Model prediction control method and device suitable for forklift drive-by-wire drum brake system

Abstract

The invention discloses a model prediction control method and device suitable for a drive-by-wire drum brake system of a forklift, and the method comprises the steps: 1) carrying out the mechanism analysis of a mechanical structure and dynamic characteristics of the drive-by-wire drum brake system of the forklift, and constructing a dynamic model of the drive-by-wire drum brake system; 2) designing a model prediction controller according to the model established in the step 1), determining a prediction time domain and a control time domain, constructing a cost function, and explicitly adding state, input, output and other constraints; (3) according to the formula established in the step (2), a model prediction control algorithm for the drive-by-wire drum brake system of the forklift is obtained on the basis of the designed controller; and 4) on the basis of the Lyapunov stability theory, proving that the control algorithm designed in the step 3 can enable the system state to converge within finite time, and carrying out a simulation experiment in an MATLAB / Simulink environment to verify the actual control effect of the control algorithm in the braking system.

Description

Technical Field

[0001] This invention belongs to the field of industrial vehicle brake-by-wire technology, and proposes a control method and device for a forklift brake-by-wire drum braking system based on Model Predictive Control (MPC). This method is based on a vehicle braking system model. Through a model predictive controller, it determines the prediction and control time domains, constructs a cost function, and explicitly incorporates the state, input, and output. By designing a constraint state estimator, it aims to estimate the system state as accurately as possible for MPC prediction under noise and model uncertainty, ensuring closed-loop feasibility and stability, and achieving precise allocation of braking force and response optimization. Compared with traditional control methods, this invention significantly improves response speed, braking accuracy, constraint handling capability, and adaptability to complex operating conditions. Background Technology

[0002] With the evolution of the logistics industry towards intelligence and automation, forklifts, as core equipment for industrial material handling, have their control system precision and anti-interference capabilities profoundly impacting overall operational efficiency and safety. However, traditional mechanical or hydraulic braking solutions exhibit significant limitations in dynamic response, control precision, and electronic system compatibility, failing to meet the stringent requirements of high-standard control performance in modern warehousing and logistics scenarios. Therefore, brake-by-wire (BBW) technology, based on electrical signal transmission, has gradually become a research and application hotspot, especially brake-by-wire drum braking systems, which offer significant advantages in structural simplification, refined control, and integration of multiple braking strategies, representing a crucial direction for forklift braking technology development. In complex real-world operating environments, forklifts often need to cope with dynamic load changes, time-varying road surface adhesion conditions, inconsistent brake drum clearances, and system unsteady-state problems caused by the coupling effect between the motor and mechanical braking. These multiple factors collectively lead to significant uncertainties in model parameters. At the same time, the brake-by-wire system itself also faces inherent limitations such as actuator response delay, strong nonlinearity of friction characteristics and sensor signal noise. It is highly susceptible to interference from external environmental disturbances, which can lead to deterioration of control accuracy and even induce potential operational safety hazards.

[0003] To address these challenges, the performance of the control strategy becomes a key indicator in system design. Among various control theories, Model Predictive Control (MMCC) has become a widely adopted high-performance control strategy in industry due to its unique advantages in accurate system modeling with weak dependencies and high control accuracy. Compared with traditional PID control methods, MMCC's characteristics stem from its model-based prediction and online optimization implementation: MMCC supports multivariable feedback control and can simultaneously handle coupling between multiple inputs and outputs, explicit constraints, and multi-objective trade-offs, which is more powerful than single-variable controllers. Furthermore, MMCC allows constraints on the system's state, inputs, and outputs during the design phase, ensuring the feasibility and stability of the closed-loop system, which is beneficial for achieving precise allocation and response optimization of forklift braking force.

