will become apparent from the following detailed description, or may be learned in part by practice of the present invention.

[0008] According to a first aspect of the present invention, there is provided an active fault-tolerant control and optimization method for an aviation fuel conditioning pump test equipment, the method comprising:

[0009] Using operating data from an aviation fuel conditioning pump test device, a linear variable parameter model of the aircraft fuel conditioning pump test device is constructed to describe the dynamic characteristics of the aircraft fuel conditioning pump test device under different operating conditions. The operating data includes the system speed, outlet pressure, and flow rate corresponding to different fuel metering valve openings.

[0010] Based on the linear variable parameter model, an affine linear variable parameter fault model is constructed by combining the actuator fault, sensor fault, and external interference characteristics. The affine linear variable parameter fault model includes an actuator fault subsystem for expressing the actuator fault and a sensor fault subsystem for expressing the sensor fault, thereby separating the actuator fault from the sensor fault.

[0011] Based on the adaptive observer gain term, the first sliding mode surface corresponding to the actuator fault observer, and the second sliding mode surface corresponding to the sensor fault observer, an adaptive upper limit sliding mode observer is configured; wherein the adaptive upper limit sliding mode observer includes: the actuator fault observer and the sensor fault observer, each of which is configured with an adaptive observer gain term;

[0012] An adaptive upper limit-based sliding mode observer configures an adaptive gain sliding mode controller for the affine linear variable parameter fault model, so as to be used for active fault-tolerant control of an aviation fuel adjustment pump test device.

[0013] In some exemplary embodiments, the method further comprises:

[0014] The target Lyapunov function is configured for the sliding mode observer with adaptive upper bound to eliminate the influence of unknown upper bound of fault on the convergence of system state.

[0015] In some exemplary embodiments, the method further comprises:

[0016] An adaptive RBF network is used to provide disturbance and estimation error for an adaptive gain sliding mode controller.

[0017] In some exemplary embodiments, the adaptive RBF network includes an input layer, a hidden layer, and an output layer arranged in sequence; wherein the input parameter of the input layer is the system state variable x, which is a g-dimensional variable; the hidden layer includes l neurons, and the activation function of each neuron has a Gaussian basis function form; the output layer is used to output an interference estimate.

[0018] In some exemplary embodiments, constructing a linear variable parameter model of the aviation fuel adjustment pump test equipment using operating data of the aviation fuel adjustment pump test equipment includes:

[0019] For the fuel metering valve between the fully closed state and the fully open state, N fuel metering valve opening values ​​are selected at equal intervals, and the corresponding system speed, outlet pressure and flow rate data are collected at a preset sampling frequency to obtain multiple sets of operating data;

[0020] Configure the fuel metering valve opening FMV as the input variable u(t); configure the system speed n, outlet pressure P out and flow rate W as output variable y(t); define system speed n as state variable x(t), and configure it as scheduling parameter;

[0021] Identify each set of operating data and fit the model parameters to obtain N sets of linear state space models S j ={A j ,B j ,C}, expressed as:

[0022]

[0023] Among them, x(t) is the state variable, u(t) is the input variable, and y(t) is the output variable; A j ,B j ,C are the state transfer matrix, control input matrix and observation matrix of the system respectively, which is a constant matrix;

[0024] The coefficient matrix is ​​solved for N groups of linear state space models and the corresponding scheduling parameters. Based on the coefficient matrix, a linear variable parameter model of the aviation fuel pump test equipment is constructed to describe the dynamic characteristics of the system under different working conditions. The model is expressed as:

[0025]

[0026] Among them, ρ(t) is the scheduling parameter; A(ρ(t)) and B(ρ(t)) are coefficient matrices respectively; the coefficient matrix can be solved based on the system speed and the multi-vertex scheduling matrix.

[0027] In some exemplary embodiments, the constructing of an affine linear variable parameter fault model based on the linear variable parameter model and combining actuator fault, sensor fault, and external interference features includes:

[0028] Define bounded functions corresponding to actuator faults, sensor faults, and system uncertainty, and transform the linear variable parameter model based on the preconfigured transformation matrix to obtain an affine linear variable parameter fault model:

[0029]

[0030] in, are the transformed system matrices, is the state variable after conversion, is the output variable after conversion; the conversion matrix T i =[T i,1 ,T i,2 ] T , the transformation matrix S i =[S i,1 ,S i,2 ] T ;

[0031] Considering the coupling between different types of faults, the affine linear variable parameter fault model is split into the actuator fault subsystem and the sensor fault subsystem to separate the actuator fault and the sensor fault.

[0032] The actuator failure subsystem includes:

[0033]

[0034] The sensor failure subsystem includes:

[0035]

[0036] Where L is the transformation matrix; f a 、f s Indicates actuator failure and sensor failure respectively.

