Model prediction control method and device based on disturbance prediction, equipment and medium

Abstract

The invention relates to a model prediction control method and device based on disturbance prediction, equipment and a medium, and relates to the technical field of system control, and the method comprises the steps: obtaining the disturbance data and actual state quantities (height, speed and attitude angle) of an aircraft at the current moment in real time, determining a disturbance prediction sequence based on the current and historical disturbance data, and obtaining a disturbance prediction result; and constructing a discrete system state prediction model, performing rolling optimization solving by taking minimization of a preset cost function as a target to obtain an optimal control sequence in a future prediction time domain, and outputting a first control quantity in the optimal control sequence as an actual control instruction at the current moment to an execution mechanism. According to the invention, automatic control of a multivariable complex system is realized, high precision and high real-time performance of control are ensured, and the ability of the system to deal with extreme conditions is improved.

Description

Technical Field

[0001] This invention relates to the technical field of system control, and in particular to a model predictive control method, apparatus, device and medium based on disturbance prediction. Background Technology

[0002] Automatic control technology has been widely and deeply applied in many fields such as modern industry, transportation, and aerospace. In industrial production, automatic control technology can automate production processes, improving production efficiency and product quality; in the transportation sector, it helps to realize intelligent transportation systems and optimize traffic flow; and in aerospace, it is crucial for ensuring the safe flight and precise control of aircraft. Especially in fields such as autonomous driving and aircraft takeoff and landing control, the controlled systems are characterized by numerous variables and complex modeling, placing higher demands on the accuracy and real-time performance of automatic control technology. For example, during the leveling-off phase of a civil aircraft's approach and landing, the aircraft is at low altitude and speed and faces complex flight environment disturbances such as crosswinds, which can easily lead to safety accidents, thus requiring extremely high control precision.

[0003] Currently, the most widely used automatic controller in industry is PID control. The target value and feedback value are calculated once by the PID controller to obtain the output value, which is then directly used as the control variable. In aircraft flight control, cascaded PID control is generally used, employing a hierarchical cascaded structure. The outer loop controls the aircraft's position and trajectory, while the inner loop controls its attitude and speed. This cascaded PID control can achieve variable control across different orders, while also decoupling the controller, facilitating separate debugging of the inner and outer loops. Due to its simple structure and low implementation cost, PID control remains the preferred technology for many low-order systems such as temperature regulation, pressure maintenance, and UAV attitude stabilization. In more complex control scenarios such as aircraft takeoff and landing control, Model Predictive Control (MPC) is gradually becoming the mainstream control technology. MPC control can control multivariable coupled, time-varying, and nonlinear systems, making it more suitable than traditional PID control in complex and highly dynamic scenarios.

[0004] However, when traditional PID controllers are applied to complex scenarios such as the approach and landing of commercial aircraft, they require separate controllers for each of the multiple coupled variables, making it difficult to address the coupling effects within the system. They also cannot construct a comprehensive model that considers the relationships between all variables, and are unable to handle unforeseen accidents and complex operating conditions in scenarios with complex environmental disturbances, compromising safety. Furthermore, PID controllers themselves cannot constrain variables, potentially leading to performance degradation or oscillations. Existing MPC control methods also have shortcomings; basic MPC controllers have weak anti-interference capabilities. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a model predictive control method, device, equipment and medium based on disturbance prediction, in order to solve at least one of the above-mentioned technical problems.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: Firstly, this application provides a model predictive control method based on disturbance prediction, employing the following technical solution: A model predictive control method based on disturbance prediction, comprising: The perturbation data and actual state quantities of the controlled object at the current moment are acquired in real time. The controlled object is an aircraft, and the actual state quantities include the altitude, speed and attitude angle of the controlled object. Based on the disturbance data of the controlled object at the current moment and historical disturbance data, a disturbance prediction sequence is determined, wherein the disturbance prediction sequence represents the disturbance amount at multiple future moments; Based on the preset time-varying system matrix sequence and the disturbance prediction sequence, a discrete system state prediction model of the controlled object is constructed. The discrete system state prediction model characterizes the relationship between the state variables of the controlled object and the control input. Based on the discrete system state prediction model, the actual state quantities at the current moment, and the constraints, rolling optimization is performed with the goal of minimizing the preset cost function to obtain the optimal control sequence in the future prediction time domain. The first control variable in the optimal control sequence is used as the actual control command at the current moment and output to the actuator of the controlled object to achieve rolling optimization control of the controlled object.