[0004] In summary, this model predictive control strategy applied to a forklift drive-by-wire drum brake system, through the use of a system predictive model, explicit constraint processing, and an online parameter adaptive update mechanism, can not only specifically attenuate external disturbances in different frequency bands, but also achieve safe control in the event of system failure or actuator performance degradation. Furthermore, under the premise of strictly meeting physical constraints and real-time calculation requirements, this method can significantly enhance braking control accuracy and system stability, thus providing strong technical support for the smooth and safe operation of intelligent forklifts. Summary of the Invention

[0005] To address the issues of slow response and insufficient control precision in forklift drive-by-wire drum brake systems under complex operating conditions, this invention proposes a forklift drive-by-wire drum brake control method and device based on Model Predictive Control (MPC). This method is an advanced control strategy that utilizes the system's mathematical model to predict the system's behavior over a future period by solving an optimization problem at each control moment, and calculates the optimal control input. Furthermore, it only implements the control action at the current moment, repeating this process at the next moment, thereby maintaining rapid response and high-precision braking force distribution under complex operating conditions, ensuring the operational safety and handling efficiency of new energy forklifts.

[0006] To address the aforementioned technical challenges, the first aspect of this invention relates to a model predictive control method for a forklift drive-by-wire drum brake system, comprising the following steps: Step 1: Conduct a mechanism analysis on the mechanical structure and dynamic characteristics of the forklift's drive-by-wire drum braking system, and then construct its dynamic model; Step 2: Based on the model established in Step 1, design a model predictive controller, determine the prediction time domain and control time domain, construct the cost function, and explicitly add constraints such as state, input, and output. Step 3: Based on the formula established in Step 2 and the designed controller, obtain the model predictive control algorithm for the forklift drive-by-wire drum brake system. Step 4: Based on Lyapunov stability theory, prove that the control algorithm designed in Step 3 can make the system state converge stably, and conduct simulation experiments in MATLAB / Simulink environment to verify the control effect of the control algorithm in the braking system.

[0007] The second technical solution of this invention proposes a control device that applies model predictive control to a forklift drive-by-wire drum braking system. This device organically integrates electro-hydraulic braking (EHB), an electronic control unit (ECU), and a high-speed data communication module, and is equipped with an online optimization solver and a state / disturbance estimator to build a collaborative operating platform. By leveraging predictive models, explicit expression of constraints, and real-time optimization calculations, it achieves closed-loop precise control of the braking process, and while ensuring real-time response, it also possesses basic fault tolerance capabilities.

[0008] To address the braking challenges of forklifts under complex operating conditions, this invention proposes a novel control strategy aimed at enhancing system control performance under strong nonlinearity, model uncertainty, and external disturbances. Compared to existing methods, this approach offers advantages including: utilizing model prediction and online constraint optimization to directly handle constraints such as actuator saturation and state limitations, generating a constrained optimal control sequence while balancing tracking accuracy and control smoothness. Even under complex operating conditions such as heavy load, no-load, or frequent start-stop cycles, the system maintains stable and controlled braking characteristics, thereby improving forklift operating efficiency and safety. Attached Figure Description

[0009] To more intuitively illustrate the technical solution of the present invention, the accompanying drawings referenced in the technical description of the present invention are now briefly introduced. It should be noted that the following drawings are only schematic diagrams of some embodiments of the present invention. For those skilled in the art, other similar illustrations can be drawn based on these drawings without any creative effort.

[0010] Figure 1 This is a schematic diagram illustrating the dynamic modeling principle of the forklift drive-by-wire drum braking system proposed in this invention. Figure 2 A schematic diagram of the overall structure of the simulation program built on the MATLAB / Simulink platform is shown. Figure 3 The figure shows the simulation results of tracking the master cylinder piston position under emergency conditions in the MATLAB / Simulink environment. Figure 4 The simulation curve of the master cylinder piston position tracking error under emergency conditions is shown. Figure 5 This is a simulation graph of the servo motor output torque during an emergency situation. Figure 6 The figure shows the simulation results of tracking the master cylinder piston position under steady operating conditions in the MATLAB / Simulink environment. Figure 7 The simulation curve of the master cylinder piston position tracking error under steady-state conditions is shown. Figure 8 This is a simulation curve of the servo motor output torque during a steady-state operation. Figure 9 This is a flowchart illustrating the implementation of the control method of the present invention. Detailed Implementation

[0011] The technical solution of the present invention will be further described below with reference to the accompanying drawings. Example 1

