An optimal control system and cumulative assessment method for vibration-induced injury of vehicle occupants

Abstract

The application discloses a kind of optimization control system and cumulative evaluation method of vibration under vehicle occupant injury, including perception layer, control layer and execution layer;Perception layer is through triaxial acceleration sensor, pressure sensor array and vehicle speed sensor real-time collection vehicle vibration signal, occupant weight and driving condition;Control layer embedded main control unit, inside integrated with vibration analysis module, comfort threshold module, cumulative injury evaluation module, dynamic target calculation module, adaptive injury gain adjustment module;Execution layer includes electromagnetic variable-damping shock absorber and attitude adjustment component.The application solves the problem that the existing seat comfort control system ignores vibration cumulative injury, control response lag and semi-active damper physical constraint is not fully considered, effectively reduces the risk of cumulative injury of occupant, improves long-term health level of riding.

Description

Technical Field

[0001] This invention relates to a control and evaluation technology, and more particularly to an optimized control system and cumulative evaluation method for latent injuries of vehicle seat occupants under vibration excitation. Background Technology

[0002] Vehicle ride-related injuries are one of the core indicators for evaluating vehicle ride quality. Vibration is a key factor in inducing these injuries. Long-term accumulation of discomfort can cause hidden injuries to the human body. During vehicle operation, road bumps, engine vibrations, and rapid acceleration and deceleration can generate vibration excitations of 10-200Hz. These vibrations are transmitted through the vehicle body and suspension to the seats, causing resonance in the occupants' bodies and leading to continuous fatigue in the muscles of the waist and buttocks.

[0003] Existing solutions have several shortcomings: traditional passive damping seats have fixed damping parameters, which can only reduce vibration within a specific frequency range and cannot adapt to complex road conditions; semi-active adjustable seats can adjust damping and angle manually or according to preset modes, but the control logic relies on manual operation or fixed programs, lacking dynamic perception of real-time vibration excitation of the vehicle body, and the adjustment response is severely lagging (response time is mostly over 500ms); most seat damping control systems do not consider individual differences of occupants, and the same adjustment parameters are difficult to meet the needs of users with different weights and sitting postures; existing seats and vehicle suspension systems work independently and do not form a collaborative optimization along the vibration transmission path, resulting in limited damping effect; in addition, current control strategies only use real-time vibration amplitude as the control target and mainly adjust according to the instantaneous comfort threshold in the ISO 2631-1 standard, failing to fully consider the cumulative effect of latent damage to occupants under vibration excitation.

[0004] Existing research on human health indicates that fatigue and damage to human tissues depend not only on the instantaneous intensity of vibration but also on the cumulative exposure dose. Long-term exposure to complex vibration conditions, even with vibration amplitudes kept to a minimum at every moment, can still lead to cumulative health risks such as micro-injuries to lumbar muscles and intervertebral disc degeneration. Therefore, there is an urgent need to establish quantitative assessment methods for the cumulative effects of vibration and, based on this, to construct control systems with active damage suppression capabilities, thereby shifting vehicle ride control from providing instantaneous comfort to long-term health protection. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide an optimized control system for latent injuries to vehicle seat occupants under vibration excitation. Another purpose of this invention is to provide a cumulative evaluation method applied to this system.

[0006] Technical solution: The optimized control system for vehicle occupant injury under vibration as described in this invention includes:

[0007] The perception layer includes a triaxial accelerometer distributed along the vehicle's suspension, a pressure sensor array installed on the seat cushion and backrest, and a vehicle speed sensor connected to the vehicle's CAN FD bus. The triaxial accelerometer collects the frequency, amplitude, and direction of vehicle vibrations. The pressure sensor array identifies the occupant's real-time posture, weight, and the location of the body's center of pressure using pressure distribution data. The vehicle speed sensor acquires the vehicle's speed.

[0008] Control layer: Equipped with an embedded main control unit, integrating a vibration analysis module, a comfort threshold module, a cumulative damage assessment module, an occupant adaptive learning module, and a damage adaptive gain module. It is connected to the sensing layer via a CAN FD bus signal, receives vibration data, occupant status data, and vehicle speed data collected by the sensing layer, and generates control commands.

[0009] The execution layer includes a seat damping adjustment mechanism, a seat posture adjustment component, an active suspension coordination module, and a position feedback sensor. The seat damping adjustment mechanism dynamically adjusts the system's damping coefficient according to control commands. The seat posture adjustment component adjusts the seat cushion angle, backrest angle, and seat height. The active suspension coordination module sends coordination control signals. The position feedback sensor provides real-time position information and transmits it to the control layer.

[0010] The vibration analysis module in the control layer converts data collected by the triaxial accelerometer in the sensing layer into frequency domain signals using Fourier transform, extracting the dominant vibration frequency and amplitude. It also performs Gaussian filtering on data collected by the pressure sensor array, calculates and identifies whether the occupant's posture is forward-leaning, backward-leaning, or sideways based on the pressure distribution entropy value, and determines the occupant's dynamic body offset based on the movement trajectory of the pressure center point. The comfort threshold module in the control layer determines the vibration interference level, presets the vibration interference level based on the ISO 2631-1 standard, and optimizes it using a preset maximum threshold.

[0011] The execution method of the cumulative damage assessment module in the control layer includes the following steps:

[0012] Set sampling period Initialize the linear cumulative weighted vibrational energy dose. Cumulative injuries to occupants Total vibration exposure time ;

[0013] Obtain the current dominant vibration frequency from the vibration analysis module. and vibration amplitude The current occupant weight is collected from the pressure sensor array. ;

[0014] Formula for determining preset vibration dose value:

[0015]

[0016] in, For a moment The vibration acceleration after being weighted by human body vibration frequency was filtered using the weighting curve specified in ISO 2631-1 standard. For integration, is a dummy variable representing the instantaneous time during the vibration process. This represents the total vibration exposure time.

[0017] Introducing intermediate variables :

[0018]

[0019] Damage is proportional to a high power of the stress amplitude, that is:

[0020]

[0021] in, If is the reciprocal of the slope of the SN curve, then:

[0022]

[0023] Frequency-weighted acceleration It can be represented as:

[0024]

[0025] in, It is the measured vibration amplitude. If it is the dominant frequency of vibration, then:

[0026]

[0027] Introducing a normalized body weight factor:

[0028]

[0029] in, For scaling exponent, For reference weight, the differential of the corrected linear cumulative amount is:

[0030]

[0031]

[0032] The nonlinear damage is calculated using Euler recursion:

[0033] Where k is the number of iterations, These are frequency-weighted coefficients. The amplitude of vibration Power of 1 Weight correction factor;

[0034] Final cumulative damage for:

[0035]

[0036] in, It is a non-linear time exponent.

[0037] The assessment model for the cumulative damage is as follows:

[0038]

[0039]

[0040] in, These are frequency-weighted coefficients. The human tissue damage index, This is a standard reference weight for the human body. The bioallometric growth weight scaling index. It is a non-linear cumulative exponent.

[0041] The occupant adaptive learning module in the control layer uses historical optimization parameters as the initial adjustment benchmark to record the optimal damping coefficient and posture parameters under different weights and sitting postures.