[0007] The beneficial effects of this invention are as follows: a predictive model applicable to complex multivariate systems such as the landing and leveling phase of an aircraft is established, which can more accurately fit the complex relationship between multiple coupled variables, thereby achieving more precise control; in addition, based on the predicted disturbance, the disturbance can be pre-compensated in advance, thereby improving the robustness, real-time performance and reliability of the control; by actively predicting and utilizing disturbances, a discrete time-varying model with disturbance input that is closer to the real scenario is established, making the prediction effect more reliable; and the MPC architecture fully considers various constraints, more effectively avoiding the risks brought about by extreme cases.

[0008] Based on the above technical solution, the present invention can be further improved as follows.

[0009] Furthermore, acquiring the disturbance data of the controlled object at the current moment includes: Real-time wind speed of the controlled object at the current moment; Based on the current wind speed and the preset disturbance model, determine the disturbance data of the controlled object at the current moment; The preset disturbance model is: ; Where the F matrix is ​​the time-varying (LPV) transformation matrix, Let be the wind speed, and e be the error.

[0010] The beneficial effects of adopting the above-mentioned further scheme are: real-time acquisition of the wind speed of the aircraft at the current moment and the disturbance model composed of the preset time-varying transformation matrix and error amount can more accurately determine the disturbance data of the aircraft at the current moment, providing a more reliable foundation for subsequent construction of discrete system state prediction model and determination of optimal control sequence, which helps to achieve more precise rolling optimization control of the aircraft.

[0011] Furthermore, determining the disturbance prediction sequence based on the disturbance data of the controlled object at the current moment and historical disturbance data includes: Based on the Gaussian process regression model, the current disturbance data, and historical disturbance data, the disturbance prediction sequence is determined.

[0012] The beneficial effects of adopting the above-mentioned further scheme are: using Gaussian processes to estimate future disturbance sequences and introducing them into MPC, a discrete time-varying system with accurate disturbance input is constructed, thereby improving the accuracy of control.

[0013] Furthermore, the step of performing rolling optimization based on the discrete system state prediction model, the actual state variables at the current moment, and the constraints, with the objective of minimizing a preset cost function, to obtain the optimal control sequence in the future prediction time domain includes: Based on the disturbance prediction sequence and the discrete system state prediction model, an MPC prediction model matrix is ​​constructed. The MPC prediction model matrix represents the prediction path of the system state under the initial state and the future control sequence within a prediction time domain. Based on the MPC prediction model matrix, the actual state variables at the current time, and the constraints, the optimal control sequence in the future prediction time domain is obtained by minimizing the cost function.

[0014] The beneficial effects of adopting the above-mentioned further solutions are: By integrating the future disturbance sequence predicted by Gaussian process regression into the prediction model of model predictive control in a feedforward manner, a DEPE-MPC controller with disturbance look-ahead capability is constructed, thereby realizing closed-loop management of disturbances through "estimation-prediction-active utilization". This method transforms the passive compensation of traditional MPC into active utilization, significantly improving the control accuracy and anti-interference capability of the system when facing time-varying and complex disturbances (such as wind disturbances during aircraft landing).

[0015] Furthermore, the cost function is a quadratic cost function, and the expression of the quadratic cost function is: ; in, For future control sequence vectors, C represents the actual state quantity at the current moment, and C and V are coefficient matrices calculated based on the MPC prediction model matrix and the preset weight matrix.

[0016] The beneficial effect of adopting the above further scheme is that by substituting the system model containing the predicted perturbation into the standard quadratic cost function and using matrix operations to accurately simplify it into... This standardized quadratic programming form achieves explicit analysis and significant simplification of the mathematical structure of optimization problems. This formal standardization allows complex constrained optimization problems to be directly computed online using efficient and mature quadratic programming solvers. This not only ensures global convergence and numerical stability of the solution process but also significantly improves the real-time performance of online rolling optimization, meeting the stringent real-time control requirements of high-speed dynamic processes such as aircraft landing. Simultaneously, the cost function organically integrates system dynamics, predicted disturbances, and performance weights through coefficient matrices C and V, enabling the optimal control law to simultaneously consider feedforward compensation for disturbances and feedback stabilization of the system state. This theoretically guarantees optimal control performance and significantly improves the computational efficiency and reliability of the controller in engineering applications.

[0017] Furthermore, the constraints include: ; ; ; Where G and H are constraint matrices, representing the coupling constraints between inputs and state variables in practical applications.

