Dynamic torsional vibration damper based on road conditions and control method

Abstract

The invention relates to a torsional vibration dynamic damper based on road conditions and a control method, and aims to effectively suppress torsional vibration of a vehicle and improve driving stability and comfort by adjusting parameters of the damper in real time. The system comprises a road surface recognition sensor, a vehicle speed sensor, an acceleration sensor, a torsional vibration monitor and the like, and is used for acquiring road conditions, vehicle motion states and vibration information in real time. And on the basis of the information, a convolutional neural network (CNN) is adopted to identify the road surface state, and an ADRC (Active Distance Rejection Control) algorithm is combined to adjust the damping and viscosity of the magnetorheological damper. The control method adapts to different road conditions and working conditions in real time by adjusting the response of the shock absorber, it is ensured that vibration is effectively restrained when the vehicle runs on an uneven road surface, and therefore more stable driving experience is provided. According to the invention, the driving stability and comfort of the vehicle under complex road conditions can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle vibration control, and in particular to a torsional vibration dynamic damper based on road conditions and a control method. BACKGROUND

[0002] The problem of torsional vibration of vehicle transmission system has always been an important problem in the field of automotive engineering. Torsional vibration refers to the vibration caused by torque fluctuation between components such as engine, transmission, drive shaft, etc. during vehicle power transmission. This vibration propagates through the components of the transmission system during vehicle driving, affecting the stability, comfort and durability of the transmission system.

[0003] The vibration characteristics of a vehicle are closely related to road conditions. Different road surfaces (such as highways, urban roads, potholed roads, mountain roads, etc.) have different effects on the torsional vibration of the vehicle transmission system. On flat roads, the vibration of the vehicle is relatively small, and the demand for torsional vibration control is low; however, on uneven or rough roads, the transmission system of the vehicle may be subjected to severe torsional vibration impact.

[0004] CN103140698A discloses an active damping device that uses the reaction force when an auxiliary mass is driven by a driver to suppress the vibration of a damping object. The active damping device has a rigid damping control unit with a control characteristic that multiplies the displacement and operating speed of the driver by a rigidity gain k1a and a damping gain c1a and feeds back the result, thereby making the natural vibration frequency of the damping device mechanical system consistent with the vibration frequency ω of the damping object, eliminating the damping of the damping device mechanical system. In the rigid damping control unit, the rigidity gain k1a and the damping gain c1a are adjusted according to the varying vibration frequency ω, so that even if the vibration frequency varies, the vibration of the vibration frequency component can be appropriately suppressed.

[0005] CN110073436A discloses an active damping system for absorbing the vibration of a vibrating component. The damping system includes a solid acoustic wave exciter having a coupling element for coupling to the vibrating component and an electric coil for moving the coupling element by means of a coil current. The solid acoustic wave exciter is used to provide a measurement signal related to an induced voltage. In addition, a damping system controller is used to determine a theoretical current strength of the coil current based on the measurement signal, and adjust the actual current strength of the coil current to the theoretical current strength for adjusting the movement of the coupling element so that the vibration of the component is at least partially absorbed. The controller can also obtain a detection component and a control component from the measurement signal, and determine the theoretical current strength based on the detection component.

[0006] Although existing active dampers can adjust damping and stiffness, they often rely on parameters such as vehicle speed, load, and torsional vibration, lack effective road condition perception and identification functions, and thus cannot respond and adjust in real time under complex road conditions. Traditional damping techniques may not effectively cope with these changes.

[0007] In addition, on the one hand, there are differences in the understanding of those skilled in the art; on the other hand, the applicant has studied a large number of literatures and patents when making the invention, but due to the limited space, all the details and contents are not listed in detail. However, this does not mean that the present invention does not have these prior art characteristics. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art in the background art. SUMMARY

[0008] To address the deficiencies of the prior art, the present invention aims to provide a road condition-based torsional vibration dynamic damper and control method. By combining road condition information, vehicle driving state, and torsional vibration data of the transmission system, the damper force and elastic coefficient of the damper are adjusted in real time to suppress torsional vibration, improve driving comfort, vehicle stability, and prolong the service life of the transmission system, thereby solving at least some of the above technical problems. The damper of the present invention can automatically adjust according to different road conditions, thereby providing more accurate and efficient vibration control on complex or uneven roads.

[0009] The present invention discloses a road condition-based torsional vibration dynamic damper, which comprises: a road condition perception module for real-time acquisition of road condition information of the current road surface, wherein the road condition information includes unevenness, bumpiness, and / or slope change; a vehicle state monitoring module for real-time monitoring of state information of the vehicle, wherein the state information includes vehicle speed, acceleration, and / or torque fluctuation; a damper adjustment module capable of adjusting the damping coefficient and stiffness coefficient of the damper according to the road condition information and vibration data; a control unit communicatively connected with the road condition perception module, the vehicle state monitoring module, and the damper adjustment module, dynamically adjusting the working state of the damper through real-time data processing and control algorithm. The control unit can use the deviation between the system desired torque fluctuation and the actual monitored torque fluctuation as the input quantity, and the calculated damping current size as the output quantity, using an ADRC controller to adjust the current size in the coil, thereby changing the viscosity of the magnetorheological fluid.