[0042] The working principle of the adjustable seat damping mechanism in the execution layer is as follows: the PWM current signal output from the control layer is input to the electromagnetic coil of the electromagnetic variable damping shock absorber, changing the viscosity of the magnetorheological fluid in the electromagnetic variable damping shock absorber, and continuously adjusting the damping coefficient of the electromagnetic variable damping shock absorber. The relationship between the damping coefficient and the current satisfies:

[0043]

[0044] in, This refers to the damping coefficient of the electromagnetic variable damping vibration damper. This is the proportionality coefficient. For current, It is the minimum damping.

[0045] The seat posture adjustment component in the execution layer includes a backrest angle adjustment unit, a seat cushion angle adjustment unit, and a height adjustment unit. The control logic is as follows:

[0046] Backrest angle adjustment unit: automatically fine-tunes according to the rate of change in vehicle speed;

[0047] Seat cushion angle adjustment unit: When encountering bumpy roads, the front of the seat cushion tilts up to improve thigh support;

[0048] Height adjustment unit: When driving on bumpy roads, raise the seat to increase the shock absorption travel; when driving at high speed and smoothly, lower the seat to optimize the center of gravity.

[0049] The backrest angle adjustment unit, seat cushion angle adjustment unit, and height adjustment unit are all equipped with position feedback sensors to collect the actual position in real time and transmit it to the control layer.

[0050] The active suspension coordination module in the execution layer communicates with the suspension controller via the CAN FD bus and sends coordination commands. If the difference between the current vibration main frequency and the seat resonance frequency is detected to be ≤5Hz, the suspension stiffness is increased.

[0051] The cumulative assessment method for vehicle occupant injury under vibration as described in this invention includes the following steps:

[0052] Preset amplitude thresholds for vibration interference level classification:

[0053]

[0054] in, The sliding root mean square value of the acceleration amplitude corresponding to the current dominant vibration frequency within a 0.1-second window is used. This classification standard is based on the threshold of subjective and physiological response of the human body to continuous vibration in the ISO 2631-1 standard, and the vibration intensity is quantified into three levels.

[0055] Select the damage adaptive gain according to the current level, as shown in the table below. :

[0056]

[0057] Calculate the target damping coefficient:

[0058]

[0059] in, The target damping coefficient, The initial preset damping coefficient, The damage-adaptive gain is determined by the vibration level. This represents the current cumulative damage value. This represents the maximum permissible cumulative damage threshold.

[0060] Obtaining the seat's absolute velocity: Vertical acceleration is measured via an accelerometer at the bottom of the seat. The absolute velocity is obtained through numerical integration and high-pass filtering. ;

[0061] Relative speed is obtained by measuring the relative displacement between the seat and the chassis using a vehicle height sensor or displacement sensor, and then calculating the relative speed using the difference. ,in This refers to the absolute speed of the chassis.

[0062] Calculate ideal ceiling forces:

[0063]

[0064] The core objective of ceiling control is to suppress the absolute motion of the seat (sprung mass). The absolute velocity is signed, with positive for upward and negative for downward.

[0065] Applying Karnopp semi-active constraints:

[0066]

[0067] in, This is the actual control force output to the damper;

[0068] Based on the force-current characteristic curve of the electromagnetic variable damping vibration damper, the actual damping force is... Mapped to control current ; to transfer current The duty cycle is converted into a PWM signal and output to the electromagnetic variable damping vibration damper in the execution layer.

[0069] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: (1) Through the closed-loop design of sensing, analysis, adjustment and feedback, the vibration characteristics of different road conditions and information such as passenger weight and sitting posture can be captured in real time, and the seat damping and posture can be dynamically adjusted to accurately offset the vibration interference of bumps, rapid acceleration and deceleration, etc., avoiding the adaptation limitations of the traditional seat fixed mode, so that passengers can obtain soft support and effective vibration reduction under various working conditions. (2) Based on the 0.01s sampling period, the Euler recursive algorithm is used for real-time calculation. The single-step iteration only depends on the cumulative value of the previous moment. The amount of calculation is small and the response is efficient, which can meet the real-time calculation requirements of the controller. The algorithm strictly follows the ISO 2631-1 standard Wk weighted curve, combines the Miner fatigue criterion, and introduces the weight correction factor according to the biological allometric growth law to adapt to the lumbar spine stress characteristics of passengers of different body types, and solves the problems of homogeneity and low accuracy of traditional evaluation methods. At the same time, a nonlinear cumulative index is introduced to accurately restore the viscoelastic fatigue creep characteristics of human muscles and intervertebral discs, avoiding the evaluation bias of linear algorithms. Based on this precise assessment capability, the system can realize personalized vibration reduction adjustment for passengers without frequent manual adjustment, effectively suppress body displacement caused by vibration, slow down the accumulation of muscle fatigue, effectively protect the health of passengers' lumbar and cervical spine, and improve riding comfort and safety. (3) According to the ISO 2631-1 standard, the vibration intensity is classified into three levels. Combined with the real-time cumulative damage value of passengers, the adaptive matching control gain and target damping coefficient are used to realize precise adaptive control of vibration intensity and human damage in two dimensions. By combining the ceiling damping algorithm with the Karnopp semi-active constraint rule, the invalid control conditions are effectively filtered, the optimal vibration reduction control force is accurately output, and the electromagnetic damper is driven by force-current mapping and PWM signal output, resulting in precise control response and strong stability. The system is equipped with a maximum damage threshold protection mechanism. After the damage exceeds the limit, it automatically switches to the health protection mode to suppress vibration damage first. The adaptive control strategy can work in conjunction with the suspension system to reduce vibration, optimize the vibration reduction effect from the end of the vibration transmission path, and rely on intelligent automatic adjustment throughout the process to adapt to personalized riding needs. It does not require frequent manual adjustment, significantly weakens vertical vibration impact, effectively suppresses the accumulation of passenger fatigue, and greatly improves the overall riding comfort and health protection performance of the vehicle. (4) Through the synergy of system and method, this invention achieves the upgrade of control target from instantaneous comfort to long-term health, effectively reduces the risk of cumulative damage to passengers, and improves the long-term riding health level. Attached Figure Description

[0070] Figure 1 This is a system architecture diagram of the present invention.

[0071] Figure 2 This is a system workflow diagram of the present invention.

[0072] Figure 3 This is a schematic diagram of the damage assessment and control principle of the present invention. Detailed Implementation

[0073] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0074] like Figure 1 As shown, the optimized control system for vehicle occupant injury under vibration proposed in this invention specifically includes:

[0075] The perception layer includes at least three triaxial accelerometers distributed along the vehicle suspension, a distributed pressure sensor array laid on the seat cushion and backrest, and a vehicle speed sensor connected to the vehicle's CAN FD bus. The triaxial accelerometers are used to collect the frequency, amplitude, and direction of vehicle vibration in the range of 10-200Hz. The pressure sensor array contains no less than 16 pressure sensing units, which are used to identify the occupant's real-time sitting posture, weight, and the position of the body pressure center point through pressure distribution data. The vehicle speed sensor is used to obtain the vehicle's driving speed to associate driving conditions such as high speed, low speed, and rapid acceleration / deceleration.

