Torque feed-forward and PID closed loop control method and system based on second-order lag modeling

Abstract

The invention relates to a torque feedforward and PID closed loop control method and system based on second-order lag modeling, and belongs to the field of automobile testing. The system comprises a controller, a motor, a driver of the motor, a force sensor installed at the load end of the motor, and an encoder. A torque feed-forward value is calculated and adjusted according to the target force, a PID target value is calculated through a second-order lag system model of a motor transmission system, and PID control of the force is constructed by reading a force value on a force sensor. The method is good in dynamic property, small in overshoot and high in steady-state precision; the method is not influenced by a target force fluctuation range, and has good robustness in a motor force range.

Description

Technical Field

[0001] This invention relates to the field of automotive testing technology, specifically to a control method and system based on second-order hysteresis modeling, torque feedforward, and PID closed-loop. Background Technology

[0002] In the automotive testing field, instruments such as chassis dynamometers, powertrain test benches, and mechanical-grade HiL test benches meet the dynamic response testing requirements of complete vehicles and components. The demands for high precision and high dynamics in motor control are also increasing. Bench testing not only needs to simulate vehicle inertia but also the resistance of vehicle movement. Accurate simulation requires not only precise electrical inertia control technology but also high-performance force control technology. Force control methods primarily rely on controllers or PLCs to collect force sensor values ​​and construct a force-controlled PID closed loop. Traditional methods use analog signals or fieldbuses to send the feedforward torque and the torque of the force closed loop to the motor driver. This control method has good control effects on servo systems. However, if the system has a large mechanical inertia or contains flywheels, the system response will deteriorate. If a high dynamic response output is required, the system may become unstable or lose control.

[0003] Based on this, the present invention is proposed. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a control method and system based on second-order hysteresis modeling, torque feedforward, and PID closed-loop control. The technical solution is as follows:

[0005] The first aspect is a control system based on second-order hysteresis modeling, torque feedforward, and PID closed loop, including a controller, a motor, a motor driver, a force sensor installed at the motor load end, and an encoder.

[0006] The motor driver operates in torque mode or current mode. The target force is linearly converted into the torque feedforward value of the driver. At the same time, the target force is given to the second-order hysteresis system model of the motor drive system. The torque feedforward value is adjusted according to the target force calculation, and the target value of the PID is obtained by calculating through the second-order hysteresis system model of the motor drive system. The force PID control is constructed by reading the force value on the force sensor. The sum of the torque feedforward value and the PID output value is given to the driver, and the closed-loop control of the force is realized on the controller.

[0007] The driver establishes vector control through an encoder.

[0008] As a further aspect of the present invention, the controller is a PID controller.

[0009] Secondly, the control method based on second-order hysteresis modeling, torque feedforward, and PID closed-loop is executed through the aforementioned control system, including the following steps:

[0010] Step 1, Torque Feedforward Value It is based on the input target force generate;

[0011] Step 2: Construct a second-order lag system model of the motor drive system, based on the input target force. Generate the setpoint signal for the PID controller; the setpoint signal for the PID controller is the target force output. After passing through the PID controller, it becomes the output of the PID controller. Provided to the driver;

[0012] Step 3: Transfer function in the complex frequency domain of the second-order lag system model of the motor drive system. The expression is:

[0013] ;

[0014] in:

[0015] Indicates the input target force The Laplace transform of;

[0016] Indicates the target force output The Laplace transform of;

[0017] S is a complex variable, representing the complex frequency domain;

[0018] It is the damping ratio of the motor drive system;

[0019] It is the cutoff frequency of the motor drive system;

[0020] It is the lag time;

[0021] It is the proportionality coefficient;

[0022] Step 4: The controller reads the force sensor data. After being processed by a notch filter and a first-order low-pass filter, the filtered result is obtained. The controller acquires encoder signals through a counting module and calculates the motor's speed and angular acceleration.