[0010] According to a preferred embodiment, the road condition perception module includes a road recognition sensor; the vehicle state monitoring module includes a vehicle speed sensor, an acceleration sensor, and a torque sensor, wherein the torque sensor is installed at the free end of the engine crankshaft of the vehicle; the damper adjustment module is configured as a magnetorheological torsional damper mechanism, capable of adjusting the viscosity and yield stress of the magnetorheological fluid in the damper according to the control signal output by the control unit.

[0011] According to a preferred embodiment, the shock absorber adjustment module comprises: a shock absorber bottom shell with a groove in the side wall and a shock absorber end cover; several bolts and nuts for fixing; a wire conduit for placing the wire; an inertial disc with a slot arranged inside the shell, which can move independently relative to the shell; a bearing for positioning the inertial disc; a sealing ring to prevent the leakage of magnetorheological fluid; a coil for changing the strength of the magnetic field and a conductive slip ring; a segmented spline shaft for connecting the shell and the inertial disc in the transmission; a check ring for positioning, connected with the torque sensor through the spline shaft, installed on the output shaft of the vehicle power source.

[0012] According to a preferred embodiment, the control unit is built-in with a road surface recognition algorithm and a shock absorber dynamic adjustment algorithm, wherein the road surface recognition algorithm adopts a convolutional neural network algorithm for classifying the road surface and monitoring the real-time data; the shock absorber dynamic adjustment algorithm adopts an ADRC algorithm for real-time calculation of the adjustment parameters of the shock absorber most suitable for the current road conditions and torsional vibration conditions.

[0013] According to a preferred embodiment, based on the recognition result of the road surface conditions, the control unit can generate a set of pre-input signals for shock absorber control as the prior adjustment reference of active control, and then use the ADRC controller to combine the real-time vehicle state data collected by the vehicle state monitoring module to close-loop adjust the output of the shock absorber to adapt to the current road condition changes, wherein when a significant change in road surface state is monitored, the control unit can re-identify the new road surface conditions and update the pre-input signals accordingly, thereby realizing continuous optimization of the shock absorber performance under different road environments.

[0014] Preferably, the shock absorber dynamic adjustment algorithm adopts an ADRC algorithm for real-time calculation of the adjustment parameters of the shock absorber most suitable for the current road conditions and torsional vibration conditions. ADRC (Active Disturbance Rejection Control) is a kind of closed-loop control that does not depend on the accurate mathematical model of the controlled object, but estimates and compensates for the internal uncertainty and external disturbances of the system to achieve high-performance control. The control unit calculates the error (the deviation between the set value and the actual value) and adjusts the input to minimize the error.

[0015] According to a preferred embodiment, the road surface recognition algorithm adopts a convolutional neural network (CNN) algorithm for classifying the road surface and monitoring the real-time data. Convolutional Neural Network (CNN) is a deep learning model specially designed for image and time series data, mainly composed of convolutional layers, activation functions, pooling layers and fully connected layers. The convolutional layer extracts local features through convolution kernels; the activation function introduces non-linear capability; the pooling layer reduces dimensionality, improves computational efficiency and feature robustness; the fully connected layer is used for final classification or regression. The CNN road information preprocessing steps include:

[0016] Using a vehicle-mounted binocular camera or multi-camera system to collect road images, including lane lines, road signs, obstacles, etc.;

[0017] The disparity map is generated using a multi-scale pyramid matching, first constructing left and right image pyramids, starting disparity estimation from the lowest layer matching, refining layer by layer, and finally generating a high-resolution disparity map at the top layer;

[0018] The disparity map is converted into a depth map to obtain geometric information such as road boundaries, road elevations, and obstacles;

[0019] The depth map is processed in layers to generate a discrete elevation map for different road levels or obstacle classification;

[0020] The data set is divided into training set, validation set and test set, and cross-validation is used to evaluate the recognition accuracy and robustness of the model under different road conditions.

[0021] The application also discloses a control method of a road condition-based torsional vibration dynamic damper, which comprises the following steps:

[0022] S1, real-time monitoring of the state information of the vehicle: using an acceleration sensor to monitor the overall acceleration of the vehicle in real time, using a speed sensor to monitor the real-time driving speed of the vehicle, and / or using a torque sensor to monitor the torque fluctuation of the vehicle transmission system;

[0023] S2, real-time acquisition of road condition information of the current road surface: using a road surface recognition sensor to real-time acquire image data of the front road surface, and analyzing and processing road surface feature data to obtain the characteristics and classification of the road surface, and further predict possible road defects and evaluate the possible vehicle vibration caused by the road defects;

[0024] S3, adjusting the working state of the damper: generating a road surface label through the road condition information, mapping the road surface label to the expected current using a lookup table or a fuzzy rule base, adjusting the current size in the coil using PWM (10-20 kHz) according to the signal of the control unit, thereby changing the viscosity of the magneto-rheological fluid, wherein the deviation between the expected torque fluctuation of the system of different road types and the actually monitored torque fluctuation is used as an input quantity, and the resistance regulating current size calculated by the control unit is used as an output quantity.