[0037] In some exemplary embodiments, the method further comprises:

[0038] Configure the sliding mode nonlinear switching term v i,1 With v i,2 ,include:

[0039]

[0040] Among them, α i,1 , α i,2 is a constant greater than 0, P1 and P3 are Lyapunov matrices to be solved, and are symmetric positive definite matrices. and are sliding mode gains respectively;

[0041] According to the adaptive law of the sliding mode gain, the upper bound constraints are relaxed, including:

[0042]

[0043] in, is the estimation error; ε1, ε2 are constants greater than 0;

[0044] Configure the sliding surface corresponding to the actuator fault observer, including:

[0045]

[0046] Configure the sliding surface corresponding to the sensor fault observer, including:

[0047]

[0048] Fault characteristic estimation is achieved using equivalent output error injection terms.

[0049] In some exemplary embodiments, configuring an adaptive gain sliding mode controller for the affine linear variable parameter fault model based on an adaptive upper limit sliding mode observer includes:

[0050] u(t)=u eq (t)+u sw (t)

[0051] Where u(t) is the adaptive gain sliding mode controller, u eq (y) is the control input item on the sliding surface, u sw (t) is the switching control item;

[0052] The control inputs on the sliding surface include disturbances estimated using an adaptive RBF network and estimation errors.

[0053] In some exemplary embodiments, the method further comprises:

[0054] In response to the fault signal, collecting current operating data of the aviation fuel adjustment pump test equipment;

[0055] The current operating data is processed using a sliding mode observer with an adaptive upper limit to obtain actuator fault estimation information, sensor fault estimation information, and estimation information corresponding to state variables;

[0056] The actuator fault estimation information, sensor fault estimation information, and estimation information corresponding to the state variables are input into the adaptive gain sliding mode controller, and the system control parameters are output to perform active fault-tolerant control.

[0057] According to a second aspect of the present invention, a computer program product is provided, on which a computer program is stored. When the computer program is executed by a processor, the active fault-tolerant control and optimization method of the aviation fuel conditioning pump test equipment is implemented.

[0058] According to a third aspect of the present invention, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the active fault-tolerant control and optimization method of the aviation fuel adjustment pump test equipment is implemented.

[0059] According to a fourth aspect of the present invention, there is provided an electronic device, comprising:

[0060] processor; and

[0061] a memory for storing executable instructions of the processor;

[0062] Wherein, the processor is configured to implement the above-mentioned active fault-tolerant control and optimization method of aviation fuel adjustment pump test equipment by executing the executable instructions.

[0063] The active fault-tolerant control and optimization method for an aviation fuel pump test equipment provided by an embodiment of the present invention establishes a linear parameter-varying (LPV) model for the aircraft fuel pump test equipment, taking into account sensor and actuator failures, and separates actuator failures from sensor failures. In response to actuator and sensor failures, the method considers the influence of possible interference and errors in the LPV system of the aircraft fuel pump test equipment and introduces an adaptive gain term into the sliding mode controller. This allows the adaptive sliding mode controller to dynamically adjust the controller structure online in response to faults and interferences, and to compensate for sudden faults and online interferences by switching nonlinear terms. This results in a fault-tolerant control effect superior to that based on robust theory suppression, and the output value can quickly track the expected value. The method considers the constraints of the adaptive gain control method and the conservatism of the fault-tolerant control, and compensates for unknown interference deviations to optimize the LPV active fault-tolerant effect and improve the robustness of the LPV active fault-tolerant control system.

[0064] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The accompanying drawings are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present invention, and together with the description, serve to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and it is clear that those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0066] Figure 1 A schematic diagram schematically illustrates an active fault-tolerant control and optimization method for aviation fuel adjustment pump test equipment according to an exemplary embodiment of the present invention;

[0067] Figure 2 A schematic diagram schematically illustrates the working principle of an adaptive sliding mode observer according to an exemplary embodiment of the present invention;

[0068] Figure 3 A schematic diagram schematically illustrates a radial basis function neural network structure according to an exemplary embodiment of the present invention;

[0069] Figure 4 A schematic diagram schematically illustrating an LPV model result according to an exemplary embodiment of the present invention;

[0070] Figure 5 A schematic diagram schematically illustrates a fault estimation result based on an adaptive sliding film observer according to an exemplary embodiment of the present invention;

[0071] Figure 6 A schematic diagram schematically illustrates an active fault-tolerant control result according to an exemplary embodiment of the present invention;

[0072] Figure 7 The figure schematically shows the composition of an electronic device in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0073] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0074] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the blocks shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0075] In the related art, the control method of the aviation fuel regulating pump test equipment has certain defects, such as: (1) the aviation fuel regulating pump test equipment has complex working conditions, and the parameters of the system model under multiple working conditions will change with the scheduling parameters at all times. A single fixed sliding surface is difficult to ensure the effect of global fault-tolerant control; (2) this method does not take sensor failure into consideration, so it is necessary to design a suitable sliding surface to achieve accurate expected tracking of the system when sensor and actuator failures occur, to ensure the safety and stability of the equipment; (3) the aviation fuel regulating pump test equipment may be in the case of complex interference and fault shock, and the use of robust theory to suppress observer errors and system interference has certain limitations.