[0076] Control layer: An embedded main control unit equipped with an ARM Cortex-A76 processor, which is connected to the sensing layer via a CANFD bus signal. It is used to receive vibration data, occupant status data and vehicle speed data collected by the sensing layer. This invention introduces the Vibration Dose Value (VDV) defined in the ISO 2631-1 standard as a quantitative indicator of cumulative damage.

[0077] The execution layer includes a seat damping adjustment mechanism, a seat posture adjustment component, and an active suspension coordination module, which are connected to the control layer via a LIN bus signal. The seat damping adjustment mechanism is used to dynamically adjust the damping coefficient of the seat damping system according to control commands. The seat posture adjustment component is used to adjust the seat cushion angle, backrest angle, and seat height. The active suspension coordination module is used to send coordination control signals to the vehicle's active suspension system to optimize the vehicle body vibration transmission path.

[0078] The system's various layers interact via bus communication, forming a dynamic, closed-loop control system. The perception layer is responsible for comprehensively collecting data on vehicle vibration, occupant status, and driving conditions. The control layer intelligently analyzes the collected data and generates targeted, optimized control commands. The execution layer adjusts seat damping, posture, and the linked suspension system according to these commands. Through real-time data acquisition, intelligent vibration analysis, multi-dimensional dynamic adjustment, and state feedback correction, the perception, control, and execution layers form a closed-loop control system that continuously counteracts the interference of vehicle vibration on passenger comfort. Figure 2 As shown, its specific implementation includes:

[0079] (1) The perception layer is the system's "sensory perception," which achieves high-precision perception of the vehicle and occupant status through multiple types of sensors, specifically including:

[0080] (11) Triaxial accelerometer

[0081] Three triaxial acceleration sensors are distributed along the vehicle's suspension, installed on the lower control arm of the front suspension, the longitudinal arm of the rear suspension, and the middle of the vehicle floor. This arrangement can simultaneously collect vibration transmission characteristics at different locations on the vehicle body, avoiding the limitations of collecting data from a single location.

[0082] Sensor parameters: sampling frequency 1kHz (ensuring the capture of high-frequency vibration details), measurement range ±16g (covering extreme vehicle vibration scenarios), resolution ≤0.001g (ensuring accurate identification of weak vibrations). Its core function is to collect vibration data in the 10-200Hz range, including vibration frequency, amplitude, and direction.

[0083] (12) Pressure sensor array

[0084] An array of 16 pressure sensors (10 on the seat cushion and 6 on the backrest) is laid on the surface of the seat cushion and backrest. Each unit can output a 0-5V analog pressure signal (corresponding to a pressure range of 0-1000N). The pressure distribution data allows for the following:

[0085] Occupant weight identification: calculated based on the sum of pressures in each unit, using the following formula:

[0086]

[0087] in, Passenger weight (unit: kg) For the first Measurement values ​​of each pressure unit (unit: N). The acceleration due to gravity is taken as 9.8 m / s². This formula quickly obtains the occupant's weight by converting the total pressure to gravity, providing a basic parameter for subsequent damping adjustment (different weights have different vibration reduction requirements).

[0088] Seated posture and pressure center point recognition: The seated posture (leaning forward, leaning backward, or leaning to the side) is determined by calculating the entropy value of the pressure distribution in each unit, and the body's center of gravity is located using the pressure center point coordinate formula.

[0089]

[0090] in, The coordinates of the pressure center point are shown in mm. For the first Installation coordinates of each pressure unit, The formula calculates the center of gravity position using a weighted average to determine the occupant's dynamic body offset, providing a basis for posture adjustment.

[0091] (13) Vehicle speed sensor

[0092] Vehicle speed signals (sampling frequency 10Hz) are acquired via the vehicle's CANFD bus and combined with acceleration sensor data to correlate driving conditions (such as high-speed stable driving, low-speed bumpy road driving, rapid acceleration / deceleration, etc.). The vehicle speed change rate formula is used to identify rapid acceleration / deceleration conditions.

[0093]

[0094] in, The rate of change of vehicle speed (unit: m / s²). Current vehicle speed The speed of the vehicle at the previous moment. The sampling interval is 0.1s. When When this occurs, the system determines it to be a rapid acceleration / deceleration condition and triggers the corresponding attitude adjustment strategy.

[0095] (2) The control layer is the "brain" of the system, equipped with an embedded main control unit (2.4GHz, supporting multi-threaded real-time processing) with an ARM Cortex-A76 processor. It integrates a vibration analysis module, a comfort threshold module, a cumulative damage assessment module, an occupant adaptive learning module, and a damage adaptive gain module. The control layer is connected to the sensing layer via a CAN FD bus signal, used to receive vibration data, occupant status data, and vehicle speed data collected by the sensing layer, and generates control commands after algorithm analysis. The specific implementation of each module includes:

[0096] (21) Vibration Analysis Module: Used to extract the dominant vibration frequency and amplitude. The original vibration data collected by the accelerometer is converted into a frequency domain signal by Fourier transform to extract the dominant vibration frequency and the energy ratio of each frequency component in the 10-200Hz frequency band; the data collected by the pressure sensor array is processed by Gaussian filtering, and the occupant's sitting posture is identified by calculating the pressure distribution entropy value to determine whether the occupant's sitting posture is forward, backward or sideways, and the dynamic offset of the occupant's body is determined based on the movement trajectory of the pressure center point.

[0097] The simplified formula for the Fourier transform is:

[0098]

[0099] in, This is a time-domain vibration acceleration signal (unit: m / s²). Frequency (unit: Hz) The amplitude of the frequency domain signal (unit: m·m·K). / Hz). Through this transformation, the system can resolve the dominant frequency in the 10-200Hz frequency band from complex vibrations. (The frequency with the highest energy percentage) provides the core basis for the main control unit to judge the vibration characteristics.

[0100] (22) Comfort Threshold Module: Used to determine the level of vibration disturbance. Includes the vibration dose value (VDV) based on the ISO 2631-1 standard, as a quantitative indicator of cumulative damage, where VDV < 0.2 m / s. 1.75 : No noticeable symptoms; VDV between 0.2 and 0.4: Noticeable symptoms; VDV between 0.4 and 0.8: Discomfort; VDV > 0.8: Health risks.

[0101] Based on the ISO 2631-1 standard (evaluation of human response to vibration), a pre-defined formula for judging the level of vibration interference is provided:

[0102]

[0103] in, The acceleration amplitude (unit: m / s²) corresponds to the dominant vibration frequency. This model quantifies the vibration intensity into three levels, providing a standard for subsequent parameter adjustments to the control strategy.

[0104] The comfort threshold module in the control layer optimizes the maximum threshold, transforming accumulated damage into adjustment commands for the suspension damping coefficient. ,in: The basic damping coefficient (corresponding to the minimum damping state) is given in the damper product manual; To achieve damage adaptive gain, this implementation uses simulation calibration. The maximum permissible cumulative damage threshold is determined according to ISO 2631-1 standard. time This confirmation process determines the adjustment command for the suspension damping coefficient. .

[0105] (23) Cumulative Damage Assessment Module: This module is used to perform the cumulative damage assessment method and output the current cumulative damage value. The cumulative damage coefficient is calculated using the amplitude and magnitude output by the vibration analysis module. The cumulative damage value is combined with the damage adaptive gain through the damping benchmark calculation to calculate the target damping coefficient required at present. This coefficient increases with the increase of the cumulative damage level and outputs the damage factor W(t).