[0025] Preferably, the road surface types can include high-frequency impact road surface, slope mutation road surface, urban expressway road surface, urban general road surface (including speed reduction belt, gravel obstacles, etc.), loose sandstone road surface, etc.

[0026] High-frequency impact road surface: With high damping and fast response as the core, the torsion system is easily excited by high-frequency excitation (>20 Hz) on uneven road surfaces such as gravel and wavy surfaces, and the risk of resonance increases. A high-damping strategy is adopted to increase the bandwidth ω0 of the ADRC controller to 100-150 rad / s to achieve fast disturbance tracking and torque response. Temperature sensor feedback is combined to compensate for temperature drift of the magnetorheological fluid (e.g., when the shear performance decreases due to temperature rise, the current is appropriately increased), and a limiting mechanism is added to prevent overshoot and reduce the risk of current saturation. Real-time monitoring of input / output shaft angular velocity fluctuations evaluates the damping effect of the magnetorheological fluid adjusted by the control current.

[0027] Slope mutation road surface: In the case of slope mutation such as on-ramp and drop road surface, the input torque changes abruptly, which can easily cause gear meshing impact and transmission shaft rebound. By combining slope prediction and pre-loading damping, the slope change trend is identified in advance, and moderate damping is pre-loaded before the torque slope rises significantly. The ADRC controller can adjust the observation gain in advance to alleviate the sudden torque disturbance caused by impact.

[0028] Urban expressway road surface: On flat and paved expressway sections, the torsional excitation is mainly low-frequency harmonic vibration. At this time, moderate damping is set to save energy and improve driving comfort. The bandwidth ω0 of the ADRC controller can be kept in the range of 30-50 rad / s, and a certain amplitude of natural vibration is allowed to avoid unnecessary energy consumption and system over-control.

[0029] Urban general road surface: In urban conditions such as speed bumps, manhole covers, and small stones, short-term torque peak impact is caused. A transient impact channel is introduced, and ω0 is increased to 200-250 rad / s. The response capability of the current changes in milliseconds (<20 ms) to detect high acceleration or high angular acceleration mutations, and the damping coefficient is temporarily increased to enhance the absorption capacity of the shock absorber to sudden conditions. By monitoring the disturbance slope in ADRC, the torsional energy is preferentially absorbed during the impact period, and then the normal control state is smoothly restored to prevent continuous over-damping.

[0030] Loose sand and gravel road surface: In conditions such as dust obstruction and limited camera, road condition recognition is prone to distortion. This road type focuses on robustness as the core strategy, and weakens the dependence on visual perception. The ADRC controller sets a conservative damping output threshold, and preferentially uses IMU, torque sensor, and wheel speed changes for disturbance estimation. The ADRC controller can appropriately reduce the ω0 value (<30 rad / s) to avoid over-damping or current mutation caused by disturbance misidentification.

[0031] Preferably, the control unit can also be built-in with consideration of the driving strategy of the driver. In complex road conditions where the road conditions frequently change, the vehicle speed is generally not more than 40 km / h (when the vehicle speed is too high, a conservative vibration suppression strategy is adopted, the ADRC controller can use a larger current to ensure comfort; when the vehicle speed is too low, the vehicle is on a certain road for a long time, no disturbance is filtered out, and the ADRC controller can adopt a damping strategy corresponding to the road type). The response time of the nonlinear ADRC controller is generally in the range of 5-300 ms, the higher the bandwidth of the ESO module, the faster the observation, but the sensitivity to noise will increase, and the response time of the control unit can be set to 50 ms after comprehensive consideration. The road recognition sensor can recognize obstacles such as potholes, damage, slope fluctuations, and speed bumps within 10 m, and the vehicle speed and acceleration information obtained by the vehicle speed sensor and the acceleration sensor are comprehensively calculated to obtain the road type switching time. When the time interval between two switches is less than 0.2 s, it is considered as a disturbance road section (the disturbance road section is a continuously changing road, and does not include speed bumps, manhole covers, and gravel obstacles as road type characteristics), and the observed v0 is regarded as z3, thereby completing the disturbance judgment. According to the calculation result, the control unit can adjust the viscosity and yield stress of the damper by adjusting the current in the coil in real time; after adjustment, the damper can dynamically respond to changes in road conditions and suppress torsional vibration.

[0032] According to a preferred embodiment, in step S2, the convolution operation can automatically identify important features in the image and extract high-level semantic information of the image. After training the deep learning model, the feature representation of different categories of images can be learned. When encountering a new image, the model can not only quickly extract its features, but also accurately judge the road type to which the image belongs by comparing with the learned category features. The CNN is trained as an intelligent road recognition algorithm in this method, which quickly identifies different road types (such as flat road, potholed road, and unpaved road) using vehicle road recognition sensor data.

