A method for vertical vibration control in integrated electric drive vehicles

Abstract

This invention discloses a vertical vibration control method for integrated electric drive vehicles, comprising: constructing a 1 / 2 seven-degree-of-freedom model based on a traditional 1 / 2 four-degree-of-freedom model, adding vertical vibrations of the front and rear axles and vibrations related to the unbalanced magnetic pull of the motor; establishing dynamic equations and transforming them into state-space expressions containing road disturbances and motor magnetic pull, thereby constructing an extended state system. Disturbances are observed using a high-gain observer to ensure error convergence, and feedback controllers are designed for different frequency bands: H∞ robust control is used to suppress body vibration at low frequencies, μ-comprehensive robust control is used to suppress wheel bounce at mid-frequency frequencies, and adaptive resonant control is used to suppress motor vibration at high frequencies; combined with feedforward compensation to offset disturbances and system delays, adaptive weights are allocated according to the vibration energy of each frequency band, and the combined feedback power and feedforward compensation force output is used as the total power. This invention effectively improves vehicle NVH performance, enhances overall vehicle ride comfort, handling stability, and driving safety.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle technology, specifically to a vertical vibration control method for integrated electric drive vehicles. Background Technology

[0002] With the rapid development of vehicle electrification, integrated electric drive axles have become an important path for vehicle electrification transformation due to their relatively low development difficulty, high integration, and minimal impact on overall vehicle design. However, they also bring new noise, vibration, and harshness (NVH) problems.

[0003] While electric motors themselves exhibit low vibration and noise, their electromagnetic excitation characteristics differ fundamentally from those of traditional engines. The tight coupling between the integrated drive system and the rear axle system, in particular, complicates the vibration transmission path. Manufacturing defects and motor shaft deformation can cause rotor eccentricity, resulting in unbalanced magnetic pull that is transmitted to the vehicle body through the motor housing, axle housing, and suspension, significantly impacting the vehicle's low-frequency vibration performance. Furthermore, the dynamic displacement changes caused by motor vibration further exacerbate air gap unevenness, intensifying electromagnetic vibration and noise, creating a vicious cycle of electromagnetic-mechanical coupling. Vertical vibration is a critical issue in vehicle NVH (Noise, Vibration, and Harshness), directly affecting ride comfort, handling stability, and driving safety.

[0004] However, with the development of integrated electric drive systems in vehicles, a series of problems have arisen in vertical vibration-related technologies. ① Traditional vertical vibration models for vehicles are no longer applicable to integrated electric drive vehicles. Furthermore, unlike hub motor-driven vehicles, the motor and its electric drive assembly in integrated electric drive vehicles are independent of the tire assembly. Therefore, current vertical vibration models for hub motor-driven vehicles that consider motor coupling effects are also unsuitable for integrated electric drive vehicles. ② Obtaining the unbalanced magnetic pull after electromechanical coupling is difficult. Currently, the motor eccentricity (eccentricity) is usually assumed to be constant, as in this patent (CN119388938A); or the relative displacement between the stator and rotor in the vibration model is directly taken as the motor eccentricity, as in this patent (CN118410634A); or an invasive sensor is used to measure the eccentricity. ③ The control architecture does not deeply consider the effects of electromechanical coupling. Current control architecture technologies focus on vibration control algorithms and only treat the motor as a vibration excitation source, meaning that the unbalanced magnetic pull is not considered in the constraints of the vibration model and control algorithm. However, the motor is also a vibration receiver in the coupled system, and the vibration frequency band of the motor also needs to be considered in the control architecture design.

[0005] Therefore, in response to the aforementioned existing problems, this invention establishes a vertical vibration model that can comprehensively describe the vertical coupling relationship between the motor, rear axle, suspension, and tires, and is applicable to integrated electric drive vehicles. It uses a state observer to observe unbalanced magnetic pull and proposes a multi-frequency parallel control method for multi-source excitation to improve the NVH performance of integrated electric drive vehicles under all operating conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a vertical vibration control method for integrated electric drive vehicles, aiming to solve the problems in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a vertical vibration control method for integrated electric drive vehicles, comprising the following steps:

[0008] Step S1: Constructing a vertical vibration model and state-space expression: Based on the traditional 1 / 2 four-degree-of-freedom vehicle model, add vertical vibrations of the front and rear axles and vibrations related to the unbalanced magnetic pull of the motor to construct a 1 / 2 seven-degree-of-freedom model; establish dynamic equations based on Newton's second law and pitch moment balance equations, and then select vehicle displacement, velocity, and motor displacement as state variables, and road disturbance and motor magnetic pull as disturbance variables, and transform them into state-space expressions;

[0009] Step S2: Joint estimation of disturbances by multiple disturbance ESO observers: Based on the state-space expression, three types of disturbances are separated: motor unbalanced magnetic pull, road displacement / velocity, etc., and an extended state system containing disturbances is constructed; a high-gain observer is designed to observe the extended state, and the observation error is checked to see if it converges. If it does not converge, the parameters are adjusted and the system is redesigned. After convergence, the disturbance observation value is output.