[0076] In view of the shortcomings and deficiencies of the prior art, this exemplary embodiment provides an active fault-tolerant control and optimization method for aviation fuel conditioning pump test equipment.

[0077] In this example implementation, reference Figure 1 As shown, the active fault-tolerant control and optimization method for aviation fuel adjustment pump test equipment may specifically include the following steps:

[0078] Step S11, constructing a linear variable parameter model of the aircraft fuel conditioning pump test equipment using operating data of the aircraft fuel conditioning pump test equipment, for describing the dynamic characteristics of the aircraft fuel conditioning pump test equipment under different operating conditions; wherein the operating data includes system speed, outlet pressure, and flow rate corresponding to different fuel metering valve openings;

[0079] Step S12: constructing an affine linear variable parameter fault model based on the linear variable parameter model and combining the actuator fault, sensor fault, and external interference characteristics; wherein the affine linear variable parameter fault model includes: an actuator fault subsystem for expressing the actuator fault and a sensor fault subsystem for expressing the sensor fault, so as to separate the actuator fault from the sensor fault;

[0080] Step S13: configuring an adaptive upper limit sliding mode observer based on the adaptive observer gain term, the first sliding mode surface corresponding to the actuator fault observer, and the second sliding mode surface corresponding to the sensor fault observer; wherein the adaptive upper limit sliding mode observer includes: an actuator fault observer and a sensor fault observer, each of which is configured with an adaptive observer gain term;

[0081] Step S14 : configuring an adaptive gain sliding mode controller for the affine linear variable parameter fault model based on an adaptive upper limit sliding mode observer, so as to be used for active fault-tolerant control of the aviation fuel conditioning pump test equipment.

[0082] The method provided by the present disclosure establishes a linear variable parameter LPV fault model for aviation fuel pump test equipment, and designs a parameter-adaptive sliding mode observer on this basis to estimate the actuator and sensor fault signals in real time. Combining the inherent robustness of variable structure control with the flexible adjustment capability of adaptive control, an adaptive sliding mode fault-tolerant control method is designed to improve the robustness, accuracy and response speed of fault-tolerant control. Based on the advantages of the adaptive RBF network such as strong nonlinear fitting ability and fast convergence speed, unknown interference is estimated, and the sliding mode control algorithm is combined to generate a dynamic compensation signal to reconstruct the control strategy. This method breaks through the limitation of traditional fault-tolerant control's insufficient adaptability to time-varying working conditions, and can adjust the observer gain and control law in real time according to changes in system parameters, thereby improving the accuracy of active fault-tolerant control, effectively suppressing the risk of system instability caused by faults, and ensuring the reliable operation of the fuel pump test equipment.

[0083] Hereinafter, each step of the active fault-tolerant control and optimization method of the aviation fuel adjustment pump test equipment in this exemplary embodiment will be described in more detail with reference to the accompanying drawings and embodiments.

[0084] In step S11, a linear variable parameter model of the aviation fuel regulating pump test equipment is constructed using the operating data of the aviation fuel regulating pump test equipment to describe the dynamic characteristics of the aviation fuel regulating pump test equipment under different working conditions; wherein the operating data includes the system speed, outlet pressure and flow corresponding to different fuel metering valve openings.

[0085] Exemplarily, the above step S11 may specifically include:

[0086] Step S21: For each fuel metering valve position between the fully closed state and the fully open state, N fuel metering valve opening values ​​are selected at equal intervals, and corresponding system speed, outlet pressure, and flow rate data are collected at a preset sampling frequency to obtain multiple sets of operating data.

[0087] Step S22, configure the fuel metering valve opening FMV as the input variable u(t); configure the system speed n, outlet pressure P outand flow rate W as output variable y(t); define system speed n as state variable x(t), and configure it as scheduling parameter;

[0088] Step S23: identify each set of operating data, fit the model parameters, and obtain N sets of linear state space models S j ={A j ,B j ,C};

[0089] In step S24 , a coefficient matrix is ​​obtained for the N groups of linear state space models and the corresponding scheduling parameters, and a linear variable parameter model of the aviation fuel conditioning pump test equipment is constructed based on the coefficient matrix to describe the dynamic characteristics of the system under different working conditions.