[0012] This embodiment relates to a model predictive control method applied to a forklift drive-by-wire drum brake system. The specific implementation steps are as follows: Step 1: Conduct a mechanism analysis on the mechanical structure and dynamic characteristics of the forklift's drive-by-wire drum braking system, and then construct its dynamic model; The actual displacement of the master cylinder piston in the system is set as follows: The actual speed is acceleration is , Indicates the gear ratio. For transmission efficiency, For the gear radius, For the internal hydraulic system, The cross-sectional area of ​​the master cylinder piston. This represents the torque output by the motor. Based on the analysis of the relationships between various physical variables within the forklift's drive-by-wire drum brake system, and combined with Newton's second law, the dynamic equations of this braking system can be derived as follows:

[0013] in, The damping coefficient is... This is the stiffness coefficient; For rack force, denoted as ; The internal hydraulic pressure of the system is denoted as . ; The spring force is denoted as . ; Friction force, denoted as Assuming that the movement of the master cylinder piston does not affect the fluid volume within the system, the dynamic relationship of the hydraulic part of the system can be expressed as:

[0014] in, Defined as The first differential of time Represents the internal brake fluid volume flow rate. This indicates the bulk modulus of brake fluid. The following describes the hydraulic pressurization model of the system's wheel cylinders:

[0015] in, Indicates wheel cylinder hydraulics, taking into account Substituting equation (4) into equation (3) yields:

[0016] By integrating equation (5), the mapping relationship between the pressure inside the master cylinder and the position of the master cylinder piston can be described as follows:

[0017] in, This is the initial pressure inside the cylinder. In practical applications, the pressure inside the master cylinder With piston displacement There exists a nonlinear relationship between them, which is affected by various factors such as dead zone effect, friction, temperature fluctuation, and brake pad wear. In the dynamic modeling process, these nonlinear effects can be regarded as disturbance terms of the system. To facilitate modeling and controller design, a second-order polynomial can be used to represent the master cylinder pressure. With piston displacement The relationship curve between pressure and displacement is fitted. This method not only reflects the trend of pressure changing with displacement, but also simplifies the construction of subsequent control strategies. The fitted pressure... With displacement The second-order relation between them is expressed as follows:

[0018] Among them, parameters It describes the master cylinder pressure. With piston displacement The important characteristic variables of the relationship between them are closely related to the overall stiffness of the system and the properties of the hydraulic medium.

[0019] The core purpose of the above modeling method is to provide a dynamic model with reasonable accuracy and easy implementation for the controller design of the forklift wire braking system, rather than pursuing a completely accurate physical characterization of the system.

[0020] Based on the above analysis, the complete dynamic model of the controlled system can be derived as follows:

[0021] in, To account for the uncertainties in the lumped model parameters and external disturbances, based on the mechanical structure and dynamic properties of the forklift's steer-by-wire braking system, it can be assumed that the lumped perturbation has an upper bound, i.e. ,in, Its upper limit.

[0022] Step 2: Design a model predictive controller, determine the prediction time domain and control time domain, construct the cost function and explicitly add constraints such as state, input, and output; From the original equation

[0023] To facilitate the application of model predictive control theory, we first linearize the point... Simplify by ignoring higher-order terms: .

[0024] Design prediction time domain and control time domain: Prediction Time Domain , representing the number of future steps to predict the system state; control time domain represents the length of the actual optimized control input sequence. Based on experience, the initial value can be set as follows:

[0025] The goal of the cost function is to minimize the performance metric considering the worst-case uncertainty scenario. Where: The control sequence to be optimized ; The model parameters are an uncertain set; For external disturbances A bounded set; For the state based on the current moment and control sequences in prediction Predicted state after the step; Use the target reference trajectory; This is the weight matrix for the state tracking error; To predict the first in the time domain Step control input; The steady-state control input required to keep the system near the desired reference point.

[0026] State, input, and output constraints:

[0027] in: The region is affected by uncertainty; It is a terminal invariant set; Step 3: Based on the designed controller, obtain the model predictive control algorithm for the forklift drive-by-wire drum brake system; Based on the obtained constraint functions, the control law of the model predictive control algorithm is derived:

[0028] Step 4: Based on Lyapunov stability theory, prove that the designed control algorithm can make the system state converge in a finite time, and conduct simulation experiments in MATLAB / Simulink environment to verify the actual control effect of the control algorithm in the braking system. Write the system's dynamic equations in affine form:

[0029] in: It is the stage cost function, usually defined as a positive definite quadratic form; Differentiate the Lyapunov function:

[0030] Substituting the control law and rearranging, we get:

[0031] The design goal of model predictive control is to make Not correct, and should hour, .therefore This model predictive control system can guarantee asymptotic stability of the system under the worst-case conditions of all energy-bounded disturbances, and has the specified robust performance.