[0018] The beneficial effects of adopting the above-mentioned further solutions are: the systematic constraint processing mechanism prevents control commands from exceeding the capabilities of the actuators or causing system instability from the source, significantly enhances the practicality and reliability of the controller under complex working conditions, and provides key guarantees for achieving autonomous and precise control of high-safety-level tasks such as aircraft landing.

[0019] Furthermore, the step of solving the problem based on the MPC prediction model matrix, the actual state variables at the current time, and the constraints, with the objective of minimizing the cost function, to obtain the optimal control sequence in the future prediction time domain, includes: The optimization problem is constructed as a quadratic programming problem, and based on the MPC prediction model matrix, the actual state variables at the current time, and the constraints, the quadratic programming solver is called to solve it, so as to obtain the optimal control sequence in the future prediction time domain.

[0020] The beneficial effects of adopting the above-mentioned further scheme are: by constructing the optimization problem containing disturbance information into a standard quadratic programming form and using the terminal weight matrix to ensure stability, the real-time solution efficiency of the controller and the strong robustness of the closed-loop system are ensured. Finally, under the premise of strictly satisfying multiple state and input constraints, the safety, stability and reliability of the control system in complex scenarios (such as automatic landing of aircraft) are effectively improved.

[0021] Secondly, this application provides a model predictive control device based on disturbance prediction, which adopts the following technical solution: A model predictive control device based on disturbance prediction, comprising: The acquisition module is used to acquire the disturbance data and actual state quantities of the controlled object at the current moment in real time. The controlled object is an aircraft, and the actual state quantities include the altitude, speed and attitude angle of the controlled object. The disturbance prediction module is used to determine a disturbance prediction sequence based on the disturbance data of the controlled object at the current moment and historical disturbance data. The disturbance prediction sequence represents the disturbance amount at multiple future moments. The discrete system state prediction model construction module is used to construct a discrete system state prediction model of the controlled object based on a preset time-varying system matrix sequence and the disturbance prediction sequence. The discrete system state prediction model characterizes the relationship between the state variables of the controlled object and the control input. The control sequence prediction module is used to perform rolling optimization based on the discrete system state prediction model, the actual state quantity at the current time, and the constraints, with the goal of minimizing the preset cost function, to obtain the optimal control sequence in the future prediction time domain. The control module is used to take the first control quantity in the optimal control sequence as the actual control command at the current moment and output it to the actuator of the controlled object to realize the rolling optimization control of the controlled object.

[0022] Thirdly, this application provides an electronic device that adopts the following technical solution: An electronic device includes a memory and a processor, wherein the memory stores a computer program capable of being loaded by the processor and executing the model control method based on disturbance prediction as described in any of the first aspects.

[0023] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the model control method based on disturbance prediction as described in any of the first aspects.

[0024] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0025] Figure 1 A schematic flowchart of a model control method based on disturbance prediction provided in one embodiment of the present invention; Figure 2 A schematic diagram of the structure of a model control device based on disturbance prediction provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0028] This application provides a model predictive control method based on disturbance prediction. This method can be executed by an electronic device, which can be a server or a mobile terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The mobile terminal device can be a laptop computer, a desktop computer, etc., but is not limited to these.

[0029] like Figure 1 As shown, a model predictive control method based on disturbance prediction mainly includes: S1, real-time acquisition of disturbance data and actual state quantities of the controlled object at the current moment, wherein the controlled object is an aircraft, and the actual state quantities include the altitude, speed and attitude angle of the controlled object; In this embodiment of the application, taking the landing leveling phase control of a large aircraft as an example, the disturbance term is a variable dominated by wind disturbance. The step of obtaining the disturbance data of the controlled object at the current moment includes: Real-time wind speed of the controlled object at the current moment; Based on the current wind speed and the preset disturbance model, determine the disturbance data of the controlled object at the current moment; The preset disturbance model is: ; Where the F matrix is ​​the time-varying (LPV) transformation matrix, Let be the wind speed, and e be the error.

[0030] In this embodiment, the wind speed acquisition device can be an anemometer mounted on the aircraft, or other high-precision wind speed measurement devices such as laser Doppler anemometers can be used to replace traditional anemometers to improve the accuracy of wind speed measurement. Acquiring actual state quantities requires different sensors; the altitude sensor can be a barometric altimeter, the attitude angle sensor can be an inertial measurement unit (IMU), and the velocity sensor can be an airspeed tube.