[0023] Step 5: Based on the feedback from the force sensor The target force output after processing by the second-order hysteresis system model of the motor drive system The controller outputs F in a closed loop. PID ;

[0024] Step 6: Calculate the controller output force ;

[0025] Step 7: Send the output force F to the driver;

[0026] Step 8: Repeat steps 1-7;

[0027] Step 9: When the motor speed is less than the preset threshold, or when exiting torque mode / current mode, exit control.

[0028] As a further aspect of the present invention, in step 7, the output force F is sent to the driver via a bus or analog signal.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] Good dynamic performance (good dynamic performance means that the system response time is less than 100ms when simulating different vehicles);

[0031] Small overshoot and high steady-state accuracy;

[0032] Unaffected by the fluctuation range of the target force (second-order model modeling, after parameter tuning, the mathematical model of the motor drive system is obtained, and the PID closed loop established by it will not oscillate, and at the same time, automatic adjustment and control are achieved through feedforward force calculation and PID closed loop as the target force changes), it has good robustness within the range of motor force.

[0033] There is no need to measure the inherent internal resistance of the equipment; it can be adjusted in real time through force-closed-loop PID control.

[0034] For systems with mechanical flywheels, it is only necessary to adjust the force of the inertia disk to the feedforward torque in real time, without the need to readjust the PID parameters, which can achieve fast response and high-precision control. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the control method and system based on second-order lag modeling, torque feedforward, and PID closed-loop.

[0036] Figure 2 The simulation results are shown when the vehicle's inertia is 702kg, and an additional 2000N excitation is applied, which is equivalent to simulating a dynamic loading process with positive acceleration.

[0037] Figure 3 This is a diagram showing the results of a dynamic loading process where the inertia of a vehicle is 702 kg, an excitation of -2000 N is applied, the acceleration is negative, and the deceleration is negative.

[0038] Figure 4 The simulation results are shown when the vehicle's inertia is 1400kg and an additional 2000N excitation is applied, which is equivalent to simulating a dynamic loading process with positive acceleration.

[0039] Figure 5 This is a diagram showing the results of a dynamic loading process where the inertia of a vehicle is 1400kg, an excitation of -2000N is applied, the acceleration is negative, and the deceleration is negative.

[0040] Figure 6 The simulation results are shown when the vehicle's inertia is 2495kg, and an additional 2000N excitation is applied, which is equivalent to simulating a dynamic loading process with positive acceleration.

[0041] Figure 7 This is a diagram showing the results of a dynamic loading process where the inertia of a vehicle is 2495 kg, an excitation of -2000 N is applied, the acceleration is negative, and the deceleration is negative. Detailed Implementation

[0042] The present invention will be described in detail below with reference to specific embodiments. These embodiments are merely some, not all, implementations of the present invention. All other implementations obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0043] Example 1

[0044] Considering that the main problems of the current mainstream control methods are slow dynamic response, large overshoot, and poor steady-state accuracy, control instability may also occur in highly dynamic systems.

[0045] This invention proposes a control method and system based on second-order hysteresis modeling, torque feedforward, and PID closed-loop to achieve force control. The motor driver operates in torque mode or current mode, and the motor load terminal is equipped with a force sensor or torque sensor. In this embodiment, a force sensor is used as an example for illustration. The target force ( The signal is linearly converted into a torque feedforward value for the driver. Simultaneously, the target force is fed into the second-order hysteresis system model of the motor drive system. Closed-loop force control is implemented on the controller or PLC, and the PID output and torque feedforward value are simultaneously provided to the driver to achieve high dynamics and precise force control. This method obtains the required force output from the motor through the torque feedforward value, and achieves high dynamics and steady-state accuracy of force control by constructing a second-order hysteresis system model of the motor drive system and a PID force control closed loop. During debugging, only parameters such as the cutoff frequency, hysteresis time, and damping ratio of the second-order hysteresis system model of the motor drive system need to be modified to construct the motor model. Multiple sets of PID parameters are not required to achieve the dynamics and control accuracy of motor force control. Furthermore, the influence of wind resistance and bearing resistance in the mechanical system does not need to be considered excessively.