[0010] Step S3: Multi-band parallel control outputs total working power: Based on disturbance observations, controllers are designed for different frequency bands: H∞ robust control is used to suppress vehicle body vibration at low frequencies, μ comprehensive robust control is used to suppress wheel bounce at mid frequencies, and adaptive resonant control is used to suppress motor vibration at high frequencies; at the same time, a feedforward controller is designed to compensate for disturbances and system delays, calculate the vibration energy of each frequency band and assign adaptive weights, and fuse the feedback working power, i.e., the weighted working power of each frequency band and the feedforward compensation force, to output the total working power to achieve vertical vibration suppression.

[0011] Furthermore, the specific process of step S1 is as follows:

[0012] Step S1.1: Build a 1 / 2 seven-degree-of-freedom model for integrated electric drive vehicles. The seven-degree-of-freedom model is based on the traditional 1 / 2 four-degree-of-freedom model of drive vehicles, with the addition of vertical vibration of the front axle, vertical vibration of the rear axle, and vertical component vibration caused by the unbalanced magnetic pull of the motor.

[0013] Step S1.2: Based on Newton's second law and the pitch moment balance equation, establish the vertical vibration dynamic equation of the 1 / 2 seven-degree-of-freedom model; the vertical vibration dynamic equation of the 1 / 2 seven-degree-of-freedom model includes the vertical vibration dynamic equation of the vehicle body, the vertical vibration dynamic equation of the motor stator, the vertical vibration dynamic equation of the front and rear axles, the vertical vibration dynamic equation of the front and rear tires, and the pitch motion dynamic equation of the vehicle body.

[0014] Step S1.3: Select state variables , as dynamic input variables and disturbance input variables ;

[0015] Step S1.4: Transform the dynamic equations in step S1.2 into state-space expressions:

[0016] ;

[0017] In the formula, express The rate of change of the state variable at time t; , , These represent the system matrix, input matrix, and disturbance matrix, respectively.

[0018] Furthermore, state variables Represented as:

[0019] ;

[0020] In the formula, Indicates the vertical displacement of the vehicle body; This represents the vertical velocity of the vehicle body, which is its vertical displacement. The first derivative; Indicates the vehicle's pitch angle; This represents the pitch rate of the vehicle body, which is the pitch angle of the vehicle body. The first derivative; This indicates the vertical displacement of the connection point between the front axle and the wheel hub bearing; This represents the vertical velocity at the connection point between the front axle and the wheel hub bearing, and the vertical displacement at the connection point between the front axle and the wheel hub bearing. The first derivative; This indicates the vertical displacement of the connection point between the rear axle and the wheel hub bearing; This represents the vertical velocity at the connection point between the rear axle and the wheel hub bearing, and the vertical displacement at the connection point between the rear axle and the wheel hub bearing. The first derivative; Indicates the vertical displacement of the motor; This represents the vertical velocity of the motor, which is its vertical displacement. The first derivative; This indicates the vertical displacement of the front tire; This represents the vertical velocity of the front tire, which is also the vertical displacement of the front tire. The first derivative; This indicates the vertical displacement of the rear tire; This represents the vertical velocity of the rear tire, which is its vertical displacement. The first derivative; Indicates transpose;

[0021] Disturbance input variables Represented as:

[0022] ;

[0023] In the formula, This represents the displacement caused by road disturbance to the front wheels; This represents the displacement caused by road disturbance to the rear wheel; This represents the vertical velocity of the road disturbance to the front wheel, which is the displacement excited by the road disturbance to the front wheel. The first derivative; This represents the vertical velocity of the road surface disturbance at the rear wheel, which is the displacement excited by the road surface disturbance at the rear wheel. The first derivative; This indicates an unbalanced magnetic pull in the motor.