[0031] S2, Based on the disturbance data of the controlled object at the current moment and the historical disturbance data, determine the disturbance prediction sequence, wherein the disturbance prediction sequence represents the disturbance amount at multiple future moments; In this embodiment of the application, determining the disturbance prediction sequence based on the disturbance data of the controlled object at the current moment and historical disturbance data includes: Based on the Gaussian process regression model, the current disturbance data, and historical disturbance data, the disturbance prediction sequence is determined.

[0032] wind speed Historical data is readily available. Based on this historical data, and using existing Gaussian process regression methods, the wind speed at time N in the future can be estimated, thus obtaining the predicted wind speed value at time N. Multiplying by the F matrix converts the wind speed into total disturbance, yielding the total disturbance at time N in the future. .

[0033] In addition to the Gaussian process regression model, other time series forecasting models such as the autoregressive integral moving average (ARIMA) model can be considered for perturbation forecasting. These models may have better forecasting performance in different scenarios.

[0034] S3. Based on the preset time-varying system matrix sequence and the disturbance prediction sequence, construct a discrete system state prediction model for the controlled object. The discrete system state prediction model characterizes the relationship between the state variables of the controlled object and the control input. In this embodiment of the application, a linear discrete time-varying model with a perturbation input is as follows: ; ; in, Let n be the n-dimensional state variables and m-dimensional control inputs at time k, respectively. Let k+1 be the state variable. , These are the state matrix and input matrix at time k, respectively. , It is the perturbation input at time k. These are two smaller quantities, representing It is a continuous and bounded input variable.

[0035] The state matrix and input matrix are time-varying system matrix sequences, which are pre-determined based on the aircraft's dynamic characteristics and flight environment. These time-varying system matrix sequences reflect the changing relationship between the aircraft's state variables and control inputs over time. By combining the disturbance prediction sequence with the pre-defined time-varying system matrix sequence, a discrete system state prediction model that accurately describes the aircraft's state changes can be constructed. In practical applications, the time-varying system matrix sequence can be adjusted and optimized according to different flight missions and environmental conditions to improve the model's accuracy.

[0036] S4. Based on the discrete system state prediction model, the actual state quantity at the current moment, and the constraints, a rolling optimization solution is performed with the goal of minimizing the preset cost function to obtain the optimal control sequence in the future prediction time domain. In this embodiment of the application, the step of performing rolling optimization based on the discrete system state prediction model, the actual state variables at the current moment, and the constraints, with the objective of minimizing a preset cost function, to obtain the optimal control sequence in the future prediction time domain includes: Based on the disturbance prediction sequence and the discrete system state prediction model, an MPC prediction model matrix is ​​constructed. The MPC prediction model matrix represents the prediction path of the system state under the initial state and the future control sequence within a prediction time domain. Based on the MPC prediction model matrix, the actual state variables at the current time, and the constraints, the optimal control sequence in the future prediction time domain is obtained by minimizing the cost function.

[0037] In this embodiment, the optimization problem is constructed as a quadratic programming problem, and based on the MPC prediction model matrix, the actual state variables at the current time, and the constraints, a quadratic programming solver is called to solve the problem, thereby obtaining the optimal control sequence in the future prediction time domain.

[0038] Wherein, the cost function is a quadratic cost function, and the expression of the quadratic cost function is: ; in, For future control sequence vectors, C represents the actual state quantity at the current moment, and C and V are coefficient matrices calculated based on the MPC prediction model matrix and the preset weight matrix.

[0039] The constraints include: ; ; ; Where G and H are constraint matrices, representing the coupling constraints between inputs and state variables in practical applications.

[0040] By acquiring relevant aircraft data in real time and fully considering disturbances during flight, advanced predictive models are used to accurately predict future disturbances. Based on this, a discrete system state prediction model and an MPC prediction model matrix are constructed. The optimal control sequence is solved through rolling optimization, and the first control variable is output as the actual control command. This effectively handles disturbances, accurately predicts and controls the future state of the aircraft, improves the accuracy and stability of aircraft control, overcomes the shortcomings of traditional control methods in adapting to complex flight environments, and provides strong support for the safe and efficient flight of aircraft.

[0041] In the above implementation, the predicted disturbance sequence is incorporated into the MPC prediction model, and the disturbance estimation is integrated into the rolling time-domain optimization to derive the desired control law.