[0033] According to a preferred embodiment, in step S3, the hyperbolic tangent model (Tanh) formula is used to describe the mechanical properties of the magnetorheological fluid:

[0034]

[0035] where c(H) is the damping coefficient related to the magnetic field strength, k is the stiffness coefficient, a(H) is the hysteresis characteristic proportional coefficient related to the magnetic field strength, β is the proportional coefficient related to the curve slope, δ is the variable describing the half-width of the hysteresis curve, and f0 is the bias force. c(H) and a(H) can be obtained by experimental fitting, and this model is integrated into the ADRC controller to adjust the yield stress according to the road type identified by the CNN and the torque fluctuation of the transmission system, that is, the damping force of the magnetorheological damper is adjusted by controlling the current.

[0036] According to a preferred embodiment, in step 3, the control unit can use the ADRC algorithm to calculate the adjustment parameters of the shock absorber:

[0037] Control law:

[0038]

[0039] State observer ESO update:

[0040]

[0041] β1=2ω0,

[0042]

[0043] e=y-z1,

[0044] where u(t) is the control input, i.e. the control signal applied to the system; b0 is an empirically selected controlled object control gain estimate, used to adjust the strength of the control input; k p is the proportional gain, used to adjust the response speed of the control system; y ref is the desired current (desired output), i.e. the target value that the system hopes to achieve; z1 and z2 are the tracking current and disturbance, respectively, estimated by the state observer (ESO); is the time derivative of z1 and z2, representing the rate of change over time; β1 and β2 are observer gains, used to adjust the response speed of the state observer; ω0 is the natural frequency, reflecting the inherent characteristics of the system; fal(e, α, δ) is a nonlinear function, used to enhance the robustness and adaptability of the system; e is the error signal, i.e. the difference between the actual output and the observed state; α is the degree of nonlinearity, with a value range of 0.5-0.8; δ is the fal switching threshold, with a value range of 0.01-0.1.

[0045] In the closed-loop system of "road condition recognition → current control → shock absorber adjustment", the ADRC controller receives the expected current signal generated by the road condition recognition module, combines the actual current or damping output collected by the sensor, estimates the system state and disturbance using the extended state observer (ESO), and calculates the control output acting on the magneto-rheological coil in real time through the nonlinear control law, thereby quickly adjusting the viscosity of the magneto-rheological fluid and achieving dynamic matching and precise control of the required damping under different complex road conditions.

[0046] According to the road surface type identified by the CNN, the ADRC controller can adjust the current of the magneto-rheological shock absorber according to the pre-set control strategy:

[0047] (1)High-frequency impact type road surface: the uneven road surface containing dense micro convex or continuous corrugation, when the vehicle drives on such road surface, the transmission system will suffer high frequency, small amplitude impact, which significantly affects the ride comfort and transmission smoothness. For example, washing board road, rough gravel road, etc.

[0048] (2)Slope mutation type road surface: the road slope or height changes significantly in a short distance, such as steep slope fluctuation, landing steps, etc. Such road surface will cause rapid changes of driving torque, inertia moment and load moment, which brings greater impact and torsional load fluctuation to the transmission system.

[0049] (3)Urban expressway road surface: refers to the high-grade road in the city, the road surface flatness is higher, but there are local joints, potholes or bridge connections, which still have certain excitation effect on the torsional system when driving at high speed.

[0050] (4)Urban general road surface (including speed reduction belt, gravel and other obstacles): refers to the ordinary road under typical urban working conditions, including relatively flat asphalt road, paved road surface, but there are speed reduction belt, scattered stones, manhole covers, damaged areas, etc. from time to time. The excitation of such road surface to the transmission system is mainly low frequency and large amplitude, especially when passing through the speed reduction belt and the stones, which will produce obvious impact load and torsional load fluctuation.

[0051] (5)Loose sand and gravel road surface: refers to the non-paved road surface covered with loose sand and gravel, dust, etc. When the vehicle drives on such road surface, the tire-ground adhesion characteristics are unstable, which leads to large driving force fluctuation, and the micro unevenness of the road surface also causes continuous high-frequency and small-amplitude excitation.

[0052] Compared with the prior art, the present application has the beneficial effects of:

[0053] 1. Improve the uniformity of the magnetic field and enhance the magneto-rheological effect.

[0054] Compared with the dynamically arranged coils, the uniformity is better. Most of the edges of the dynamically arranged coils are inside the magneto-rheological liquid, which causes magnetic field distortion due to the edge effect, and there is a problem of mutual influence between the coils. The side wall arranged coil can make the magnetic field uniformly distributed along the axial direction, and the coil has a lengthening compensation at both ends, so that the magnetic field distributed inside the magneto-rheological liquid is more uniform, ensuring that the magneto-rheological liquid is uniformly stressed, and the single coil does not involve the mutual influence between the coils, improving the stability and control precision of the damping adjustment, and making the damping effect more smooth and reliable.