[0106] (24) Occupant Adaptive Learning Module: Based on historical optimization parameters as the initial adjustment benchmark. Used to record the optimal damping coefficient and posture parameters for different weights (50-120kg) and different sitting postures (e.g., the optimal damping coefficient for a 60kg occupant under bumpy road conditions is 5N·s / m). When the same occupant is identified by the pressure sensor (based on weight and sitting posture characteristics matching), the historical parameters are automatically called as the initial adjustment benchmark, and then combined with real-time data for fine-tuning, shortening the response time (30% faster than the first adjustment).

[0107] (25) Damage Adaptive Gain Module: Based on the ISO 2631-1 standard (evaluation of human response to vibration), the amplitude threshold of vibration interference level is preset, and the system dynamically adjusts the adaptive gain parameters according to the vibration interference level.

[0108] In step (23), the specific execution method of the cumulative damage assessment module includes the following steps:

[0109] (231) Initialization settings: Set the sampling period Initialize the linear cumulative weighted vibrational energy dose. Cumulative injuries to occupants Total vibration exposure time .

[0110] (232) Real-time data acquisition: Obtain the current dominant vibration frequency from the vibration analysis module. and vibration amplitude (Unit: m / s²) Current occupant weight collected from pressure sensor array (Unit: kg)

[0111] (233) Vibration reference parameters preset: Based on ISO 2631-1 standard (evaluation of human response to vibration), the preset formula for judging vibration dose value is as follows:

[0112]

[0113] in, For a moment Vibration acceleration (unit: m / s²) after being weighted by human body vibration frequency. The weighting filter adopts the method specified in ISO 2631-1. The weighted curve is specifically designed to evaluate the vertical vibration transmitted from the seat surface to the human body, and can fit the human body's vibration perception and damage sensitivity characteristics at different frequency bands. For integration, is a dummy variable, representing the instantaneous time during the vibration process; The total vibration exposure time is defined as the integration interval from the vibration initiation time 0 to the current observation time. First, take the fourth power of the instantaneous weighted acceleration. Then, the total cumulative vibration damage is obtained by integrating over the entire process time, and finally, the vibration dose value is obtained by taking the fourth root of the whole. The fourth power form of VDV amplifies the contribution of large-amplitude vibrations, which is more in line with the fatigue accumulation characteristics of human tissues.

[0114] However, VDV is a holistic evaluation metric and is not suitable for direct use in real-time control. To facilitate damage accumulation and summarization, an intermediate variable is introduced. :

[0115]

[0116] This can be broken down into a form that can be accumulated second by second and correlated with occupant characteristics. In the field of materials fatigue, Miner's linear accumulation theory states that damage is proportional to a high power of the stress amplitude, that is:

[0117]

[0118] in It is the reciprocal of the slope of the SN curve. For metallic materials, it is usually taken as 3, while for biological tissues (such as intervertebral discs and muscles), related studies have shown that damage is correlated with the 2nd to 4th powers of the load. Therefore:

[0119]

[0120] According to ISO 2631-1, frequency-weighted acceleration It can be expressed as the measured vibration amplitude. With frequency weighting coefficients The product of:

[0121]

[0122] in It is the measured vibration amplitude. It is the dominant vibrational frequency. Therefore...

[0123]

[0124] (234) Damage Correction: The force on the lumbar spine is directly proportional to the mass of the upper body, and body weight is a direct reflection of mass. Under the same vibration, the heavier the occupant, the greater the load on the lumbar spine and the higher the risk of injury. Considering the occupant's weight... The impact on the stress on the lumbar spine. Based on the allometric growth law in biomechanics, the cross-sectional area of ​​an organ is related to its body weight. It is directly proportional to body weight, so the compressive stress on the lumbar spine is related to body weight. It is directly proportional. Therefore, a normalized weight factor is introduced:

[0125]

[0126] in, For scaling exponent, . For reference weight Assuming a standard body weight, the differential of the corrected linear cumulative mass is:

[0127]

[0128]

[0129] (235) Nonlinear Damage Conversion: Biological tissues (such as muscles and intervertebral discs) are viscoelastic, and their fatigue damage does not increase completely linearly with exposure time. For example, continuous vibration for 1 hour may result in different cumulative damage than vibration for two 30-minute intervals (there may be "creep" or "recovery" effects). Introducing a time index... This causes the damage to increase non-linearly with exposure time. Here It is a constant between 0.5 and 1, and this implementation takes... This indicates that the rate of damage accumulation slows down slightly with exposure time (i.e., the incremental damage caused by the same vibration in the later stage is slightly less than that in the earlier stage).

[0130] In the calculation, we first use Euler's recurrence relation, that is...

[0131]

[0132] in:

[0133] k: Number of iterations

[0134] Frequency weighting coefficients (based on ISO 2631-1 Wk curve table)

[0135] : vibration amplitude power ( )

[0136] Weight Correction Factor

[0137] =0.01: Sampling period.

[0138] Define final cumulative damage for:

[0139]

[0140] in It is a nonlinear time exponent, reflecting the nonlinear creep characteristics of biological tissue fatigue.

[0141] (236) The cumulative damage assessment model is as follows:

[0142]

[0143]

[0144] in, ISO 2631-1 standard Frequency weighting coefficients (dimensionless); Human tissue damage index (referencing Miner's fatigue accumulation criterion); Standard reference weight for human body; Allometric growth weight scaling index; Nonlinear cumulative index, characterizing the fatigue creep decay characteristics of human tissue.

[0145] In practical digital controllers, the sampling period is used as the basis for calculation. Real-time computation using Discrete Euler recursion:

[0146]

[0147]

[0148] Initial conditions: , This recursion relies only on the accumulated value from the previous time step at each step, resulting in high computational efficiency.

[0149] In step (25), such as Figure 3 As shown, the specific implementation method of damage adaptive control includes the following steps:

[0150] (251) Based on the ISO 2631-1 standard (evaluation of human response to vibration), preset amplitude thresholds for vibration interference level classification:

[0151]

[0152] in, The current dominant vibration frequency The corresponding acceleration amplitude is the root mean square value (m / s²) within a 0.1-second window. This classification standard is based on the subjective and physiological response thresholds of the human body to continuous vibration in ISO 2631-1, quantifying vibration intensity into three levels to provide a grading basis for the selection of damage adaptive gain K in subsequent control strategies.

[0153] (252) Select the damage adaptive gain according to the current level, as shown in the table below. :

[0154]

[0155] (253) Calculate the target damping coefficient

[0156]

[0157] in:

[0158] Target damping coefficient (unit: N·s / m)

[0159] : Foundation damping coefficient (unit: N·s / m), in this embodiment, we take

[0160] Damage adaptive gain, determined by vibration level.

[0161] : Current cumulative damage value, the current cumulative damage output in step 4.5

[0162] Maximum permissible cumulative damage threshold

[0163] when When this happens, the system enters health protection mode, forcing the target to be the minimum vibration amplitude (i.e., ), preferentially suppress damage.