[0032] The dynamic model parameters of the forklift drive-by-wire drum brake system are given below:

[0033] Two verification conditions were designed in the simulation. The first is smooth braking, with the expected position of the master cylinder piston. The expression is as follows:

[0034] The second scenario is emergency braking, where the forklift's drive-by-wire braking system is positioned as the expected master cylinder piston. It is its extreme position, that is =0.06 m.

[0035] Simulation programs built on the MATLAB / Simulink software platform, such as Figure 2As shown, the simulation results are as follows: Figures 3 to 8 As shown.

[0036] Simulation results show that the designed model predictive control strategy can achieve stable tracking of the master cylinder piston to the reference position, demonstrating good control stability. This further illustrates that the model predictive control algorithm has excellent tracking performance when facing different types of master cylinder piston target trajectories, effectively verifying the applicability and reliability of this control method in forklift brake-by-wire systems.

[0037] Example 2 This embodiment relates to a model predictive control device for a forklift drive-by-wire drum brake system. It utilizes a pedal displacement sensor, a data acquisition module, a host computer core computing platform with built-in model predictive control, a servo motor encoder, a CAN data bus, an electronic control unit (ECU) as the system's lower-level unit, a master cylinder servo motor, brake fluid lines, an in-cylinder pressure sensor, and a data communication unit to achieve precise closed-loop tracking control of the expected master cylinder piston position. The specific implementation steps are as follows: Step 1: Receive the braking demand input signal from the forklift; When the driver presses the brake pedal, the pedal displacement sensor converts the displacement change into a corresponding electrical signal output.

[0038] Step 2: Calculate the target braking response signal based on the input information; After receiving the electrical signal from the pedal displacement sensor, the host computer uses the correspondence between the forklift pedal travel and the target master cylinder piston displacement to determine the required desired piston displacement, and uses this amount as the reference input for the model predictive control (MPC) system. The host computer then runs the model predictive control algorithm based on the current braking demand and the real-time system status (including the actual position of the piston in the master cylinder). With speed The system calculates and outputs the control signal, i.e., the servo motor torque. The actual displacement of the master cylinder piston... Instead of direct measurement, the angular displacement signal is obtained by measuring the angular displacement signal from the servo motor encoder, and then indirectly calculated based on the conversion relationship between angular displacement and piston displacement.

[0039] Step 3: Transmit the calculated control commands to the execution unit; The host computer sends relevant signals to the electronic control unit (ECU) via the CAN bus, including the desired position signal of the master cylinder piston and the torque command signal for controlling the servo motor.

[0040] Step 4: Drive the actuator to complete the braking operation; After receiving the control command transmitted from the host computer, the electronic control unit (ECU) will start the servo motor inside the EHB system. The servo motor will drive the master cylinder piston through the internal reduction gear transmission mechanism to achieve the feed motion.

[0041] Step 5: Generate the corresponding braking force; The feed motion of the master cylinder piston compresses the brake fluid in the cylinder, which is then delivered to the brake lines and further acts on the wheel cylinders, thereby generating the braking force required to brake the forklift wheels.

[0042] Step 6: Adjust the control precision through a closed-loop feedback mechanism; The real-time monitoring of the system status is jointly handled by the in-cylinder pressure sensor and the servo motor angular displacement sensor, with the core objective of determining the current position of the master cylinder piston. The electronic control unit (ECU) transmits these sensor-acquired signals to the host computer via the CAN bus. The data communication module uses this feedback information, combined with model predictive control algorithms, to adjust the control input (i.e., the target output torque generated by the servo motor) in real time. Subsequently, the updated control signal is sent back to the lower-level ECU to apply dynamic corrections to the entire braking process. This process forms a closed-loop feedback control structure, ensuring that the master cylinder piston accurately follows the set target position while maintaining the required hydraulic pressure within the master cylinder, thereby efficiently responding to and meeting the forklift's braking needs.