[0046] like Figure 1 As shown: A torque feedforward and PID closed-loop control system based on second-order hysteresis modeling, including a controller, a motor, a motor driver, a force sensor installed at the motor load end, and an encoder;

[0047] The motor driver operates in torque mode or current mode. The target force is linearly converted into the torque feedforward value of the driver. At the same time, the target force is given to the second-order hysteresis system model of the motor drive system. The closed-loop control of force is realized on the controller or PLC. The PID output and the torque feedforward value are given to the driver at the same time to achieve force control.

[0048] Encoders are mainly used by controllers to calculate speed and acceleration, while drivers use encoders to establish vector control.

[0049] The control method of this invention can be used not only for the control of electric inertia, but also for high-dynamic force control in hardware-in-the-loop simulation. This method constructs a second-order hysteresis system model of the motor drive system and obtains this model by adjusting parameters such as cutoff frequency, hysteresis time, and damping ratio. During control, the torque feedforward value is calculated and adjusted based on the target force, and the target value of the PID controller is calculated using the second-order hysteresis system model of the motor drive system. Force PID control is constructed by reading the force value from the force sensor. The torque feedforward value and the PID output value (…) The sum of the values ​​is given to the driver, and the PID control closed loop will automatically compensate for the inherent internal resistance and wind resistance of the system, so there is no need to perform very precise internal resistance measurement.

[0050] This control method can be used in systems where the motor has a roller or flywheel, or in systems where the motor is directly connected to the test piece. The steps are as follows:

[0051] Step 1, Torque Feedforward Value It is based on the input target force The generated force is the input signal of the driver; the target force described in this invention comes from the target force required to control the motor, which is calculated by simulating the road load mode of the vehicle and applied to the motor.

[0052] Step 2: Construct a second-order lag system model (referred to as the second-order model) of the motor drive system based on the input target force. The given signal for generating the PID controller is the target force output. After passing through a PID controller (PID controller is simply called controller), it becomes the output of the PID controller. Provided to the driver;

[0053] Step 3: Transfer function in the complex frequency domain of the second-order lag system model of the motor drive system. The expression is:

[0054]

[0055] in:

[0056] Indicates the input target force The Laplace transform of;

[0057] Indicates the target force output The Laplace transform of;

[0058] S is a complex variable, representing the complex frequency domain (s-domain).

[0059] It is the damping ratio of the motor drive system;

[0060] It is the cutoff frequency of the motor drive system;

[0061] It is the lag time;

[0062] It is the proportionality coefficient;

[0063] Step 4: The controller reads the force sensor data. After being processed by a notch filter and a first-order low-pass filter, the filtered result is obtained. The controller collects encoder signals through the counting module and calculates the motor's speed and angular acceleration.

[0064] Step 5: Based on the feedback from the force sensor and the target force output after processing by the second-order model Closed-loop output via controller ;

[0065] Step 6: Calculate the controller output force F. ;

[0066] Step 7: Send the output force F to the driver via bus or analog signal;

[0067] Step 8: Repeat steps 1-7;

[0068] Step 9: When the motor speed is less than the preset threshold (e.g., 0.5 km / h, the parameter can be set), or when exiting torque mode / current mode, exit control.

[0069] Example 2 (Characteristic Experiment)

[0070] The following data, after adopting the second-order model described in the embodiment, were obtained on a chassis dynamometer with a basic inertia of 1400 kg. Three different vehicles (702 kg, 1400 kg, and 2495 kg) were simulated, and the system control response and maximum overshoot were measured respectively when the excitation was 2000 N.

[0071] Figure 2 , 3 middle:

[0072] "—Trigger": indicates a 2000N stimulus;

[0073] "—BBK": indicates the algorithm corresponding to the second-order model described in this invention;

[0074] "-0.90": indicates 90% of the target value;

[0075] "--Comparison 1": Control algorithm 1 using feedforward force + PID (only feedforward force and PID closed-loop control are used, and a second-order model of motor drive is not established).

[0076] "-- Comparison 2": The control algorithm 2 adopts feedforward force + PID (with feedforward force, with lag time + PID closed-loop control, but without a complete physical model of motor drive).