[0164] (254) Obtaining the absolute velocity of the seat: Measure the vertical acceleration using an acceleration sensor at the bottom of the seat. The absolute velocity is obtained through numerical integration and high-pass filtering. (Upward is positive);

[0165] (255) Obtaining relative speed: The relative displacement between the seat and the chassis is measured by a vehicle height sensor or displacement sensor, and the relative speed is obtained by differential measurement. ,in This represents the absolute speed of the chassis (positive for upward movement).

[0166] (256) Calculate the ideal ceiling force:

[0167]

[0168] The core objective of ceiling control is to suppress the absolute motion of the seat (sprung mass). The absolute velocity is signed, with positive for upward and negative for downward.

[0169] (257) Applying Karnopp semi-active constraints:

[0170]

[0171] in, This is the actual control force output to the damper;

[0172] (258) Force-current mapping: Since the relationship between the damping force and the current of the electromagnetic variable damper is determined by the microstructure of the magnetorheological fluid, it has complex characteristics such as nonlinearity, hysteresis and temperature sensitivity, and is difficult to describe accurately with a single theoretical formula. Therefore, this invention adopts a calibration method to establish the force-current mapping relationship.

[0173] The calibration experiment was conducted on an electro-hydraulic servo excitation table: the damper was fixed to the excitation table, a constant piston speed (e.g., 0.1 m / s) and excitation amplitude (e.g., ±10 mm) were set, and control currents of different intensities were applied. (Increase the current gradually from 0A to the saturation current, with a step size of 0.1A or 0.2A). Let be the control current at the i-th calibration point. Measure the steady-state damping force at each current value. A series of data points were obtained. , The damping force at the i-th calibration point is obtained from measurements taken during bench testing. (Draw...) and The characteristic curve typically exhibits two segments: in the low current range (e.g., ... Within the range of high current, the damping force increases approximately linearly with the current; within the range of high current, the increase in damping force slows down and gradually approaches saturation.

[0174] Within the aforementioned linear working interval, the least squares method is used to perform linear fitting on the data points, yielding:

[0175]

[0176] in:

[0177] slope This is the force-current coefficient (unit: N / A).

[0178] for The passive damping force at that time (generated by the passive valve system inside the damper).

[0179] I represents the control current.

[0180] During reverse mapping, based on the target damping force Calculate the required control current:

[0181]

[0182] (259) The calculated current The duty cycle is converted into a PWM signal and output to the electromagnetic damper in the execution layer. To accommodate temperature changes and hysteresis characteristics, this mapping relationship is stored in the form of a piecewise linear lookup table, which is retrieved in real time by the control layer.

[0183] (3) The execution layer is the "executive organ" of the system, which realizes the dynamic adjustment of the seat state through the coordinated action of multiple components, including the seat damping adjustment mechanism, the seat posture adjustment component, the active suspension coordination module, and the position / current feedback sensor, as follows:

[0184] (31) Adjustable seat damping mechanism

[0185] An electromagnetic variable damping vibration damper is employed, consisting of a cylinder, piston, electromagnetic coil, and magnetorheological fluid. Its working principle is as follows: a 0.5-5A PWM current signal is output from the control layer and input to the electromagnetic coil of the electromagnetic variable damping vibration damper, changing the viscosity of the magnetorheological fluid within the damper (magnetic field strength is positively correlated with current), thereby continuously adjusting the damping coefficient of the electromagnetic variable damping vibration damper. The relationship between the damping coefficient and the current satisfies:

[0186]

[0187] in, The damping coefficient of the electromagnetic variable damping vibration damper (unit: N·s / m). This is the proportionality coefficient. Current (unit: A) This represents the minimum damping. Through this relationship, the system can achieve stepless adjustment of the damping coefficient within the range of 0.1-10 N·s / m, with a response time ≤50ms (the switching time from minimum to maximum damping), ensuring rapid suppression of high-frequency vibrations.

[0188] (32) Seat posture adjustment assembly

[0189] Three adjustment units are driven by servo motors. The specific parameters and control logic are as follows:

[0190] Backrest angle adjustment: Range 100°-160° (corresponding to upright to semi-reclined), adjustment accuracy ±0.5°. Adjustment will adjust according to the rate of change of vehicle speed during rapid acceleration / deceleration. Automatic fine-tuning: rapid acceleration ( When decelerating rapidly, lean the backrest forward 3°-5° (to reduce backward leaning); When leaning back, tilt the backrest back 3°-5° (to reduce forward leaning).

[0191] Seat cushion angle adjustment: range -5° (front tilt down) +5° (front tilt up), accuracy ±0.3°. On bumpy roads (vibration frequency 10-50Hz and amplitude >0.5g), the front of the seat cushion tilts up 2°-3° to improve thigh support.

[0192] Height adjustment: range 0-100mm, accuracy ±1mm. When driving on bumpy roads, raise the seat by 10-30mm to increase the shock absorption travel; when driving at high speed and smoothly, lower the seat by 10mm to optimize the center of gravity.

[0193] Each adjustment unit is equipped with a position feedback sensor (accuracy ±0.1mm) to collect the actual position in real time and transmit it to the control layer. The feedback correction formula is then used.

[0194] Actual target

[0195]

[0196] in, For correction, target For instruction location, actual To provide feedback position, the adjustment accuracy is corrected (correction error ≤1%).

[0197] (33) Active suspension coordination module

[0198] When the vehicle is equipped with active suspension, this module communicates with the suspension controller via the CAN bus and sends coordination commands (including vibration frequency). (Direction and suggested adjustment amount). If detected If the difference between the frequency and the seat resonance frequency (preset 60-80Hz) is ≤5Hz (which easily causes resonance), then the suspension stiffness is increased by 10%-20%, as shown in the formula:

[0199]

[0200] in, To adjust the rear suspension stiffness, The original stiffness is maintained. By changing the suspension stiffness, the vibration frequency of the vehicle body is shifted, avoiding resonance with the seats and reducing the vibration energy transmitted to the occupants.

[0201] (34) Position or current feedback sensor

[0202] Position feedback sensors (accuracy ±0.1mm) and current feedback sensors are arranged at the drive end of the seat posture adjustment component to collect the actual position and drive current of the adjustment mechanism in real time, and transmit the feedback data to the control layer to form a closed loop for adjustment accuracy correction (correction error ≤1%).

[0203] The specific workflow of this system is as follows:

[0204] (1) Data Acquisition Stage: This stage is executed collaboratively by various sensors in the perception layer, providing basic data support for subsequent analysis. Three-axis accelerometers, located on the lower control arm of the front suspension, the longitudinal arm of the rear suspension, and the center of the vehicle floor, collect vibration signals from different locations on the vehicle body at a sampling frequency of 1kHz, obtaining the frequency, amplitude, and direction information of vehicle vibration. Simultaneously, an array of pressure sensors (containing no fewer than 16 pressure sensing units) installed on the seat cushion and backrest collects real-time pressure distribution data between the occupant and the seat contact surface. The occupant's weight is identified by summing the pressure values ​​of each unit, the current sitting posture (forward, backward, or sideways) is determined by the pressure distribution pattern, and the dynamic offset of the occupant's body is monitored by the movement trajectory of the pressure center point. Furthermore, the vehicle speed sensor acquires the current driving speed signal via the vehicle's CAN FD bus, used to identify different driving conditions such as high-speed cruising, low-speed bumps, rapid acceleration, or rapid deceleration. Vibration data, occupant status data, and vehicle speed data are collected and transmitted to the control layer via the CAN FD bus, with a transmission delay of no more than 10 milliseconds to ensure real-time data transmission.