[0090] Specifically, N fuel metering valve opening values ​​FMV are selected at equal intervals (e.g., 2.5%) from the fully closed state (0% opening) to the fully open state (100% opening) of the fuel metering valve. j ,j=1,2,…,N. Different fuel metering valve opening values ​​correspond to different fuel flow rates. Data is collected at a certain sampling frequency to obtain the operating data of the aviation fuel adjustment pump test equipment under different fuel flow rates, with a total of N groups of data. Each group of operating data includes the fuel metering valve opening FMV j , system speed n j , outlet pressure P out,j and flow rate W j .

[0091] For N groups of aviation fuel pump test equipment operating data, the fuel metering valve opening FMV is selected as the input variable u(t), the system speed n, the outlet pressure P out The flow rate W is the output variable y(t), and the system speed n is the state variable x(t). Since the operating conditions of the aviation fuel pump test equipment (such as basic operating conditions, acceleration conditions, deceleration conditions, and idle conditions, etc.) are closely related to the system speed n, the system speed n is configured as the scheduling parameter ρ. For each set of data, the least squares method is used to identify the data and fit the model parameters to obtain a set of linear state space models S j ={A j ,B j ,C}, further expressed as:

[0092]

[0093] Among them, x(t) is the state variable, u(t) is the input variable, and y(t) is the output variable; A j ,B j, C are the system's state transfer matrix, control input matrix, and observation matrix, respectively, and are constant matrices of a certain dimension. Ultimately, we obtain N sets of linear state-space models, which are linear time-invariant systems that can be used to describe the local dynamic characteristics of aviation fuel conditioning pump test equipment.

[0094] Based on N sets of known linear state space models and corresponding known scheduling variables, we use interpolation fitting to solve the coefficient matrix of the linear equation system and construct an affine linear variable parameter model of the aviation fuel pump test equipment as shown in the following formula to better describe the dynamic characteristics of the system under different operating conditions. The model is expressed as:

[0095]

[0096] Where ρ(t) is a scheduling parameter that can be measured at any time and is within a bounded set of admissible parameter trajectories. The matrices A(ρ(t)) and B(ρ(t)) are related to the scheduling variable ρ(t) and have the form of affine polynomials, i.e., A(ρ(t)) = A0 + ρ(t)A1 + ρ(t)A2, B(ρ(t)) = B0 + ρ(t)B1 + ρ(t)B2. The coefficient matrices A0, A1, A2, B0, B1, B2 of the affine polynomials can be obtained based on the N sets of known linear state space models S = {S1, S2, …, S N} and the corresponding known scheduling variables ρ(t)={n1,n2,…,n N}, obtained by solving the following linear equations:

[0097]

[0098] In step S12, an affine linear variable parameter fault model is constructed based on the linear variable parameter model and combined with the actuator fault, sensor fault and external interference characteristics; wherein the affine linear variable parameter fault model includes: an actuator fault subsystem for expressing the actuator fault, and a sensor fault subsystem for expressing the sensor fault, so as to separate the actuator fault and the sensor fault.

[0099] Exemplarily, the above step S12 may specifically include:

[0100] Step S31, defining bounded functions corresponding to actuator faults, sensor faults, and system uncertainties, and transforming the linear variable parameter model based on a preconfigured transformation matrix to obtain an affine linear variable parameter fault model;

[0101] In step S32 , the coupling between different types of faults is considered, and the affine linear variable parameter fault model is split to obtain an actuator fault subsystem and a sensor fault subsystem, so as to separate the actuator fault and the sensor fault.

[0102] Specifically, based on the established affine linear variable parameter model of the aviation fuel pump test equipment, an affine linear variable parameter fault model with actuator fault, sensor fault and external interference is established, which can be expressed as:

[0103]

[0104] Among them, f a and f s represent actuator fault and sensor fault respectively; M(ρ(t))=B(ρ(t)), N=I; D and ε(t) are the constant interference distribution matrix and interference variable respectively.

[0105] Actuator failure, sensor failure and system uncertainty are all bounded functions, i.e. ||f a ||≤λ a ,||f s ||≤λ s ,||ε(t)||≤λ ε ; Among them, λ a ,λ s ,λ ε is an unknown constant greater than zero.

[0106] Assuming the system (A(ρ(t)), B(ρ(t))) is controllable, when both actuator and sensor faults exist, accurate fault estimation is difficult due to the coupling between different types of faults in the system. To address this, the affine linear variable parameter fault model can be split into two subsystems: the actuator fault subsystem and the sensor fault subsystem, thereby separating the actuator and sensor faults.

[0107] Design transformation matrix T i =[T i,1 ,T i,2 ] T , S i =[S i,1 ,S i,2 ] T and L, perform coordinate transformation on the affine linear variable parameter fault model to obtain the new state variable and output variables

[0108]

[0109] Get the new system matrix And the matrix has the following form:

[0110]

[0111]

[0112] Based on the above, the affine linear variable parameter fault model is converted to:

[0113]

[0114] in, are the transformed system matrices, is the state variable after conversion, is the output variable after conversion; the conversion matrix T i =[T i,1 ,T i,2 ] T , the transformation matrix S i =[S i,1 ,S i,2 ] T ; u is the system input variable, namely the fuel metering valve opening FMV.