[0024] Furthermore, the specific process of step S2 is as follows:

[0025] Step S2.1: Perform perturbation separation modeling on the state-space expression in step S1.3, and define the output vector. , Represents the identity matrix; the disturbance is divided into motor unbalanced magnetic pull disturbance. Road surface unevenness displacement disturbance and road surface unevenness and speed disturbance The model after perturbation separation is obtained;

[0026] Step S2.2: Construct an extended state system that includes road surface disturbance and motor unbalanced magnetic pull, and... , As an extended state variable This yields an extended system dynamics model. Represents the perturbation subvector;

[0027] Step S2.3: Design a high-gain observer to observe the extended state variables;

[0028] Step S2.4: Perform a convergence check on the observation error and set the convergence criterion. , Indicates observation error. Indicates tolerance error; if , To represent an expanded output matrix, adjust the gain parameter. And redesign the high-gain observer model , This represents a matrix designed using pole placement; if The observed value of the unbalanced magnetic pull disturbance of the output motor Observed values ​​of road surface unevenness displacement disturbance Road surface unevenness velocity disturbance observation values .

[0029] Furthermore, the high-gain observer model is as follows:

[0030] ;

[0031] In the formula, This represents the estimated value of the extended state variable; The time derivative of the extended state variable estimate; Indicates the gain parameter; Represents the gain matrix. Represents the gain scaling matrix. This represents a matrix designed using pole placement; Represents the extended system matrix; This represents the extended input matrix.

[0032] Furthermore, the specific process of step S3 is as follows:

[0033] Step S3.1: Design a low-frequency band H∞ robust controller to output low-frequency band power. ;

[0034] Step S3.2: Design a mid-frequency μ-band robust controller to output mid-frequency power. ;

[0035] Step S3.3: Design a high-frequency adaptive resonant controller to output high-frequency power. ;

[0036] Step S3.4: Design a feedforward controller to output the predicted feedforward compensation force. ;

[0037] Step S3.5: Calculate the adaptive weights and combine them with the power output of each frequency band to generate the total power;

[0038] The total operating power includes the feedback control operating power. and feedforward regulation control force , is represented as:

[0039] ;

[0040] In the formula, , , These represent the weights for low, medium, and high frequency bands, respectively.

[0041] Furthermore, the specific process of step S3.1 is as follows:

[0042] Design a low-frequency H∞ robust controller closed-loop system:

[0043] For rigid body vibration of the vehicle body, based on the state-space expression and disturbance observations, the output index of the low-frequency H∞ robust controller closed-loop system is defined. , This represents the vertical acceleration of the vehicle body, which is also the vertical displacement of the vehicle body. The second derivative, This represents the vehicle pitch angle acceleration, which is the vehicle pitch angle. The second derivative; constructing a low-frequency H∞ robust controller closed-loop system;

[0044] Design an H∞ robust control structure:

[0045] Using partial states The feedback gain matrix of the low-frequency H∞ robust controller is as follows: ;

[0046] Solve This ensures that the closed-loop system of the low-frequency H∞ robust controller satisfies the H∞ norm constraint: , Represents the H∞ norm; The preset upper limit threshold for the H∞ norm;

[0047] By constructing Lyapunov generalized functions, linear matrix inequalities that satisfy the H∞ norm constraint are obtained;

[0048] The feedback gain is obtained by solving the linear matrix inequality. Output low frequency band as power .

[0049] Furthermore, the specific process of step S3.2 is as follows:

[0050] Constructing a generalized controlled system:

[0051] For wheel hop, the vehicle mass parameters are modeled as bounded uncertainties:

[0052] ;

[0053] In the formula, Indicates vehicle mass parameters; Indicates nominal mass; Indicates the weight of uncertainty; Represents a normalized uncertain block;

[0054] Define the output performance of the mid-frequency μ-integrated robust controller closed-loop system. ;

[0055] Determine the closed-loop system of the mid-frequency μ-integrated robust controller;

[0056] Integrate into a generalized controlled system: , Represents the matrix of the generalized controlled system;

[0057] Design a μ-synthetic robust control structure:

[0058] Dynamic output feedback controller parameters using a mid-frequency μ-synthetic robust controller ;

[0059] Determine robust performance conditions: satisfy structural singular value constraints , Represents structural singular values. Represents the matrix of the generalized controlled system With feedback controller parameters The transfer function matrix after loop closure;

[0060] Solve using the DK iterative algorithm Output mid-frequency band as power .

[0061] Furthermore, the specific process of step S3.3 is as follows:

[0062] Estimated motor vibration frequency: Observations of motor unbalanced magnetic pull disturbance Perform a Fourier transform and find the peak frequency of the transform result to obtain the peak vibration frequency of the motor. ;

[0063] Design an adaptive notch filter: Determine the target notch frequency The transfer function of the notch filter is derived using the zero-pole placement method. ;

[0064] Design of a resonant controller: A resonant controller based on the internal mode principle is designed to address residual vibrations after notch filtering. By connecting resonant controllers of different frequencies in parallel, a comprehensive resonant controller is obtained. , Represents a complex frequency variable;

[0065] Solving for high-frequency dynamics: Calculating high-frequency dynamics by connecting an adaptive notch filter and a resonant controller in parallel. .