[0205] (2) Analysis and decision-making stage: This stage is completed collaboratively by various functional modules within the control layer and is the core calculation and decision-making link of the system.

[0206] After receiving the data from the sensing layer, the vibration analysis module first performs a Fourier transform on the original time-domain vibration signal collected by the accelerometer, converting it into a frequency-domain signal. It then searches for the frequency component with the highest energy proportion in the 10 to 200 Hz frequency band, thereby resolving the dominant frequency of the current vibration and its corresponding amplitude.

[0207] After vibration analysis is completed, the cumulative damage assessment module uses the frequency weighting curve in ISO 2631-1 to look up the frequency weighting coefficients from the current dominant vibration frequency, and performs biomechanical corrections based on occupant weight. It then updates the cumulative damage value recursively and obtains the current cumulative damage level through nonlinear transformation. Simultaneously, the occupant adaptive learning module uses the weight and posture characteristics identified by the pressure sensor to determine if the occupant is a previously recorded occupant. If so, it automatically retrieves the occupant's historical optimization parameters as the initial adjustment benchmark.

[0208] The control layer performs two tasks and makes judgments simultaneously: On the one hand, the vibration level judgment module classifies the vibration interference into three levels—slight, moderate, or severe—based on the short-time sliding root mean square value of the current vibration amplitude, according to the ISO 2631-1 standard, and outputs the level results to the gain selection module and the human-machine interface respectively; on the other hand, the control layer determines whether it is a rapid acceleration or deceleration condition based on the vehicle speed change rate, and determines whether it is a bumpy road condition or close to the seat resonance frequency based on the vibration dominant frequency and amplitude.

[0209] The gain selection module selects the corresponding damage adaptive gain based on the current vibration level. Then, the damping benchmark calculation module combines the accumulated damage value with the damage adaptive gain to calculate the required target damping coefficient, which increases with the level of accumulated damage. After the target damping coefficient is determined, the ceiling force calculation module obtains the absolute velocity of the seat by integrating the seat acceleration. It then multiplies the target damping coefficient by the seat's absolute velocity to calculate the ideal damping force. However, since the electromagnetic damper is a semi-active element that can only consume energy and cannot actively output energy, the Karnopp constraint module corrects the ideal damping force based on the directional relationship between the seat's absolute velocity and the relative velocity between the seat and the chassis, ensuring that the output damping force can be physically achieved by the damper.

[0210] Simultaneously, the control layer generates attitude adjustment commands and suspension coordination commands based on vibration level identification results and vehicle speed data: under bumpy road conditions, commands to raise seat height and adjust seat cushion angle are generated to increase damping travel; under rapid acceleration or deceleration conditions, commands to fine-tune backrest angle are generated to reduce occupant body offset; when the dominant vibration frequency approaches the seat resonant frequency, active suspension stiffness adjustment commands are generated to change the vibration transmission path. Finally, the force-current mapping module converts the corrected damping force into control current for the electromagnetic coil, thereby generating a PWM drive signal, which, along with the attitude adjustment commands and suspension coordination commands, is output to the execution layer.

[0211] (3) Execution feedback stage: In this stage, each mechanism of the execution layer receives instructions and executes actions, while forming a closed-loop correction through sensor feedback.

[0212] After receiving commands from the control layer via the LIN bus, the electromagnetic coil of the electromagnetic variable damper adjusts the excitation current according to the PWM signal, changing the viscosity of the magnetorheological fluid at the damping orifice, thereby achieving continuous stepless adjustment of the damping coefficient. Simultaneously, the servo motors in the seat posture adjustment assembly drive the backrest angle adjustment unit, seat cushion angle adjustment unit, and height adjustment unit to perform corresponding angle or height adjustments. The active suspension coordination module then sends suspension stiffness adjustment commands to the vehicle's active suspension controller via the CAN FD bus, optimizing the vibration transmission path of the vehicle body.

[0213] While executing the action, position feedback sensors located at the drive end of the attitude adjustment component monitor the actual position of each adjustment mechanism in real time, and current feedback sensors collect the actual drive current of the damper's electromagnetic coil in real time. These feedback data are transmitted back to the control layer in real time. The control layer uses this data to determine the deviation between the actual execution state and the target command, and corrects the PWM output and adjustment amount to form a high-precision closed-loop control, with the correction error controlled within one percent.

[0214] (4) Dynamic adaptation stage: The system executes the above data acquisition, analysis and decision-making, execution and feedback process in a cycle of 10 milliseconds, continuously sensing changes in vibration conditions and dynamically adapting to control strategies.

[0215] During the loop, the control layer employs a dual-timescale strategy: outer loop tasks such as cumulative damage assessment and damping benchmark calculation are updated with a period of 0.1 seconds because cumulative damage changes relatively slowly; while inner loop tasks such as ceiling force calculation and Karnopp constraints are executed with a period of 5 milliseconds to ensure a rapid response to instantaneous vibrations.

[0216] When a vehicle switches from a smooth road surface to a bumpy road surface, the sensing layer captures the changes in vibration amplitude and dominant frequency within the next sampling cycle (no more than 10 milliseconds); the vibration level judgment module completes the level switching within 0.1 seconds; the gain selection module synchronously updates the damage adaptive gain; the damping benchmark calculation module completes the target damping coefficient update within the current outer loop cycle; and the electromagnetic damper in the execution layer completes the full adjustment of the damping coefficient within 50 milliseconds. Considering the response time of each component, the system can complete the full adjustment from vibration sensing to damping and attitude within 200 milliseconds, achieving rapid dynamic adaptation to different road conditions.

[0217] This invention achieves dynamic optimization of seat damage under automotive vibration conditions through multi-dimensional perception, intelligent algorithm optimization, and the synergy of rapid actuators, significantly improving the riding experience under different road conditions and occupant characteristics. Specifically, this is reflected in the following aspects:

[0218] (1) Comprehensive improvement of riding injury, covering all vibration conditions in all scenarios.

[0219] Traditional vehicle seats often rely on fixed modes or manual adjustment for vibration damping and posture control, making it difficult to cope with complex and varied vibration conditions. For example, on bumpy roads, a seat with fixed damping will directly transmit most of the vibration to the occupant, causing continuous vibration in the buttocks and lower back. During rapid acceleration or deceleration, the occupant's body is prone to leaning forward or backward due to inertia, leading to tension in the lower back muscles. This system, however, uses multi-dimensional sensors (a three-axis accelerometer, a pressure sensor array, and a vehicle speed sensor) in the perception layer to capture vibration frequency, amplitude, and direction in real time within the 10-200Hz range, as well as details such as the occupant's weight, posture, and center of gravity shift. Then, the intelligent algorithm in the control layer analyzes the vibration characteristics (such as extracting the dominant frequency through Fourier transform), combines the ISO 2631-1 damage standard to determine the vibration interference level, and finally drives the dynamic adjustment in the execution layer.