[0043] The above description is merely one embodiment of the present invention, intended to illustrate the technical solution of the present invention, and does not constitute a limitation on its application scope. For those skilled in the art, various modifications, substitutions, or optimizations can be made without departing from the core ideas and basic principles of the present invention, and all equivalent variations of these technical solutions should be covered within the protection scope of the present invention.

Claims

1. A model predictive control method applicable to a forklift drive-by-wire drum brake system, specifically including the following implementation steps: Step 1: Conduct a mechanism analysis on the mechanical structure and dynamic characteristics of the forklift's drive-by-wire drum braking system, and then construct its dynamic model; Step 2: Based on the model established in Step 1, design a model predictive controller, determine the prediction time domain and control time domain, construct the cost function, and explicitly add constraints such as state, input, and output. Step 3: Based on the formula established in Step 2, and the designed controller, obtain the model predictive control algorithm for the forklift drive-by-wire drum brake system. Step 4: Based on Lyapunov stability theory, prove that the control algorithm designed in Step 3 can make the system state converge stably, and conduct simulation experiments in MATLAB / Simulink environment to verify the control effect of the control algorithm in the braking system.

2. The model predictive control method as described in claim 1, characterized in that, Step 1 further includes the following: setting the actual displacement of the master cylinder piston in the system as... The actual speed is acceleration is , Indicates the gear ratio. For transmission efficiency, For the gear radius, For the internal hydraulic system, The cross-sectional area of ​​the master cylinder piston. The torque output by the motor; based on the analysis of the relationships between various physical variables within the forklift's drive-by-wire drum brake system, and combined with Newton's second law, the dynamic equations of the braking system can be derived as follows: in, The damping coefficient is... This is the stiffness coefficient; For rack force, denoted as ; The internal hydraulic pressure of the system is denoted as . ; The spring force is denoted as . ; Friction force, denoted as , Assuming that the movement of the master cylinder piston does not affect the fluid volume within the system, the dynamic relationship of the hydraulic part of the system can be expressed as: in, Defined as The first differential of time Represents the internal brake fluid volume flow rate. This indicates the bulk modulus of brake fluid. The following describes the hydraulic pressurization model of the system's wheel cylinders: in, Indicates wheel cylinder hydraulics, taking into account Substituting equation (4) into equation (3) yields: By integrating equation (5), the mapping relationship between the pressure inside the master cylinder and the position of the master cylinder piston can be described as follows: in, This is the initial pressure inside the cylinder. In practical applications, the pressure inside the master cylinder With piston displacement There exists a nonlinear relationship between them, which is affected by various factors such as dead zone effect, friction, temperature fluctuation, and brake pad wear. In the dynamic modeling process, these nonlinear effects can be regarded as disturbance terms of the system. To facilitate modeling and controller design, a second-order polynomial can be used to represent the master cylinder pressure. With piston displacement The relationship curve between the pressure and displacement is fitted; this method not only reflects the trend of pressure changing with displacement, but also simplifies the construction of subsequent control strategies; the fitted pressure With displacement The second-order relation between them is expressed as follows: Among them, parameters It describes the master cylinder pressure. With piston displacement An important characteristic variable in the relationship between them, its physical meaning is a constant related to the overall stiffness of the system and the properties of the hydraulic medium; The core purpose of the above modeling method is to provide a dynamic model with reasonable accuracy and easy implementation for the controller design of the forklift wire braking system, rather than pursuing a completely accurate physical characterization of the system. Based on the above analysis, the complete dynamic model of the controlled system can be derived as follows: in, To account for the uncertainties in the lumped model parameters and external disturbances, based on the mechanical structure and dynamic properties of the forklift's steer-by-wire braking system, it can be assumed that the lumped perturbation has an upper bound, i.e. ,in, Its upper limit.

3. The model predictive control method as described in claim 1, characterized in that, Step 2 specifically includes: From the original equation To facilitate the application of model predictive control theory, we first linearize the point... Simplify by ignoring higher-order terms: Rewritten as a standard state-space model: Design prediction time domain and control time domain: Set the prediction time domain This represents the number of future steps to predict the system state; it sets the control time domain. , representing the actual length of the control input sequence being optimized, can be initially set as follows based on experience: Construct a cost function whose goal is to minimize the performance metric considering the worst-case uncertainty scenario: in: The control sequence to be optimized ; The model parameters are an uncertain set; For external disturbances A bounded set; For the state based on the current moment and control sequences in prediction Predicted state after the step; Use the target reference trajectory; This is the weight matrix for the state tracking error; To predict the first in the time domain Step control input; The steady-state control input required to keep the system near the desired reference point; State, input, and output constraints: in: The region is affected by uncertainty; It is a terminal invariant set.