[0066] Furthermore, the specific process of step S3.4 is as follows:

[0067] Design a feedforward control force based on disturbance estimation: using disturbance observations Calculate real-time feedforward compensation force ;

[0068] Design disturbance prediction compensation: prediction delay Subsequent perturbation observations: , Indicates the rate of change of the observed disturbance;

[0069] Obtain the predicted feedforward compensation force .

[0070] Compared with existing technologies, the present invention has the following advantages:

[0071] (1) In terms of model construction, the present invention innovatively builds a 1 / 2 seven-degree-of-freedom model suitable for integrated electric drive vehicles, which solves the problem that the current traditional four-degree-of-freedom model and the vibration model of wheel hub motor driven vehicles are not suitable for integrated electric drive vehicles. It accurately describes the vertical coupling relationship between motor-rear axle-suspension-tire and adds the unbalanced magnetic pull component of motor, thereby significantly improving the accuracy and applicability of vibration analysis.

[0072] (2) The present invention uses a multi-disturbance high-gain extended state (ESO) observer to realize the joint estimation of road disturbance and motor unbalanced magnetic pull. By extending the disturbance into a state variable and designing adaptive high-gain parameters, the problem of eccentricity not being directly obtained is solved, the use of intrusive sensors is avoided, the measurement cost is reduced, and the estimation error is ensured to converge, thereby improving anti-interference capability and real-time performance.

[0073] (3) The present invention designs a multi-band parallel adaptive strategy in the control architecture, which enables the control action to automatically focus on the frequency band with the most severe vibration at present, and can simultaneously suppress vehicle body vibration and wheel bounce caused by uneven road surface, as well as high-frequency resonance caused by unbalanced magnetic pull of motor, so as to effectively suppress the vibration of integrated electric drive vehicle across the entire frequency band. In addition, combined with feedforward compensation and robust control technology, while ensuring system stability, it effectively suppresses the electromagnetic coupling effect of vertical vibration motor of integrated electric drive vehicle, improves NVH performance, and enhances the ride comfort, handling stability and driving safety of the whole vehicle. Attached Figure Description

[0074] Figure 1 This is a flowchart of the method of the present invention.

[0075] Figure 2This is a schematic diagram of a 1 / 2 seven-degree-of-freedom model for integrated electric drive vehicles according to the present invention. Detailed Implementation

[0076] like Figure 1 As shown, the present invention provides a technical solution: a vertical vibration control method for integrated electric drive vehicles, comprising the following steps:

[0077] Step S1: Constructing a vertical vibration model and state-space expression: Based on the traditional 1 / 2 four-degree-of-freedom vehicle model, add vertical vibrations of the front and rear axles and vibrations related to the unbalanced magnetic pull of the motor to construct a 1 / 2 seven-degree-of-freedom model; establish dynamic equations based on Newton's second law and pitch moment balance equations, and then select vehicle displacement, velocity, motor displacement, etc. as state variables, and road disturbance and motor magnetic pull as disturbance variables, and transform them into state-space expressions.

[0078] The specific process of step S1 is as follows:

[0079] Step S1.1: Build a 1 / 2 (half-vehicle) seven-degree-of-freedom model for integrated electric drive vehicles, such as... Figure 2 As shown, the seven-degree-of-freedom model, based on the traditional 1 / 2 four-degree-of-freedom model of a driving vehicle, adds vertical vibrations of the front axle, rear axle, and vertical component vibrations caused by the unbalanced magnetic pull of the motor. The seven-degree-of-freedom model includes sprung mass. Front axle quality Rear axle quality Motor rotor mass Motor stator quality Front unsprung mass excluding axles Rear unsprung mass excluding axles Vertical displacement of vehicle body Vertical displacement of the connection point between the front suspension and the body Vertical displacement of the connection point between the rear suspension and the body Vertical displacement of the connection point between the front axle and the wheel hub bearing Vertical displacement of the connection point between the rear axle and the wheel hub bearing Vertical displacement of the motor Vertical displacement of front tires Vertical displacement of rear tires Road disturbance excitation displacement of the front wheels Road disturbance excitation displacement of the rear wheels Equivalent stiffness of the front suspension Equivalent stiffness of the rear suspension The equivalent damping coefficient of the front suspension The equivalent damping coefficient of the rear suspension Motor bearing stiffness Front wheel hub bearing equivalent stiffness Equivalent stiffness of rear wheel hub bearing front tire stiffness Rear tire stiffness The equivalent damping coefficient of the front tire The equivalent damping coefficient of the rear tire Adjustable front suspension for power Adjustable rear suspension for power Unbalanced magnetic pull of the motor Vehicle pitch angle Vehicle body pitch moment of inertia Distance from the center of gravity to the center of the front wheel Distance from the center of gravity to the center of the rear wheel .