[0220] Specifically, on bumpy roads (vibration frequency 10-50Hz and amplitude >0.5g), the system quickly increases the seat damping coefficient (0.1-10N·s / m stepless adjustment) through an electromagnetic damping adjustment mechanism, while raising the seat by 10-30mm to increase the damping travel. Combined with the seat cushion tilting up 2°-3° to improve thigh support, the vibration energy transmitted to the occupant is reduced by more than 40%. During rapid acceleration / deceleration (vehicle speed change rate >2m / s²), the backrest angle is slightly adjusted 3°-5° towards the occupant's body to reduce the body's offset due to inertia and avoid hard friction between the waist and the backrest. When driving at high speed and smoothly, the system lowers the seat height by 10mm to optimize the center of gravity and reduces the damping coefficient (20% reduction in comfort mode), allowing the occupant to experience softer support. This "on-demand adjustment" mode completely breaks the limitations of the traditional "one-size-fits-all" approach to seats, allowing occupants to receive appropriate support and vibration reduction in any vibration scenario, significantly reducing the discomfort caused by vibration.

[0221] (2) Achieve personalized adaptation to meet the injury needs of different occupants.

[0222] Occupants of different weights and sitting postures have significantly different needs for seat damping: for example, heavier occupants (over 100kg) require a higher damping coefficient to prevent the seat from sinking excessively, while lighter occupants (under 50kg) need softer support; occupants in a forward-leaning posture (such as while driving) need a closer fit to their lower back, while occupants in a reclining posture (such as while resting) need more relaxation space at a greater backrest angle. Traditional seats cannot specifically adapt to these differences, often resulting in some occupants feeling uncomfortable.

[0223] This system accurately identifies occupant weight using a distributed pressure sensor array (16 sensing units) (via the total pressure formula). (Calculation) and sitting posture (through the coordinates of the center of pressure) The system determines the optimal adjustment parameters for occupants with different characteristics (e.g., the optimal damping coefficient for a 60kg, forward-leaning occupant on bumpy roads is 5 N·s / m, and the backrest angle is 110°) using the "occupant adaptive learning module" in the control layer. When the same occupant rides again, the system automatically calls up historical parameters as the initial benchmark and then fine-tunes them using real-time vibration data to make the adjustment more suitable for individual habits. This personalized adaptation capability transforms the seat from a "standardized product" into a "customized service," allowing occupants, whether adults, children (requiring a car seat), or users of different body types, to obtain an injury experience that meets their individual needs.

[0224] (3) Enhance riding safety and reduce potential risks caused by vibration and inertia.

[0225] Vehicle vibrations and inertia during rapid acceleration / deceleration not only cause damage but can also pose safety hazards. For example, on continuously bumpy roads, the occupant's body may slide relative to the seat due to continuous vibration, leading to imbalance. During sudden braking, if the backrest angle is not appropriate, the occupant's upper body may lean excessively forward, potentially colliding with the steering wheel or the backrest of the front seat. Long-term exposure to high-frequency vibration environments can cause fatigue in the occupant's lumbar and cervical spine due to repeated stress, increasing the risk of chronic injury.

[0226] This system mitigates these risks through multiple mechanisms: Firstly, it monitors the body's center of gravity shift in real time using pressure sensors. When it detects a tendency for the occupant to slide due to vibration, it immediately increases friction by adjusting the seat cushion angle (front end tilting upwards) and improves the damping coefficient to reduce seat sway, keeping the body sliding amplitude within 5mm. Secondly, during rapid acceleration / deceleration, the dynamic fine-tuning of the backrest angle (3°-5° forward / backward tilt) can offset some of the inertial force through the backrest's support, reducing the amplitude of forward / backward tilt and minimizing the risk of collision. Furthermore, the system's effective vibration suppression (controlling the vibration amplitude in the 10-200Hz frequency band within the "slight" level of the ISO 2631-1 standard, i.e., <0.315m / s²) significantly reduces the risk of muscle fatigue and joint damage during long-term sitting, shifting from "passive protection" to "active health protection."

[0227] (4) Improve system response speed and adjustment accuracy to avoid discomfort caused by lag.

[0228] Traditional seat adjustments are mostly mechanical, manual, or semi-automatic. They are slow (it takes 1-2 seconds from the time the command is issued to the completion of the adjustment) and have low precision (angle adjustment error of ±2° or more). They are difficult to cope with rapidly changing vibration conditions. For example, when a vehicle suddenly drives into a pothole, the lag in the adjustment of a traditional seat will cause the occupant to experience a strong vibration before feeling the adjustment effect, which will only exacerbate the discomfort.

[0229] This system achieves "millisecond-level response + high-precision adjustment" through the collaborative design of hardware and algorithms: the triaxial accelerometer in the sensing layer captures vibration details at a sampling frequency of 1kHz and transmits the data to the control layer via CAN FD bus (transmission rate 8Mbps) with a delay of ≤10ms; the ARM Cortex-A76 processor (2.4GHz) in the control layer can complete vibration analysis and command generation within 20ms; the electromagnetic damping adjustment mechanism in the execution layer has a response time of ≤50ms (switching from minimum to maximum damping), and the seat posture adjustment unit has an angle accuracy of ±0.3° and a height accuracy of ±1mm, and forms a closed-loop correction through a position feedback sensor (error ≤1%). This rapid response capability allows the system to complete the adjustment before the vibration is transmitted to the occupant's body (the human body's perception of vibration is delayed by about 100ms), truly achieving "predictive vibration reduction" and avoiding the "lagging discomfort" of traditional adjustments.

[0230] (5) Synergize with vehicle systems to enhance overall vibration reduction effect

[0231] Seat damage depends not only on the seat itself, but also on the transmission path of vehicle body vibrations. If the vehicle body vibrates strongly due to insufficient suspension stiffness, adjusting the seat alone is unlikely to completely eliminate discomfort. Traditional seats and suspension systems often work independently and cannot work in synergy, resulting in reduced vibration damping.

[0232] The "active suspension coordination module" of this system breaks this independent state: when the control layer detects that the vehicle's main vibration frequency (e.g., 65Hz) is close to the seat resonance frequency (60-80Hz), it will send a coordination command to the active suspension controller via the CAN bus to control the suspension stiffness to increase by 10%-20%. This mechanism alters the vehicle's vibration frequency characteristics to prevent resonance. On extremely bumpy roads (vibration amplitude > 0.63 m / s²), the suspension and seat damping adjust synchronously (increasing suspension stiffness and increasing seat damping), suppressing vibration from both the "vibration source" and the "transmission end," resulting in a vibration reduction effect more than 30% higher than that of adjusting only the seat. This "vehicle-level collaborative" mechanism elevates damage optimization from a "local improvement" to a "systems engineering" approach, significantly improving the overall ride quality of the vehicle.

[0233] (6) Balancing automation and personalized control to enhance user experience.

[0234] Traditional seat damage adjustment often requires users to manually operate multiple buttons (such as damping adjustment, angle adjustment, and height adjustment), which is cumbersome and difficult to accurately match the working conditions; while fully automated systems may ignore the user's subjective preferences (such as some users who prefer slightly stiffer damping to enhance road feel when driving on mountain roads).