4. The model predictive control method as described in claim 1, characterized in that, Step 3 specifically includes: Based on the designed controller, a model predictive control algorithm for a forklift drive-by-wire drum brake system is obtained. Based on the obtained constraint functions, the control law of the model predictive control algorithm is derived: 。 5. The model predictive control method as described in claim 1, characterized in that, Step 4 specifically includes: Write the system's dynamic equations in affine form: The design goal of model predictive control is to make Not correct, and should hour, ; therefore This model predictive control system can guarantee asymptotic stability of the system under the worst-case conditions of all energy-bounded disturbances, and has the specified robust performance.

6. A model predictive control device suitable for a forklift drive-by-wire drum brake system, which achieves precise closed-loop tracking control of the target master cylinder piston position through a host computer core computing platform—an electronic control unit (ECU) with embedded model predictive control algorithms; the specific implementation steps are as follows: Step 1: Receive the braking demand input signal from the forklift; Step 2: Calculate the target braking response signal based on the input information; Step 3: Transmit the calculated control commands to the execution unit; Step 4: Drive the actuator to complete the braking operation; Step 5: Generate the corresponding braking force; Step 6: Adjust the control precision through a closed-loop feedback mechanism.

7. The model prediction control device as described in claim 6, characterized in that, Step 1 specifically includes: when the driver presses the brake pedal, the pedal displacement sensor will convert the displacement change into a corresponding electrical signal output.

8. The model prediction control device as described in claim 6, characterized in that, Step 2 specifically includes: After receiving the electrical signal from the pedal displacement sensor, the host computer uses the correspondence between the forklift pedal travel and the target master cylinder piston displacement to determine the required desired piston displacement, and uses this amount as the reference input for the model predictive control (MPC) system; the host computer runs the model predictive control algorithm, calculates and outputs the control signal, i.e., the servo motor torque, based on the current braking demand and the real-time system status (including the actual position and speed of the piston in the master cylinder); the actual displacement of the master cylinder piston is not directly measured, but is indirectly calculated by measuring the angular displacement signal obtained by the servo motor encoder and based on the conversion relationship between angular displacement and piston displacement.

9. The model prediction control device as described in claim 6, characterized in that, Step 3 specifically includes: the host computer sending relevant signals to the electronic control unit (ECU) via the CAN bus, including the desired position signal of the master cylinder piston and the torque command signal for controlling the servo motor.

10. The model prediction control device as described in claim 6, characterized in that, Step 4 specifically includes: After receiving the control command transmitted from the host computer, the electronic control unit (ECU) will start the servo motor inside the EHB system, and the servo motor will drive the master cylinder piston to achieve the feed motion through the internal reduction gear transmission mechanism.

11. The model prediction control device as described in claim 6, characterized in that, Step 5 specifically includes: the feed motion of the master cylinder piston compresses the brake fluid in the cylinder, which is then delivered to the brake line and further acts on the wheel cylinder, thereby generating the braking force required to brake the forklift wheels.

12. The model prediction control device as described in claim 6, characterized in that, Step 6 specifically includes: The real-time monitoring of the system status is jointly handled by the in-cylinder pressure sensor and the servo motor angular displacement sensor, with the core purpose of knowing the current position of the master cylinder piston; the electronic control unit (ECU) transmits these signals acquired by the sensors to the host computer via the CAN bus; the data communication module uses this feedback information, combined with the model predictive control algorithm, to adjust the control quantity (i.e., the target output torque generated by the servo motor) in real time; subsequently, the updated control signal is sent to the lower-level ECU again to apply dynamic correction to the entire braking execution process; this process forms a closed-loop feedback control structure, ensuring that the master cylinder piston can accurately follow the set target position, while maintaining the required hydraulic pressure inside the master cylinder, thereby efficiently responding to and meeting the braking needs of the forklift.

Images