[0080] Step S1.2: Based on Newton's second law and the pitch moment balance equation, establish the vertical vibration dynamic equation of the 1 / 2 seven-degree-of-freedom model, including:

[0081] Vehicle body vertical vibration dynamics equation:

[0082] ;

[0083] In the formula, This represents the vertical acceleration of the vehicle body, which is also the vertical displacement of the vehicle body. The second derivative; It represents the vertical velocity at the connection point between the front suspension and the vehicle body, and is also the vertical displacement at that connection point. The first derivative; This represents the vertical velocity at the connection point between the front axle and the wheel hub bearing, and the vertical displacement at the connection point between the front axle and the wheel hub bearing. The first derivative; It represents the vertical velocity at the connection point between the rear suspension and the vehicle body, and is also the vertical displacement at that connection point. The first derivative; This represents the vertical velocity at the connection point between the rear axle and the wheel hub bearing, and the vertical displacement at the connection point between the rear axle and the wheel hub bearing. The first derivative.

[0084] Dynamic equations of vertical vibration of motor stator:

[0085] ;

[0086] In the formula, This represents the vertical acceleration of the motor stator, which is the vertical displacement of the motor. The second derivative of .

[0087] The dynamic equations of vertical vibration of the front and rear axles are as follows:

[0088] ;

[0089] In the formula, This represents the vertical acceleration of the front axle, which is the vertical displacement of the connection point between the front axle and the wheel hub bearing. The second derivative; This represents the vertical acceleration of the rear axle, which is the vertical displacement of the connection point between the rear axle and the wheel hub bearing. The second derivative of .

[0090] The dynamic equations of vertical vibration of the front and rear tires are as follows:

[0091] ;

[0092] In the formula, This represents the vertical acceleration of the front tire and its vertical displacement. The second derivative; This represents the vertical velocity of the front tire, which is also the vertical displacement of the front tire. The first derivative; This represents the vertical velocity of the road disturbance to the front wheel, which is the displacement excited by the road disturbance to the front wheel. The first derivative; This represents the vertical acceleration of the rear tire, and its vertical displacement. The second derivative; This represents the vertical velocity of the rear tire, which is its vertical displacement. The first derivative; This represents the vertical velocity of the road surface disturbance at the rear wheel, which is the displacement excited by the road surface disturbance at the rear wheel. The first derivative.

[0093] Vehicle pitch motion dynamics equations:

[0094] ;

[0095] In the formula, This represents the vehicle pitch angle acceleration, which is the vehicle pitch angle. The second derivative of .

[0096] Among them, the vehicle pitch angle Satisfying the small angle assumption, the vertical displacement of the connection point between the front suspension and the vehicle body Vertical displacement of the connection point between the rear suspension and the body It can be represented as:

[0097] .

[0098] Step S1.3: Select state variables , as dynamic input variables and disturbance input variables ;

[0099] State variables Represented as:

[0100] ;

[0101] In the formula, This represents the vertical velocity of the vehicle body, which is its vertical displacement. The first derivative; This represents the pitch rate of the vehicle body, which is the pitch angle of the vehicle body. The first derivative; This represents the vertical velocity of the motor, which is its vertical displacement. The first derivative.

[0102] Disturbance input variables Represented as:

[0103] .

[0104] Step S1.4: Transform the dynamic equations in step S1.2 into state-space expressions:

[0105] ;

[0106] In the formula, express The rate of change of the state variable at time t; , , These represent the system matrix, input matrix, and disturbance matrix, respectively.

[0107] Step S2: Joint estimation of disturbances by multiple disturbance ESO observers: Based on the state-space expression, three types of disturbances are separated: motor unbalanced magnetic pull, road displacement / velocity, etc., and an extended state system containing disturbances is constructed. A high-gain observer is designed to observe the extended state and check whether the observation error converges (less than the tolerance value). If it does not converge, the parameters are adjusted and the system is redesigned. After convergence, the disturbance observation value is output.

[0108] The specific process of step S2 is as follows:

[0109] Step S2.1: Perform perturbation separation modeling on the state-space expression in step S1.3, and define the output vector. , The identity matrix is ​​represented by the state variables, which are all measurable. The disturbance is categorized into motor unbalanced magnetic pull disturbance. Road surface unevenness displacement disturbance and road surface unevenness and speed disturbance The model after perturbation separation is obtained:

[0110] ;

[0111] In the formula, , , All are matrices after perturbation separation, derived from the perturbation matrix. It was obtained by splitting it.