[0055] 2. Improve the heat dissipation capacity and prolong the service life of the coil.

[0056] When the magnetorheological damper works for a long time, the coil generates heat, and high temperature can change the viscosity characteristic curve of the magnetorheological liquid, thereby reducing the adjustment capacity of the magnetorheological liquid. The coil arranged on the side wall facilitates air convection or forced heat dissipation, effectively reduces the temperature, and improves the stability and service life of the system.

[0057] 3. Pre-adjustment improves response speed.

[0058] The present application detects the road surface features in advance through the road surface identification sensor, and adjusts the pre-damping parameters before vibration occurs in combination with the data of the speed and acceleration sensors, so that the system has a pre-judgment ability. Compared with the semi-active or active damping system which depends on real-time feedback adjustment, the method reduces the response delay, so that the damping effect is more accurate and efficient. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 is the front structure schematic diagram of the damper of the present application.

[0060] Figure 2 is the sectional view of the damper of the present application.

[0061] Figure 3 is the exploded view of the damper of the present application.

[0062] Figure 4 is the logic block diagram of the control method of the present application.

[0063] Figure 5 is the control flow chart of the ADRC controller with interference judgment of the present application.

[0064] LIST OF REFERENCE NUMERALS

[0065] 1: groove; 2: damper bottom shell; 3: damper end cover; 4: bolt and nut; 5: wire pipe; 6: slotted; 7: inertia disc; 8: bearing; 9: sealing ring; 10: coil; 11: conductive slip ring; 12: spline shaft; 13: check ring; 14: bolt hole; 15: magnetorheological liquid. DETAILED DESCRIPTION

[0066] The following will be described in detail in combination with the drawings.

[0067] As Figures 1-3As shown, the present application discloses a road condition-based (magneto-rheological type) torsional vibration dynamic damper for suppressing fluctuations in a vehicle transmission system, wherein the transmission system can include an engine, a transmission, a torsional vibration dynamic damper (hereinafter referred to as a damper) and the like. The magneto-rheological type torsional vibration dynamic damper is an intelligent damping device based on magneto-rheological fluid 15 (MRF), which controls the magnetic field strength by adjusting the input current of the electromagnetic coil, thereby changing the viscosity of the magneto-rheological fluid 15 in real time to achieve dynamic adjustment of the damping force. When the current increases, the viscosity of the magneto-rheological fluid 15 increases rapidly, and the damping force increases, thereby effectively suppressing torsional vibration; when the current decreases, the magneto-rheological fluid 15 returns to a low viscosity state, making the system have a lower damping. It has fast response speed, adjustable damping and strong adaptability, and can provide accurate and rapid vibration suppression under complex working conditions, improving the stability and reliability of the system.

[0068] Preferably, the damper can include a damper adjustment module, wherein the damper adjustment module can be configured as a magneto-rheological device (i.e., a magneto-rheological type torsional vibration damping mechanism) capable of adjusting the performance of the damper in real time based on signals generated by the control unit.

[0069] Preferably, the damper adjustment module can include a damper bottom shell 2 with a groove 1 opened in the side wall for loading the coil 10, the inside of the damper bottom shell 2 is provided with a wire guide pipe 5 for placing the wire, and the damper bottom shell 2 is fixed with the torque sensor through the spline shaft 12. The damper end cover 3 and the damper bottom shell 2 have bolt holes 14 corresponding to the arrangement position, which are fixed by bolt nuts 4 to assemble the shell of the damper adjustment module.

[0070] Preferably, the damper adjustment module can include an inertial disc 7 with a slot 6 arranged in the center of the shell interior, which can move independently relative to the shell. The inertial disc 7 interacts with the shell and the entire transmission system by absorbing and releasing energy to maintain system stability. The groove 1 is responsible for cutting across the direction of the ferromagnetic particle arrangement to facilitate the work of the inertial disc 7. The gap channel between the inertial disc 7 and the shell is used to store the magneto-rheological fluid 15, wherein the magneto-rheological fluid 15 can change its viscosity properties based on the size of the magnetic field.

[0071] Preferably, the damper adjustment module can include a sealing ring 9 for preventing the magneto-rheological fluid 15 from leaking, which is tightly attached to the edge of the damper bottom shell 2 and the damper end cover 3.

[0072] Preferably, the damper adjustment module can include a conductive slip ring 11 for preventing the wire group from winding.

[0073] Preferably, the damper adjustment module can include a segmented spline shaft 12 for connecting the shell, the inertial disc 7 and other components of the transmission system in the transmission.

[0074] Preferably, the damper adjustment module can include a retaining ring 13 for positioning, wherein the retaining ring 13 is connected with the torque sensor through the spline shaft 12 and can be installed on the output shaft of the vehicle power source.