[0115] The affine linear variable parameter fault model is divided into subsystem 1 affected by the actuator fault. This subsystem is only affected by the actuator fault. The actuator fault subsystem includes:

[0116]

[0117] The affine linear variable parameter fault model is divided into sensor fault subsystem 2, which is only affected by the sensor fault. The sensor fault subsystem includes:

[0118]

[0119] Where L is the transformation matrix; f a 、f s Indicates actuator failure and sensor failure respectively.

[0120] Define the following new state variables:

[0121]

[0122] in,

[0123] The system can be further expressed as:

[0124]

[0125] In step S13, an adaptive upper limit sliding mode observer is configured based on the adaptive observer gain term, the first sliding mode surface corresponding to the actuator fault observer, and the second sliding mode surface corresponding to the sensor fault observer; wherein the adaptive upper limit sliding mode observer includes: an actuator fault observer and a sensor fault observer, each of which is configured with an adaptive observer gain term.

[0126] Exemplarily, the method further includes: configuring a target Lyapunov function for the sliding mode observer with an adaptive upper limit, so as to eliminate the influence of an unknown upper bound of a fault on the convergence of the system state.

[0127] Specifically, refer to Figure 2 As shown in the figure, when designing an adaptive sliding mode observer, the traditional sliding mode observer design method has certain limitations in application because it requires knowing the upper bound information of the fault. The sliding mode observer with adaptive upper bound relaxes the upper bound condition constraints through the adaptive law and constructs a suitable Lyapunov function to eliminate the influence of the unknown upper bound of the fault on the convergence of the system state.

[0128] Specifically,

[0129] 1) Design two sliding mode observers with the following structures, representing the actuator fault observer and the sensor fault observer respectively:

[0130]

[0131]

[0132] The superscript ^ indicates an estimated value. is the stable matrix to be designed, K i is the observer gain to be designed, v i,1 、v i,2 is the sliding mode nonlinear switching term to be designed.

[0133] 2) Design the sliding mode nonlinear switching term v i,1 With v i,2 :

[0134]

[0135] Among them, α i,1 , α i,2 is a constant greater than 0, P1 and P3 are Lyapunov matrices to be solved, and are symmetric positive definite matrices. and are the designed sliding mode gains respectively.

[0136] 3) Design and The adaptive law is used to relax the upper bound constraint:

[0137]

[0138] in, is the estimation error. ε1,ε2 are constants greater than 0.

[0139] 4) Design the sliding surface as:

[0140]

[0141] Among them, s1 is the sliding surface designed by the actuator fault observer, and s2 is the sliding surface designed by the sensor fault observer.

[0142] 5) Solve the sliding mode observer to be designed matrix and K i If there exists a positive symmetric matrix P and a full rank matrix H with the following form:

[0143]

[0144]

[0145] Furthermore, the following linear matrix inequality (LMI) constraints are satisfied, including:

[0146]

[0147] in, Then the system state is asymptotically stable and robust to disturbances.

[0148] 6) Using the equivalent output error injection term to realize fault characteristic estimation, the actuator fault f a and sensor failure f s Estimated value of and It can be calculated as:

[0149]

[0150] where δ1 and δ2 are small positive scalars.

[0151] In addition, the observer can output estimated information

[0152] according to The estimated value of the state variable x can be obtained

[0153] Exemplarily, the method further includes: using an adaptive RBF network to provide interference and estimation error for the adaptive gain sliding mode controller.

[0154] Specifically, the fault estimation information obtained based on the LPV adaptive sliding mode observer and Estimated information of state variable x Adaptive sliding mode control technology and adaptive RBF network are used to design the LPV active fault-tolerant controller. The adaptive gain change law of the sliding mode controller and the adaptive update law of the weight matrix of the RBF network are designed to realize active fault-tolerant control and optimization in the event of failure of the aviation fuel pump test equipment.

[0155] 1) For affine linear variable parameter fault model:

[0156]

[0157] Design an adaptive gain sliding mode controller u(t):

[0158] u(t)=u eq (t)+u sw (t)

[0159] Among them, u eq (t) Control input item on the sliding surface (equivalent control input item), u sw (t) is the switching control item.

[0160] They have the following forms:

[0161]

[0162] Where E = xx d is the state error, x d For the expected state, For B i The pseudo-inverse matrix of . are the estimated variables of interference and estimation error d (estimated by the adaptive RBF network); Li,2=[Bi+MiBi+D], k1,j is a positive definite symmetric matrix.

[0163]

[0164] Among them, P5 and K are symmetric positive definite matrices, S = KE is the designed sliding surface, and φ>0, in is the designed adaptive law, satisfying: Where σ is a positive scalar.