[0112] Step S2.2: Construct an extended state system that includes road surface disturbance and motor unbalanced magnetic pull, and... , As an extended state variable , Representing the perturbation subvector, we obtain the extended system dynamics model:

[0113] ;

[0114] In the formula, Represents the time derivative of the extended state variable; Represents the extended system matrix; Represents the extended input matrix; Represents the extended perturbation matrix; Indicates the extended output matrix; This represents the dynamic vector of the disturbance. Indicates motor unbalanced magnetic pull disturbance The time derivative.

[0115] Step S2.3: Design a high-gain observer to observe the extended state variables. The observer model is as follows:

[0116] ;

[0117] In the formula, This represents the estimated value of the extended state variable; The time derivative of the extended state variable estimate; This indicates that the gain parameter can be increased; Represents the gain matrix. Represents the gain scaling matrix. This represents a matrix designed using pole placement, such that... It is a Hurwitz matrix.

[0118] Step S2.4: Perform a convergence check on the observation error and set the convergence criterion. , Indicates observation error. Indicates tolerance error; if Then adjust the adjustable gain parameter. And redesign ;like The observed value of the unbalanced magnetic pull disturbance of the output motor Observed values ​​of road surface unevenness displacement disturbance Road surface unevenness velocity disturbance observation values .

[0119] Step S3: Multi-band parallel control outputs total working power: Based on disturbance observations, controllers are designed for different frequency bands: H∞ robust control is used to suppress vehicle body vibration at low frequencies, μ comprehensive robust control is used to suppress wheel bounce at mid frequencies, and adaptive resonant control is used to suppress motor vibration at high frequencies; at the same time, a feedforward controller is designed to compensate for disturbances and system delays, calculate the vibration energy of each frequency band and assign adaptive weights, and fuse the feedback working power, i.e., the weighted working power of each frequency band and the feedforward compensation force, to output the total working power to achieve vertical vibration suppression.

[0120] The specific process of step S3 is as follows:

[0121] Step S3.1: Design a low-frequency band H∞ robust controller to output low-frequency band power;

[0122] 1. Design a low-frequency band H∞ robust controller closed-loop system:

[0123] For rigid body vibration of the vehicle body, based on the state-space expression and disturbance observations, the output index of the low-frequency H∞ robust controller closed-loop system is defined. ;

[0124] The low-frequency H∞ robust controller closed-loop system is represented as:

[0125] ;

[0126] In the formula, This represents the low-frequency output matrix, determined by the vehicle's physical parameters, and describes... arrive The mapping relationship; This represents the low-frequency input-output matrix, determined by the vehicle's physical parameters, and describes... arrive The direct impact relationship.

[0127] 2. Design a robust control structure for H∞:

[0128] Using partial states The feedback gain matrix of the low-frequency H∞ robust controller is as follows: Partial state feedback refers to selecting only key states that are strongly related to "vehicle rigid body vibration", such as vehicle vertical displacement and vehicle pitch angle, rather than all system states.

[0129] Solve This ensures that the closed-loop system of the low-frequency H∞ robust controller satisfies the H∞ norm constraint:

[0130] ;

[0131] In the formula, Represents the H∞ norm; This represents the preset upper limit threshold of the H∞ norm.

[0132] 3. By constructing Lyapunov functionals, we obtain linear matrix inequalities (LMIs) that satisfy the H∞ norm constraint.

[0133] 4. Solve the linear matrix inequalities to obtain the feedback gain. Output low frequency band as power .

[0134] Step S3.2: Design a mid-frequency μ-integrated robust controller to output mid-frequency power;

[0135] 1. Construct a generalized controlled system:

[0136] For wheel hop, the vehicle mass parameters (sprung mass and unsprung mass) are modeled as bounded uncertainties:

[0137] ;

[0138] In the formula, Indicates vehicle mass parameters (sprung mass, unsprung mass). Indicates nominal mass; Indicates the weight of uncertainty; Represents a normalized uncertain block;

[0139] Define the output performance of the mid-frequency μ-integrated robust controller closed-loop system. :

[0140] ;

[0141] The closed-loop system of the mid-frequency μ-synthetic robust controller is defined as follows:

[0142] ;

[0143] In the formula, This represents the mid-frequency output matrix, determined by the vehicle's physical parameters. arrive The mapping relationship; This represents the mid-frequency input-output matrix, determined by the vehicle's physical parameters, and describes... arrive The direct impact relationship.

[0144] Integrate into a generalized controlled system:

[0145] ;

[0146] In the formula, This represents the matrix of the generalized controlled system.