[0042] The cost function is: ; Where Q and R are weight matrices, and The weight matrix is ​​determined by observing the actual performance and adjusting the weights accordingly. N is the prediction step size. The P matrix satisfies the discrete Lyapunov equation: ; Where A and B matrices are the state matrix and input matrix of the system at the corresponding time, K is a proposed feasible solution for state feedback, and P matrix can be obtained by solving this equation. P, Q and R are all positive definite matrices.

[0043] Consider system constraints: , , ; Where G and H are constraint matrices, representing the coupling constraints between inputs and state variables in practical applications.

[0044] The cost function considering constraints is solved, transforming it into a quadratic programming form: ; in: ; ; ; ; ; in, This term is determined by the initial state and is not affected by the control variables. Therefore, it can be ignored during optimization calculations to obtain a new performance index: ; definition: ; ; The optimization function can be simplified to: .

[0045] The equality constraints are then introduced in the form of the Lagrange multiplier method, and the inequality constraints are solved by software. By differentiating this matrix equation with respect to U and setting the derivative to zero, the optimal solution of the function can be obtained. ; At time k, extract As the output value, this process is repeated at each subsequent time step to achieve a rolling prediction process.

[0046] Predictive models in optimization control problems take into account disturbance information and introduce a sequence of predicted disturbance values ​​at N time points. This makes it possible to fully utilize disturbances to improve control performance without violating constraints.

[0047] S5, the first control quantity in the optimal control sequence is used as the actual control command at the current moment and output to the actuator of the controlled object to realize the rolling optimization control of the controlled object.

[0048] In this embodiment, the actuator can be the aircraft's servo motor, engine, etc. The actuator adjusts the aircraft's attitude and power according to the received control commands to achieve optimized roll control of the aircraft.

[0049] This method establishes a predictive model applicable to complex multivariate systems such as the landing and leveling phase of aircraft. It can more accurately fit the complex relationships between multiple coupled variables, thereby achieving more precise control. In addition, it can pre-compensate disturbances based on the predicted disturbances, thereby improving the robustness, real-time performance, and reliability of the control. By actively predicting and utilizing disturbances, a discrete time-varying model with disturbance inputs that is closer to the real scenario is established, making the prediction results more reliable. The MPC architecture fully considers various constraints, effectively avoiding the risks brought about by extreme cases.

[0050] Figure 2 A schematic diagram of a model predictive control device 200 based on disturbance prediction is shown.

[0051] like Figure 2 As shown, a model predictive control device 200 based on disturbance prediction mainly includes: The acquisition module 201 is used to acquire the disturbance data and actual state quantities of the controlled object at the current moment in real time. The controlled object is an aircraft, and the actual state quantities include the altitude, speed and attitude angle of the controlled object. The disturbance prediction module 202 is used to determine a disturbance prediction sequence based on the disturbance data of the controlled object at the current moment and historical disturbance data, wherein the disturbance prediction sequence represents the disturbance amount at multiple future moments; The discrete system state prediction model construction module 203 is used to construct a discrete system state prediction model of the controlled object based on a preset time-varying system matrix sequence and the disturbance prediction sequence. The discrete system state prediction model characterizes the relationship between the state variables of the controlled object and the control input. The control sequence prediction module 204 is used to perform rolling optimization based on the discrete system state prediction model, the actual state quantity at the current time, and the constraints, with the goal of minimizing the preset cost function, to obtain the optimal control sequence in the future prediction time domain. The control module 205 is used to take the first control quantity in the optimal control sequence as the actual control command at the current moment and output it to the actuator of the controlled object to realize the rolling optimization control of the controlled object.

[0052] In one example, the module in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0053] For example, when modules in a device can be implemented via a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these modules can be integrated together as a system-on-a-chip (SOC).

[0054] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.

[0055] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0056] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0057] Figure 3 This is a structural block diagram of an electronic device 300 according to an embodiment of this application.

[0058] like Figure 3As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0059] The processor 301 controls the overall operation of the electronic device 300 to complete all or part of the steps in the aforementioned model predictive control method based on disturbance prediction. The memory 302 stores various types of data to support the operation of the electronic device 300. This data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0060] I / O interface 303 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 304 is used to test wired or wireless communication between electronic device 300 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 304 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0061] The communication bus 305 may include a path for transmitting information between the aforementioned components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 may be divided into an address bus, a data bus, a control bus, etc.

[0062] The electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the model predictive control method based on disturbance prediction given in the above embodiments.

[0063] The following describes the computer-readable storage medium provided in the embodiments of this application. The computer-readable storage medium described below can be referred to in correspondence with the model predictive control method based on disturbance prediction described above.