[0165] 2) Solve the sliding mode controller to be designed matrix P5, K and k 1,j , by solving the following LMI:

[0166]

[0167] in,

[0168]

[0169] So that the parameters η, φ, If the conditions are met, the sliding mode LPV fault-tolerant control system is stable and can reach the sliding surface and stabilize on the sliding surface within a certain time.

[0170] Based on the above, u(t) can be expressed as:

[0171]

[0172] Exemplary, reference Figure 3 As shown in FIG, the adaptive RBF network includes an input layer, a hidden layer, and an output layer, which are arranged in sequence; wherein the input parameter of the input layer is the system state variable x, which is a g-dimensional variable; the hidden layer includes l neurons, and the activation function of each neuron has the form of a Gaussian basis function; the output layer is used to output the interference estimation value.

[0173] Specifically, input x RBF =[x RBF,1 ,x RBF,2 ,…,x RBF,i …,x RBF,g ], i=1,2,…,g is a g-dimensional variable. The activation function of each neuron has the form of Gaussian basis function g=[g1,g2,…,g i …,h l ].in,

[0174]

[0175] Among them, c i is the center vector of the i-th neuron, b i is the width of the basis function, x RBF is the input variable of the adaptive RBF network, and is the state variable x of the system.

[0176] The output variable is the disturbance estimate

[0177]

[0178] Among them, W is the weight matrix, and its update rate is designed as:

[0179]

[0180] in, D is the interference distribution matrix; γ is a positive scalar; H represents the Gaussian function.

[0181] refer to Figure 4As shown in the figure, by using an adaptive RBF network to design an LPV active fault-tolerant controller, the advantages of the RBF network such as strong nonlinear fitting ability and fast convergence speed are used to estimate unknown disturbances, and the sliding mode control algorithm is combined to generate a dynamic compensation signal to reconstruct the control strategy. This method breaks through the limitation of traditional fault-tolerant control that is not adaptable to time-varying working conditions, and can adjust the observer gain and control law in real time according to changes in system parameters, thereby improving the accuracy of active fault-tolerant control and effectively suppressing the risk of system instability caused by faults.

[0182] Exemplarily, the method further includes:

[0183] Step S41, in response to the fault signal, collecting current operating data of the aviation fuel adjustment pump test equipment;

[0184] Step S42: Processing the current operating data using a sliding mode observer with an adaptive upper limit to obtain actuator fault estimation information, sensor fault estimation information, and estimation information corresponding to the state variables;

[0185] In step S43, the actuator fault estimation information, the sensor fault estimation information, and the estimation information corresponding to the state variables are input into the adaptive gain sliding mode controller, and the system control parameters are output to perform active fault-tolerant control.

[0186] Specifically, the aviation fuel conditioning pump test equipment can collect real-time operating data during operation. When a system fault is detected, active fault-tolerant control calculations can be performed based on the operating data at that moment. The current fuel metering valve opening, system speed, outlet pressure, and flow rate are input into an adaptive upper limit sliding mode observer. The input parameters are processed using the actuator fault observer and sensor fault observer, respectively, to obtain actuator fault estimation information, sensor fault estimation information, and corresponding state variable estimation information. The estimated data is then processed using an adaptive gain sliding mode controller to obtain system control parameters, implementing active fault-tolerant control.

[0187] For example, a pressure differential valve stuck fault with an amplitude of 0.5 occurs in the system at 10 seconds. When the speed is 0.8, the control parameter K = -0.8265; when the speed is 0.9, the control parameter K = -0.1533, σ = 0.008, φ = 0.002, γ1=20. Among them, σ, φ, are predefined constants.

[0188] At this time, the system output response overshoot is less than 1.8%, and the adjustment time is less than 1.5s. The adaptive sliding mode controller based on the adaptive RBF network retains the rapidity of the sliding mode control response in the dynamic response stage. Figure 5 、 Figure 6As shown in the figure, due to the superior disturbance fitting ability of the adaptive RBF network, the output of the aircraft engine system is restored to the control expectation value within 1.8 seconds after the actuator constant deviation hard fault occurs and within 0.08 seconds after the sensor constant deviation hard fault occurs. Therefore, the designed LPV active controller has good performance.

[0189] The method provided by the embodiments of the present invention obtains fault estimation information by designing a linear parameter-varying adaptive sliding mode observer. It then uses adaptive sliding mode control technology and an adaptive RBF neural network to design an LPV active fault-tolerant controller. First, a linear parameter-varying (LPV) model for an aircraft fuel pump test rig is established, taking into account sensor and actuator failures. Adaptive gain terms are introduced into the sliding mode controller to address actuator and sensor failures, taking into account the impact of potential disturbances and errors in the LPV system. An adaptive gain algorithm is designed to enable the adaptive sliding mode controller to dynamically adjust the controller structure online after encountering faults and disturbances. Nonlinear term switching is used to compensate for sudden faults and online disturbances, resulting in superior fault-tolerant control performance compared to robust theory-based suppression methods, enabling the output value to quickly track the expected value. Considering the constraints of the adaptive gain control method and the conservative nature of fault-tolerant control, an adaptive RBF network is introduced to compensate for unknown disturbance deviations. An adaptive update law for the adaptive RBF is designed to optimize the LPV active fault-tolerant performance and enhance the robustness of the LPV active fault-tolerant control system.