[0075] Preferably, the damper can include a road condition sensing module for real-time acquisition of road condition information such as current road unevenness, bump degree and slope change, wherein the road condition sensing module can include a road identification sensor. Optionally, the road identification sensor can be In-Sight 7000.

[0076] Preferably, the damper can include a vehicle state monitoring module for real-time monitoring of state information of the vehicle, wherein the vehicle state monitoring module can include various sensors such as a speed sensor, an acceleration sensor, a torque sensor, etc. to obtain different types of sensing data. Optionally, the acceleration sensor can be ADXL345, the speed sensor can be E6B2-CWZC6C, and the torque sensor can be Kistler 4503b.

[0077] Illustratively, the acceleration sensor ADXL345 can be installed at the center of the vehicle frame; the speed sensor E6B2-CWZC6C can be installed on the wheel; the road identification sensor In-Sight 7000 can be installed near the wheel; and the torsional vibration monitor Kistler 4503b can be fixed between the engine and the magneto-rheological torsional vibration dynamic damper by using a mounting bracket. The axis of the sensor is aligned with the transmission axis to ensure accurate measurement. When installing, ensure that the input and output shafts of the sensor rotate in the same direction, and the sensor is accurately centered to prevent errors caused by eccentricity or misalignment.

[0078] Further, the acceleration sensor can be connected to the vehicle power supply system (usually 3.3V or 5V, determined according to the sensor specification) and the vehicle ground, and connected to the ECU through the I2C interface for transmission of synchronous data; the speed sensor can be connected to the vehicle power supply system (usually 5V or 12V) and the vehicle ground, and connected to the ECU through the digital interface for input of digital signals (such as PWM signals or frequency signals), and the ECU calculates the vehicle speed according to the frequency or duty cycle of the signal received by the speed sensor, which is further used for adjustment of the damping system; the data of the road identification sensor can be connected to the ECU through the CAN bus, and the ECU analyzes the road by using a high-speed processor or an image processing algorithm to identify potholes, obstacles, etc.

[0079] According to a preferred embodiment, the magneto-rheological torsional vibration dynamic damper with the above structure can perform damping on the torque fluctuation of the transmission system in the following manner: after detecting the road surface characteristics, the road surface identification sensor generates preliminary pre-damping parameters and transmits them to the control unit; the control unit receives real-time signals from the vehicle speed sensor and the acceleration sensor, dynamically adjusts the pre-damping parameters to adapt to the current driving state; the control unit calculates and outputs the optimized final damping signal to the damper; after receiving the signal, the damper changes the magnetic field strength by adjusting the current size in the coil 10, thereby accurately controlling the viscosity of the magneto-rheological liquid 15 to achieve active control of torsional vibration. At the same time, the torque sensor detects the torque fluctuation amplitude of the transmission system in real time and compares it with the preset target range. If the torque fluctuation exceeds the set threshold, the error value is calculated and a corresponding correction signal is sent to the damper to adjust the viscosity of the magneto-rheological liquid 15 to optimize the damping effect. This process continues to circulate under the closed-loop control mechanism until the torque fluctuation amplitude meets the preset requirements, so that the system enters a stable state. When the torque fluctuation amplitude meets the set standard, the damping of the damper remains constant until the road surface identification sensor or the vehicle speed sensor and the acceleration sensor detect new environmental changes. At this time, the system restarts the adjustment process to adapt to the new working conditions, ensuring the smooth operation of the transmission system and the best damping effect.

[0080] Figure 4 is the logic block diagram of the control method of the application. The following is a step-by-step implementation example of applying a magneto-rheological torsional vibration dynamic damper for transmission system damping, i.e. the steps involved in the control method of the torsional vibration dynamic damper based on road conditions:

[0081] S1, real-time monitoring of the state information of the vehicle: using an acceleration sensor to monitor the overall acceleration of the vehicle in real time, using a vehicle speed sensor to monitor the real-time driving speed of the vehicle, and / or using a torque sensor to monitor the torque fluctuation of the vehicle transmission system;

[0082] S2, real-time acquisition of the road condition information of the current road surface: using a road surface identification sensor to acquire image data of the front road surface in real time, and analyzing and processing the road surface feature data to obtain the characteristics and classification of the road surface, and then predicting possible road defects and evaluating the vehicle vibration they may cause;

[0083] S3, adjusting the working state of the damper: generating road surface labels through road condition information, mapping the road surface labels to the expected current using a lookup table or a fuzzy rule base, and adjusting the current size in the coil 10 using PWM (10-20 kHz) according to the signal of the control unit to change the viscosity of the magneto-rheological liquid, wherein the deviation between the expected torque fluctuation of the system for different road surface types and the actual monitored torque fluctuation is used as the input quantity, and the resistance regulating current size calculated by the control unit is used as the output quantity.

[0084] Preferably, the road surface types can include high-frequency impact road surface, slope mutation road surface, urban expressway road surface, urban general road surface (including speed bump, gravel and other obstacles), loose sand and stone road surface, etc.