[0064] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described model predictive control method based on disturbance prediction.

[0065] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0067] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A model predictive control method based on disturbance prediction, characterized in that, The method comprises the steps of: real-time acquisition of disturbance data and actual state quantity of the controlled object at the current time, the controlled object being an aircraft, and the actual state quantity including height, speed and attitude angle of the controlled object; determination of a disturbance prediction sequence based on the disturbance data and historical disturbance data of the controlled object at the current time, the disturbance prediction sequence representing disturbance quantities at multiple future times; construction of a discrete system state prediction model of the controlled object based on a preset time-varying system matrix sequence and the disturbance prediction sequence, the discrete system state prediction model representing the relationship between the state quantity and the control input of the controlled object; rolling optimization solution based on the discrete system state prediction model, the actual state quantity at the current time and the constraint condition, with the objective of minimizing a preset cost function, to obtain an optimal control sequence in a future prediction time domain; output of the first control quantity in the optimal control sequence as the actual control instruction at the current time to an execution mechanism of the controlled object, so as to realize rolling optimization control of the controlled object.

2. The model predictive control method based on disturbance prediction according to claim 1, characterized in that, The method comprises the steps of: real-time acquisition of wind speed of the controlled object at the current time; determination of the disturbance data of the controlled object at the current time based on the wind speed at the current time and a preset disturbance model; the preset disturbance model is: ； Wherein, F matrix is time-varying (LPV) conversion matrix, is the wind speed, and e is the error amount.

3. The model predictive control method based on disturbance prediction according to claim 2, characterized in that, The method comprises the steps of: determination of the disturbance prediction sequence based on a Gaussian process regression model, the disturbance data at the current time and the historical disturbance data.

4. The model predictive control method based on disturbance prediction according to claim 1, characterized in that, The method comprises the steps of: construction of an MPC prediction model matrix based on the disturbance prediction sequence and the discrete system state prediction model, the MPC prediction model matrix representing the prediction path of the system state under the initial state and the future control sequence in a prediction time domain; solution based on the MPC prediction model matrix, the actual state quantity at the current time and the constraint condition, with the objective of minimizing the cost function, to obtain the optimal control sequence in the future prediction time domain.

5. The model predictive control method based on disturbance prediction according to claim 4, characterized in that, The cost function is a quadratic cost function, and the expression of the quadratic cost function is: ； wherein, is a future control sequence vector, is an actual state quantity at a current time, and C and V are coefficient matrices calculated based on the MPC prediction model matrix and a preset weight matrix.

6. The model predictive control method based on disturbance prediction according to claim 4, characterized in that, The constraint condition comprises: ； ； ； wherein G and H are constraint matrices, representing the coupling constraint between the input and the state quantity in actual application.

7. The model predictive control method based on disturbance prediction according to claim 4, characterized in that, The method comprises the steps of: construction of a quadratic programming problem based on the MPC prediction model matrix, the actual state quantity at the current time and the constraint condition, and calling of a quadratic programming solver to solve the quadratic programming problem, so as to obtain the optimal control sequence in the future prediction time domain.

8. A model predictive control apparatus based on disturbance prediction, characterized by, The method comprises the steps of: acquisition of disturbance data and actual state quantity of the controlled object at the current time, the controlled object being an aircraft, and the actual state quantity including height, speed and attitude angle of the controlled object; a disturbance prediction module, configured to determine a disturbance prediction sequence based on disturbance data of the controlled object at a current time and historical disturbance data, the disturbance prediction sequence representing disturbance amounts at multiple future times; a discrete system state prediction model construction module, configured to construct a discrete system state prediction model of the controlled object based on a preset time-varying system matrix sequence and the disturbance prediction sequence, the discrete system state prediction model representing a relationship between state amounts and control inputs of the controlled object; a control sequence prediction module, configured to perform rolling optimization based on the discrete system state prediction model, actual state amounts at the current time, and constraint conditions, to obtain an optimal control sequence in a future prediction time domain, with a preset cost function being minimized as an objective; a control module, configured to output a first control amount in the optimal control sequence as an actual control instruction at the current time to an execution mechanism of the controlled object, to implement rolling optimization control on the controlled object.

9. An electronic device, comprising: comprising a processor coupled to a memory; the processor is configured to execute a computer program stored in the memory, so that the electronic device performs the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, comprising computer programs or instructions which, when executed on a computer, cause the computer to perform the method of any one of claims 1-7.

Images