[0190] The beneficial effects of the present invention include:

[0191] (1) Compared with the existing fault-tolerant control method based on fixed sliding surface, the present invention adopts parameter-scheduled sliding mode observer and combines the linear variable parameter system model to construct a dynamic sliding surface, so that the switching gain of the fault observer can be adjusted in real time according to the speed operating parameters of the fuel control pump.

[0192] (2) In order to solve the problem that the prior art ignores sensor failure, the present invention decomposes and decouples sensor failure and actuator failure through subsystem decomposition.

[0193] (3) Due to the adoption of error feedback compensation strategy, the present invention can suppress sliding mode chattering while ensuring the dynamics of the system after the fault.

[0194] (4) Since the introduction of adaptive radial basis function neural network compensation makes up for the error estimated by the observer and returned to the controller, the sliding mode controller based on adaptive radial basis function neural network proposed in the present invention improves the fault suppression capability of the system fault-tolerant controller.

[0195] It should be noted that the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention and are not intended to be limiting. It is readily understood that the processes illustrated in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0196] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0197] Figure 7 A schematic diagram of an electronic device suitable for implementing an embodiment of the present invention is shown.

[0198] It should be noted that Figure 7 The electronic device 1000 shown is merely an example and should not limit the functionality or scope of use of the embodiments of the present invention. The electronic device can be a data processing device mounted on an aircraft fuel conditioning pump test device. Alternatively, it can be a host computer connected to the aircraft fuel conditioning pump test device for controlling and processing operational data.

[0199] like Figure 7 As shown, electronic device 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in read-only memory (ROM) 1002 or the program loaded from storage portion 1008 into random access memory (RAM) 1003. Various programs and data required for system operation are also stored in RAM 1003. CPU 1001, ROM 1002 and RAM 1003 are connected to each other via bus 1004. Input / output (I / O) interface 1005 is also connected to bus 1004.

[0200] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 1008 including a hard disk and the like; and a communication section 1009 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. Removable media 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 1010 as needed, so that computer programs read therefrom can be installed into the storage section 1008 as needed.

[0201] In particular, according to an embodiment of the present invention, the process described below with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product that includes a computer program carried on a storage medium, the computer program containing program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1009 and / or installed from a removable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the various functions defined in the system of the present application are performed.

[0202] It should be noted that the storage medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any storage medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0203] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0204] The units involved in the embodiments of the present invention may be implemented in software or hardware, and the units described may also be provided in a processor. In some cases, the names of these units do not limit the units themselves.

[0205] It should be noted that, as another aspect, the present application also provides a storage medium, which can be included in an electronic device; or it can exist independently without being installed in the electronic device. The above storage medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device can implement the following Figure 1 The individual steps of the method are shown.

[0206] In one embodiment, the present application provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0207] Furthermore, the above-described figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention and are not intended to be limiting. It is readily understood that the processes illustrated in the above-described figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0208] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the claims.

[0209] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof, which is limited only by the appended claims.

Claims

1. An active fault-tolerant control and optimization method for aviation fuel conditioning pump test equipment, characterized in that: The method comprises: Using operating data from an aviation fuel conditioning pump test device, a linear variable parameter model of the aircraft fuel conditioning pump test device is constructed to describe the dynamic characteristics of the aircraft fuel conditioning pump test device under different operating conditions. The operating data includes the system speed, outlet pressure, and flow rate corresponding to different fuel metering valve openings. Based on the linear variable parameter model, an affine linear variable parameter fault model is constructed by combining the actuator fault, sensor fault, and external interference characteristics. The affine linear variable parameter fault model includes an actuator fault subsystem for expressing the actuator fault and a sensor fault subsystem for expressing the sensor fault, thereby separating the actuator fault from the sensor fault. Based on the adaptive observer gain term, the first sliding mode surface corresponding to the actuator fault observer, and the second sliding mode surface corresponding to the sensor fault observer, an adaptive upper limit sliding mode observer is configured; wherein the adaptive upper limit sliding mode observer includes: the actuator fault observer and the sensor fault observer, each of which is configured with an adaptive observer gain term; An adaptive upper limit-based sliding mode observer configures an adaptive gain sliding mode controller for the affine linear variable parameter fault model, so as to be used for active fault-tolerant control of an aviation fuel adjustment pump test device.