[0085] High-frequency impact road surface (i ref = 0.6A / 3.0kNm): With high-frequency uneven road surface such as gravel and waves, the torsion system is prone to high-frequency excitation (>20Hz), and the risk of resonance is intensified. A high-damping strategy is adopted to increase the bandwidth ω0 of the ADRC controller to 100-150 rad / s to achieve fast disturbance tracking and torque response. The temperature sensor feedback is combined to compensate for the temperature drift of the magnetorheological fluid 15 (e.g., when the shear performance decreases due to temperature rise, the current is moderately increased), and a limiting mechanism is added to prevent overshoot and reduce the risk of current saturation. The input / output shaft angular velocity fluctuation is monitored in real time to evaluate the damping effect of the control current on the magnetorheological fluid 15.

[0086] Slope mutation road surface (i ref = 1.0A / 4.5kNm): In the case of slope mutation such as on-ramp and drop road surface, the input torque changes abruptly, which can cause gear meshing impact and transmission shaft rebound. By combining slope prediction and pre-loading damping, the slope change trend is identified in advance, and moderate damping is pre-loaded before the torque slope rises significantly. The ADRC controller can adjust the observation gain in advance to alleviate the sudden torque disturbance caused by impact.

[0087] Urban expressway road surface (i ref = 0.1A / 1.2kNm): On a flat and paved expressway section, the torsional excitation is mainly low-frequency harmonic vibration. At this time, moderate damping is set to save energy and improve driving comfort. The bandwidth ω0 of the ADRC controller can be kept in the range of 30-50 rad / s, and a certain amplitude of natural vibration is allowed to avoid unnecessary energy consumption and system over-control.

[0088] Urban general road surface (i ref = 0.3A / 1.8kNm; after passing through speed bumps, stones and other obstacles, i ref = 1.6A / 7.5kNm): In urban working conditions such as speed bumps, manhole covers, and small stones, short-term torque peak impact is caused. A transient impact channel is introduced, and ω0 is increased to 200-250 rad / s. The response capability of the current is less than 20ms. When high acceleration or high angular acceleration mutation is detected, the damping coefficient is temporarily increased to enhance the absorption capacity of the shock absorber to sudden working conditions. By monitoring the disturbance slope in ADRC, the torsional energy is preferentially absorbed within the impact period, and then the normal control state is smoothly restored to prevent continuous over-damping. Loose sandstone pavement (i ref Loose sandstone pavement (i

[0089] Loose sandstone pavement (i Loose sandstone pavement (i

[0090] Loose sandstone pavement (i Figure 5 Loose sandstone pavement (i Loose sandstone pavement (i

[0091] It should be noted that the above-mentioned embodiments are only examples, and those skilled in the art can think of various solutions under the inspiration of the disclosure of the present application, and these solutions also belong to the disclosed range of the present application and fall within the protection scope of the present application. Those skilled in the art should understand that the specification and drawings of the present application are illustrative and not limiting to the claims. The protection scope of the present application is defined by the claims and their equivalents. The specification of the present application contains multiple inventive concepts, such as "preferably" or "according to a preferred embodiment", which means that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application according to each inventive concept. Throughout the text, the features introduced by "preferably" are only optional ways, and should not be understood as necessarily set, therefore the applicant reserves the right to abandon or delete the relevant preferred features at any time.

Claims

1. A dynamic vibration damper for torsional vibration based on road conditions, characterized in that, It includes: The road condition perception module is used to collect real-time road condition information, including unevenness, bumpiness and / or slope changes. The vehicle status monitoring module is used to monitor the vehicle's status information in real time, including vehicle speed, acceleration, and / or torque fluctuations. The damper adjustment module can adjust the damping coefficient and stiffness coefficient of the damper according to road condition information and vibration data; The control unit communicates with the road condition sensing module, vehicle status monitoring module, and shock absorber adjustment module, and dynamically adjusts the working state of the shock absorbers through real-time data processing and control algorithms. The control unit can use the deviation between the system's expected torque fluctuation and the actual monitored torque fluctuation as input, and the calculated resistance current as output, to use the ADRC controller to adjust the current in the coil (10), thereby changing the viscosity of the magnetorheological fluid (15).

2. The vibration damper according to claim 1, characterized in that, The road condition perception module includes a road surface recognition sensor; the vehicle status monitoring module includes a vehicle speed sensor, an acceleration sensor and a torque sensor, wherein the torque sensor is installed at the free end of the engine crankshaft of the vehicle; the damper adjustment module is configured as a magnetorheological torsional damping mechanism, which can adjust the viscosity and yield stress of the magnetorheological fluid (15) in the damper according to the control signal output by the control unit.