[0235] In summary, this invention achieves comprehensive improvements in terms of damage, personalization, safety, response speed, system synergy, and user experience through multi-dimensional perception, intelligent algorithm optimization, rapid execution adjustment, and whole-vehicle system collaboration. It effectively solves the problems of insufficient damage, poor adaptability, and delayed response of traditional vehicle seats under vibration conditions, providing an efficient and reliable technical solution for optimizing vehicle passenger injury.

Claims

1. An optimal control system for reducing occupant injury in a vehicle subjected to vibration, characterized by comprising: include: The perception layer includes a triaxial accelerometer distributed along the vehicle's suspension, a pressure sensor array installed on the seat cushion and backrest, and a vehicle speed sensor connected to the vehicle's CAN FD bus. The triaxial accelerometer collects the frequency, amplitude, and direction of vehicle vibrations. The pressure sensor array identifies the occupant's real-time posture, weight, and the location of the body's center of pressure using pressure distribution data. The vehicle speed sensor acquires the vehicle's speed. Control layer: Equipped with an embedded main control unit, integrating a vibration analysis module, a comfort threshold module, a cumulative damage assessment module, an occupant adaptive learning module, and a damage adaptive gain module. It is connected to the sensing layer via a CAN FD bus signal, receives vibration data, occupant status data, and vehicle speed data collected by the sensing layer, and generates control commands. The execution layer includes a seat damping adjustment mechanism, a seat posture adjustment component, an active suspension coordination module, and a position feedback sensor; the seat damping adjustment mechanism dynamically adjusts the system's damping coefficient according to control commands; the seat posture adjustment component adjusts the seat cushion angle, backrest angle, and seat height. The active suspension coordination module sends a coordination control signal; the position feedback sensor provides real-time position information and transmits it to the control layer.

2. The optimal control system of claim 1, wherein, The vibration analysis module in the control layer converts the data collected by the triaxial accelerometer in the sensing layer into a frequency domain signal by performing Fourier transform, and extracts the vibration main frequency and amplitude; and performs Gaussian filtering on the data collected by the pressure sensor array, and uses the pressure distribution entropy value to identify whether the occupant's sitting posture is forward, backward or sideways, and judges the dynamic offset of the occupant's body based on the movement trajectory of the pressure center point.

3. The optimal control system of claim 1, wherein, The comfort threshold module in the control layer is used to determine the vibration interference level. It presets the vibration interference level based on the ISO 2631-1 standard and optimizes it using the preset maximum threshold.

4. The optimal control system of claim 1, wherein, The execution method of the cumulative damage assessment module in the control layer includes the following steps: Setting a sampling period , initializing a weighted vibration energy dose linear accumulation , occupant cumulative injury , total vibration exposure duration ; Obtain the current dominant vibration frequency from the vibration analysis module. and vibration amplitude The current occupant weight is collected from the pressure sensor array. ; Formula for determining preset vibration dose value: in, For a moment The vibration acceleration after being weighted by human body vibration frequency was filtered using the weighting curve specified in ISO 2631-1 standard. For integration, is a dummy variable representing the instantaneous time during the vibration process. This represents the total vibration exposure time. Introducing an intermediate variable : Damage is proportional to a high power of the stress amplitude, that is: wherein is the inverse of the slope of the S-N curve, then: Frequency-weighted acceleration may be expressed as: in, It is the measured vibration amplitude. If it is the dominant frequency of vibration, then: Introducing a normalized body weight factor: wherein, is the scaling exponent, is the reference body weight, the corrected linear cumulative amount differential is: Nonlinear damage is calculated using Euler's recursion: wherein k is the iteration number, is a frequency weighting coefficient, is a vibration amplitude to the power of n, is a body weight correction factor; Final accumulated damage Is: wherein is a non-linear time exponent.

5. The optimal control system of claim 4, wherein, The assessment model for the cumulative damage is as follows: in, These are frequency-weighted coefficients. The human tissue damage index, This is a standard reference weight for the human body. The bioallometric growth weight scaling index. It is a non-linear cumulative exponent.

6. The optimal control system of claim 1, wherein, The occupant adaptive learning module in the control layer uses historical optimization parameters as the initial adjustment benchmark to record the optimal damping coefficient and posture parameters under different weights and sitting postures.

7. The optimal control system of claim 1, wherein, The working principle of the adjustable seat damping mechanism in the execution layer is as follows: the PWM current signal output from the control layer is input to the electromagnetic coil of the electromagnetic variable damping shock absorber, changing the viscosity of the magnetorheological fluid in the electromagnetic variable damping shock absorber, and continuously adjusting the damping coefficient of the electromagnetic variable damping shock absorber. The relationship between the damping coefficient and the current satisfies: wherein is the damping coefficient of the electromagnetic variable-damping shock absorber, is a proportionality coefficient, is the current, is the minimum damping.

8. The optimal control system of claim 1, wherein, The seat posture adjustment component in the execution layer includes a backrest angle adjustment unit, a seat cushion angle adjustment unit, and a height adjustment unit. The control logic is as follows: Backrest angle adjustment unit: automatically fine-tunes according to the rate of change in vehicle speed; Seat cushion angle adjustment unit: When encountering bumpy roads, the front of the seat cushion tilts up to improve thigh support; Height adjustment unit: When driving on bumpy roads, raise the seat to increase the shock absorption travel; when driving at high speed and smoothly, lower the seat to optimize the center of gravity. The backrest angle adjustment unit, seat cushion angle adjustment unit, and height adjustment unit are all equipped with position feedback sensors to collect the actual position in real time and transmit it to the control layer.

9. The optimal control system of claim 1, wherein, The active suspension coordination module in the execution layer communicates with the suspension controller via the CAN FD bus and sends coordination commands. If the difference between the current vibration main frequency and the seat resonance frequency is ≤5Hz, the suspension stiffness is increased.

10. A method of cumulative assessment of injury to a vehicle occupant under vibration, characterized by, Includes the following steps: Preset amplitude thresholds for vibration interference level classification: in, The sliding root mean square value of the acceleration amplitude corresponding to the current dominant vibration frequency within a 0.1-second window is used. This classification standard is based on the threshold of subjective and physiological response of the human body to continuous vibration in the ISO 2631-1 standard, and the vibration intensity is quantified into three levels. The damage adaptive gain is selected according to the current level, selected according to the following table : Calculate the target damping coefficient: in, The target damping coefficient, The initial preset damping coefficient, The damage-adaptive gain is determined by the vibration level. This represents the current cumulative damage value. This represents the maximum permissible cumulative damage threshold. Obtaining the seat's absolute velocity: Vertical acceleration is measured via an acceleration sensor at the bottom of the seat. The absolute velocity is obtained through numerical integration and high-pass filtering. ; Relative speed is obtained by measuring the relative displacement between the seat and the chassis using a vehicle height sensor or displacement sensor, and then calculating the difference to obtain the relative speed. ,in This refers to the absolute speed of the chassis. Calculate ideal ceiling forces: The core objective of skyhook control is to suppress the absolute motion of the seat (sprung mass), Signed absolute velocity, positive up, negative down; Applying Karnopp semi-active constraints: wherein, Factual is the control force actually output to the damper; Based on the force-current characteristic curve of the electromagnetic variable damping vibration damper, the actual damping force is... Mapped to control current ; to transfer current The duty cycle is converted into a PWM signal and output to the electromagnetic variable damping vibration damper in the execution layer.

Images