2. The method according to claim 1, characterized in that The method further comprises: The target Lyapunov function is configured for the sliding mode observer with adaptive upper bound to eliminate the influence of unknown upper bound of fault on the convergence of system state.

3. The method according to claim 1, characterized in that The method further comprises: An adaptive RBF network is used to provide disturbance and estimation error for an adaptive gain sliding mode controller.

4. The method according to claim 3, characterized in that The adaptive RBF network includes an input layer, a hidden layer, and an output layer, which are arranged in sequence; the input parameter of the input layer is the system state variable x, which is a g-dimensional variable; the hidden layer includes l neurons, and the activation function of each neuron has the form of a Gaussian basis function; the output layer is used to output the interference estimation value.

5. The method according to claim 1, wherein The method of constructing a linear variable parameter model of the aviation fuel regulating pump test equipment using the operating data of the aviation fuel regulating pump test equipment includes: For the fuel metering valve between the fully closed state and the fully open state, N fuel metering valve opening values ​​are selected at equal intervals, and the corresponding system speed, outlet pressure and flow rate data are collected at a preset sampling frequency to obtain multiple sets of operating data; Configure the fuel metering valve opening FMV as the input variable u(t); configure the system speed n, outlet pressure P out and flow rate W as output variable y(t); define system speed n as state variable x(t), and configure it as scheduling parameter; Identify each set of operating data and fit the model parameters to obtain N sets of linear state space models S j ={A j ,B j ,C}, expressed as: Among them, x(t) is the state variable, u(t) is the input variable, and y(t) is the output variable; A j ,B j ,C are the state transfer matrix, control input matrix and observation matrix of the system respectively, which is a constant matrix; The coefficient matrix is ​​solved for N groups of linear state space models and the corresponding scheduling parameters. Based on the coefficient matrix, a linear variable parameter model of the aviation fuel pump test equipment is constructed to describe the dynamic characteristics of the system under different working conditions. The model is expressed as: Among them, ρ(t) is the scheduling parameter; A9ρ(t)) and B9ρ(t)) are coefficient matrices respectively; the coefficient matrix can be solved based on the system speed and the multi-vertex scheduling matrix.

6. The method according to claim 1, characterized in that The affine linear variable parameter fault model is constructed based on the linear variable parameter model and combined with the actuator fault, sensor fault and external interference characteristics, including: Define bounded functions corresponding to actuator faults, sensor faults, and system uncertainty, and transform the linear variable parameter model based on the preconfigured transformation matrix to obtain an affine linear variable parameter fault model: in, are the transformed system matrices, is the state variable after conversion, is the output variable after conversion; the conversion matrix T i =[T i,1 ,T i,2 ] T , the transformation matrix S i =[S i,1 ,S i,2 ] T ; Considering the coupling between different types of faults, the affine linear variable parameter fault model is split into the actuator fault subsystem and the sensor fault subsystem to separate the actuator fault and the sensor fault. The actuator failure subsystem includes: The sensor failure subsystem includes: Where L is the transformation matrix; f a 、f s Indicates actuator failure and sensor failure respectively.

7. The method according to claim 6, characterized in that The method further comprises: Configure the sliding mode nonlinear switching term v i,1 With v i,2 ,include: Among them, α i,1 , α i,2 is a constant greater than 0, P1 and P3 are Lyapunov matrices to be solved, and are symmetric positive definite matrices. and are sliding mode gains respectively; According to the adaptive law of the sliding mode gain, the upper bound constraints are relaxed, including: in, is the estimation error; ε1, ε2 are constants greater than 0; Configure the sliding surface corresponding to the actuator fault observer, including: Configure the sliding surface corresponding to the sensor fault observer, including: Fault characteristic estimation is achieved using equivalent output error injection terms.

8. The method according to claim 1, characterized in that An adaptive upper limit-based sliding mode observer is used to configure an adaptive gain sliding mode controller for the affine linear variable parameter fault model, including: u(t)=u eq (t)+u sw (t) Where u(t) is the adaptive gain sliding mode controller, u eq (t) is the control input on the sliding surface, u sw (t) is the switching control item; The control inputs on the sliding surface include disturbances estimated using an adaptive RBF network and estimation errors.

9. The method according to claim 1, characterized in that The method further comprises: In response to the fault signal, collecting current operating data of the aviation fuel adjustment pump test equipment; The current operating data is processed using a sliding mode observer with an adaptive upper limit to obtain actuator fault estimation information, sensor fault estimation information, and estimation information corresponding to state variables; The actuator fault estimation information, sensor fault estimation information, and estimation information corresponding to the state variables are input into the adaptive gain sliding mode controller, and the system control parameters are output to perform active fault-tolerant control.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the active fault-tolerant control and optimization method for aviation fuel adjustment pump test equipment according to any one of claims 1 to 8 is implemented.

Images