3. The vibration damper according to claim 1 or 2, characterized in that, The damper adjustment module includes: a damper bottom shell (2) with grooves (1) on its side wall and a damper end cap (3); several bolts and nuts (4) for fixing; a wire conduit (5) for placing wires; an inertia disk (7) with slots (6) arranged inside the shell, which can move independently relative to the shell; a bearing (8) for positioning the inertia disk (7); a sealing ring (9) to prevent leakage of magnetorheological fluid (15); a coil (10) and a conductive slip ring (11) for changing the magnetic field strength; a segmented spline shaft (12) for connecting the shell and the inertia disk (7) in the transmission; and a retaining ring (13) for positioning, which is connected to a torque sensor through the spline shaft (12) and installed on the output shaft of the vehicle power source.

4. The vibration damper according to any one of claims 1 to 3, characterized in that, The control unit incorporates a road surface recognition algorithm and a shock absorber dynamic adjustment algorithm. The road surface recognition algorithm uses a convolutional neural network algorithm to classify the road surface and monitor data in real time. The shock absorber dynamic adjustment algorithm uses the ADRC algorithm to calculate in real time the most suitable working state of the shock absorber for the current road conditions and torsional vibration.

5. The vibration damper according to any one of claims 1 to 4, characterized in that, Based on the identification results of road conditions, the control unit can generate a set of pre-input signals for vibration reduction control as a priori adjustment reference for active control. Then, using the ADRC controller combined with the real-time vehicle status data collected by the vehicle status monitoring module, the control unit performs closed-loop adjustment of the shock absorber output to adapt to changes in current road conditions. When a significant change in road conditions is detected, the control unit can re-identify the new road conditions and update the pre-input signals accordingly, thereby achieving continuous optimization of vibration reduction performance under different road environments.

6. The vibration damper according to any one of claims 1 to 5, characterized in that, The steps of the control unit employing the road surface recognition algorithm include: Use vehicle-mounted binocular cameras or multi-camera systems to capture road images; Disparity maps are generated using multi-scale pyramid matching. First, left and right image pyramids are constructed. Disparity estimation starts from the lowest layer and is refined layer by layer until a high-resolution disparity map is generated at the top layer. Convert the disparity map into a depth map to obtain geometric information; The depth map is layered to generate discrete elevation maps, which can be used for different road grades or obstacle classification. The dataset was divided into training, validation and test sets, and cross-validation was used to evaluate the recognition accuracy and robustness of the model under different road conditions.

7. A control method for a dynamic vibration damper for torsional vibration based on road conditions, characterized in that, It includes the following steps: S1. Real-time monitoring of vehicle status information: Real-time monitoring of the overall acceleration of the vehicle using an acceleration sensor, real-time driving speed of the vehicle using a vehicle speed sensor, and / or torque fluctuation of the vehicle's transmission system using a torque sensor. S2. Real-time acquisition of road condition information: Use road surface recognition sensors to acquire image data of the road surface ahead in real time, and analyze and process the road surface feature data to obtain the road surface characteristics and classification, thereby predicting possible road surface defects and assessing the vehicle vibrations they may cause. S3. Adjust the working state of the shock absorber: generate road surface labels through road condition information, apply a lookup table or fuzzy rule library to map the road surface labels to the desired current, and use PWM to adjust the current in the coil (10) according to the signal of the control unit, thereby changing the viscosity of the magnetorheological fluid. The deviation between the system's desired torque fluctuation and the actual monitored torque fluctuation for different road surface types is used as the input quantity, and the magnitude of the adjustable current calculated by the control unit is used as the output quantity.

8. The control method according to claim 7, characterized in that, In each control cycle, the ADRC controller can calculate the output based on the latest error and adjust the working state of the shock absorber. When the vehicle travels to a new road condition or undergoes dynamic changes, the sensor can provide new data in real time. The control unit recalculates and adjusts the working state of the shock absorber. The magnetic field strength changes with the input current, which causes the viscosity of the magnetorheological fluid to change, thereby dynamically adjusting the damping and ensuring that the vehicle's transmission system is always in the best damping state. The coil (10) is powered by an independent power supply and an external charging port.

9. The control method according to claim 7 or 8, characterized in that, In step S2, the CNN algorithm can automatically identify important features in the image and extract high-level semantic information of the image; after training the deep learning model, it learns the feature representations of different categories of images; When a new image is encountered, its features are quickly extracted and compared with the learned category features to accurately determine the category to which the image belongs.

10. The control method according to any one of claims 7 to 9, characterized in that, In step S3, the control unit can use the ADRC algorithm to calculate the adjustment parameters of the shock absorber: Control Law: State observer ESO update: β1=2ω0, e = y - z1, Where u(t) is the control input; b0 is the empirically selected control gain estimate of the controlled object; k p For proportional gain; y ref z1 represents the desired current; z2 represents the tracking current and z3 represents the disturbance, respectively. β1 and β2 are the time derivatives of z1 and z2, respectively; β1 and β2 are the observer gains; ω0 is the natural frequency; fal(e,α,δ) is the nonlinear function; e is the error signal; α is the degree of nonlinearity; and δ is the fal switching threshold.

Images