[0077] The motor drive itself is a second-order lag system. "Without a complete physical model" means that only the lag is considered, only the system lag is taken into account, so that the PID controller has a lag, which reduces the system overshoot, but the response becomes slower.

[0078] 2.1 Simulate a vehicle inertia of 702 kg, and implement response control by increasing and decreasing the excitation by 2000 N, such as... Figure 2 , 3 As shown, the response time obtained by the method of the present invention is 75ms, which is better than the algorithms of Comparison 1 and Comparison 2, and the maximum overshoot is 109% of the target value, which is much lower than the control of Comparison 1 and Comparison 2.

[0079] 2.2 Simulate a vehicle with an inertia of 1400 kg, and implement response control with increases and decreases of 2000 N excitation. For example... Figure 4 , 5 As shown, the response time obtained by the method of the present invention is 75ms, which is better than the algorithms of Comparison 1 and Comparison 2, and the maximum overshoot is 109% of the target value, which is much lower than the control of Comparison 1 and Comparison 2.

[0080] 2.3 Simulate a vehicle inertia of 2495 kg, and implement response control with increases and decreases of 2000 N excitation. (For example...) Figure 6 , 7 As shown, the response time obtained by the method of the present invention is 75ms, which is better than the algorithms of Comparison 1 and Comparison 2, and the maximum overshoot is 1.09% of the target value, which is much lower than the control of Comparison 1 and Comparison 2.

[0081] As can be seen from the above, by simulating different vehicle inertias, the control method of the present invention can achieve high control response and small overshoot, and high control steady-state accuracy.

[0082] Furthermore, it should be understood that those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A control system based on second-order hysteresis modeling, torque feedforward, and PID closed-loop, characterized in that, Includes controller, motor, motor driver, force sensor installed at the motor load end, and encoder; The motor driver operates in torque mode or current mode. The target force is linearly converted into the torque feedforward value of the driver. At the same time, the target force is given to the second-order hysteresis system model of the motor drive system. The torque feedforward value is adjusted according to the target force calculation, and the target value of the PID is obtained by calculating through the second-order hysteresis system model of the motor drive system. The force PID control is constructed by reading the force value on the force sensor. The sum of the torque feedforward value and the PID output value is given to the driver, and the closed-loop control of the force is realized on the controller. The driver establishes vector control through an encoder.

2. The control system based on second-order hysteresis modeling, torque feedforward, and PID closed-loop as described in claim 1, characterized in that: The controller is a PID controller.

3. A control method based on second-order lag modeling, torque feedforward, and PID closed-loop, characterized in that: This is executed by a control system based on second-order hysteresis modeling and PID closed-loop as described in claim 1 or 2, comprising the following steps: Step 1, Torque Feedforward Value It is based on the input target force generate; Step 2: Construct a second-order lag system model of the motor drive system, based on the input target force. Generate the setpoint signal for the PID controller; the setpoint signal for the PID controller is the target force output. After passing through the PID controller, it becomes the output of the PID controller. Provided to the driver; Step 3: Transfer function in the complex frequency domain of the second-order lag system model of the motor drive system. The expression is: ； in: Indicates the input target force The Laplace transform of; Indicates the target force output The Laplace transform of; S is a complex variable, representing the complex frequency domain; It is the damping ratio of the motor drive system; It is the cutoff frequency of the motor drive system; It is the lag time; It is the proportionality coefficient; Step 4: The controller reads the force sensor data. After being processed by a notch filter and a first-order low-pass filter, the filtered result is obtained. The controller acquires encoder signals through a counting module and calculates the motor's speed and angular acceleration. Step 5: Based on the feedback from the force sensor The target force output after processing by the second-order hysteresis system model of the motor drive system The controller outputs F in a closed loop. PID ; Step 6: Calculate the controller output force ; Step 7: Send the output force F to the driver; Step 8: Repeat steps 1-7; Step 9: When the motor speed is less than the preset threshold, or when exiting torque mode / current mode, exit control.

4. The control method according to claim 3, characterized in that: In step 7, the output force F is sent to the driver via bus or analog signal.

Images