[0147] 2. Design of μ-synthetic robust control structure:

[0148] Dynamic output feedback controller parameters using a mid-frequency μ-synthetic robust controller ;

[0149] Determine robust performance conditions: satisfy structural singular value constraints This ensures the internal stability of the closed-loop system of the mid-frequency μ-integrated robust controller and satisfies H∞ performance for all permissible parameter uncertainties. Represents structural singular values. Represents the matrix of the generalized controlled system With feedback controller parameters The transfer function matrix after loop closure.

[0150] 3. Solve using the DK iterative algorithm. Output mid-frequency band as power .

[0151] Step S3.3: Design a high-frequency adaptive resonant controller to output high-frequency power;

[0152] 1. Estimate the motor vibration frequency: based on the observed values ​​of the motor's unbalanced magnetic pull disturbance. Perform a Fourier transform and find the peak frequency of the transform result to obtain the peak vibration frequency of the motor. :

[0153] ;

[0154] In the formula, This indicates the Fourier transform operation; This indicates an operation to find the peak frequency.

[0155] 2. Design an adaptive notch filter: Determine the target notch frequency. The transfer function of the notch filter is derived using the zero-pole placement method. :

[0156] ;

[0157] In the formula, Represents a complex frequency variable; This represents the damping ratio of the molecular part of the notch filter; This represents the damping ratio in the denominator of the notch filter.

[0158] 3. Design of a resonant controller: To address the residual vibration after notch filtering, a resonant controller is designed based on the internal mode principle. The transfer function is:

[0159] ;

[0160] In the formula, This represents the gain coefficient of the resonant controller; This indicates the damping ratio of the resonant controller; This indicates the resonant angular frequency corresponding to the resonant controller;

[0161] By connecting resonant controllers of different frequencies in parallel, a composite resonant controller is obtained. To suppress fundamental frequency and harmonic frequency vibrations.

[0162] 4. Solving for high-frequency dynamics: Calculate the high-frequency dynamics by connecting the adaptive notch filter and the resonant controller in parallel. :

[0163] .

[0164] Step S3.4: Design a feedforward controller to output the predicted feedforward compensation force;

[0165] 1. Design feedforward control force based on disturbance estimation: using disturbance observations Calculate real-time feedforward compensation force :

[0166] ;

[0167] In the formula, express The transpose of .

[0168] 2. Design disturbance prediction compensation (to solve the time delay problem): prediction delay Subsequent perturbation observations:

[0169] ;

[0170] In the formula, Indicates the rate of change of the observed disturbance;

[0171] Obtain the predicted feedforward compensation force :

[0172] .

[0173] Step S3.5: Calculate the adaptive weights and combine them with the power output of each frequency band to generate the total power;

[0174] 1. Calculate the vibration energy in each frequency band:

[0175] ;

[0176] In the formula, , , These represent the vibrational energy in the low, medium, and high frequency bands, respectively. Indicates the low-frequency range; Indicates the mid-frequency range; Indicates the frequency range of the high-frequency band; Represents frequency variables; This represents a differential operator.

[0177] 2. Dynamically allocate adaptive weights based on the vibration energy of each frequency band:

[0178] ;

[0179] In the formula, , , These represent the weights for low, medium, and high frequency bands, respectively. This represents a regularization term used to ensure numerical stability.

[0180] 3. Output total operating power to achieve vertical vibration suppression of integrated electric drive vehicles; the total operating power includes feedback control operating power. and feedforward regulation control force , is represented as:

[0181] .

[0182] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A vertical vibration control method for an integrated electric drive vehicle, characterized by, Comprise the following steps: Step S1, build vertical vibration model and state space expression: based on the traditional vehicle 1 / 2 four degree of freedom model, add front axle, rear axle vertical vibration and motor imbalance magnetic pull related vibration, build 1 / 2 seven degree of freedom model; Based on Newton's second law and pitch moment balance equation, the dynamic equation is established, and then the body displacement, speed, motor displacement is selected as the state variable, the road disturbance, motor magnetic pull is the disturbance variable, and the state space expression is converted; Step S2, joint estimation of disturbance: based on the state space expression, separate the motor imbalance magnetic pull, road displacement / speed three kinds of disturbance, build the extended state system containing disturbance; Design high gain observer to observe the extended state, check whether the observation error converges, if not, adjust the parameters and redesign, and output the disturbance observation value after convergence; Step S3, multi-band parallel control output total actuator: based on the disturbance observation value, design controller by frequency band: low frequency uses H∞robust control to suppress vehicle body vibration, medium frequency uses μ synthesis robust control to suppress wheel vibration, high frequency uses adaptive resonance control to suppress motor vibration; At the same time, design feedforward controller to compensate disturbance and system delay, calculate vibration energy of each frequency band and distribute adaptive weight, fuse feedback actuator, that is, weighted actuator of each frequency band and feedforward compensation force, output total actuator to realize vertical vibration suppression.

2. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 1, characterized in that: The specific process of step S1 is: Step S1.1: build 1 / 2 seven degree of freedom model for integrated electric drive vehicle, which adds front axle vertical vibration, rear axle vertical vibration and vertical component vibration caused by motor imbalance magnetic pull on the basis of traditional drive vehicle 1 / 2 four degree of freedom model; Step S1.2: based on Newton's second law and pitch moment balance equation, the vertical vibration dynamics equation of the 1 / 2 seven degree of freedom model is established; The vertical vibration dynamics equation of the 1 / 2 seven degree of freedom model includes vehicle body vertical vibration dynamics equation, motor stator vertical vibration dynamics equation, front and rear axle vertical vibration dynamics equation, front and rear tire vertical vibration dynamics equation, vehicle body pitch motion dynamics equation; Step S1.3: Selecting state variables , actuating force input variables , and disturbance input variables ; Step S1.4: convert the dynamic equation in step S1.2 into state space expression: ； In the formula, represents the rate of change of the state variable at the time instant; , , respectively represent a system matrix, an input matrix, and a disturbance matrix.

3. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 2, characterized in that: State variable is represented as: ； In the formula, represents the vertical displacement of the vehicle body; represents the vertical velocity of the vehicle body, which is the first derivative of the vertical displacement of the vehicle body ; represents the pitch angle of the vehicle body; represents the pitch angular velocity of the vehicle body, which is the first derivative of the pitch angle of the vehicle body ; represents the vertical displacement of the front axle and the hub bearing connection point; represents the vertical velocity of the front axle and the hub bearing connection point, which is the first derivative of the vertical displacement of the front axle and the hub bearing connection point ; represents the vertical displacement of the rear axle and the hub bearing connection point; represents the vertical velocity of the rear axle and the hub bearing connection point, which is the first derivative of the vertical displacement of the rear axle and the hub bearing connection point ; represents the vertical displacement of the motor; represents the vertical velocity of the motor, which is the first derivative of the vertical displacement of the motor ; represents the vertical displacement of the front tire; represents the vertical velocity of the front tire, which is the first derivative of the vertical displacement of the front tire ; represents the vertical displacement of the rear tire; represents the vertical velocity of the rear tire, which is the first derivative of the vertical displacement of the rear tire ; represents the transpose; Disturbance input variable is represented as: ； wherein represents a road surface disturbance excitation displacement of the front wheel; represents a road surface disturbance excitation displacement of the rear wheel; represents a vertical velocity of the front wheel road surface disturbance, which is a first derivative of the road surface disturbance excitation displacement of the front wheel ; represents a vertical velocity of the rear wheel road surface disturbance, which is a first derivative of the road surface disturbance excitation displacement of the rear wheel ; represents an unbalanced magnetic pull of the motor.

4. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 3, characterized in that: The specific process of step S2 is: Step S2.1: perturbation separation modeling on the state space expression in step S1.3, define the output vector , denotes the identity matrix; divide the perturbation into motor unbalanced magnetic pull perturbation , road unevenness displacement perturbation , and road unevenness speed perturbation , to obtain the perturbation separated model; Step S2.2: Construct an extended state system comprising road perturbations and motor unbalanced magnetic pull, yielding , as extended state variables , resulting in an extended system dynamics model, denotes the perturbation sub-vector; Step S2.3: design high gain observer to observe the extended state variable; Step S2.4: Convergence check on observation error, set convergence criterion , represents observation error, represents tolerance error; if , represents extended output matrix, adjust gain parameter and redesign , represents matrix designed by pole placement; if , output motor unbalance magnetic pull disturbance observation value , road unevenness displacement disturbance observation value , road unevenness speed disturbance observation value .

5. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 4, characterized in that: The high gain observer model is: ； wherein denotes the extended state variable estimate; denotes the time derivative of the extended state variable estimate; denotes the gain parameter; denotes the gain matrix, denotes the gain scaling matrix, denotes the matrix designed by pole placement; denotes the extended system matrix; denotes the extended input matrix.

6. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 5, characterized in that: The specific process of step S3 is: Step S3.1: design a low frequency band H∞ robust controller, output low frequency band actuator force ; Step S3.2: design a comprehensive robust controller for the mid-frequency band μ, output the mid-frequency band actuator force ; Step S3.3: design a high-frequency adaptive resonant controller, output high-frequency band as the driving force ; Step S3.4: Design a feedforward controller to output a predicted feedforward compensation force ; Step S3.5: calculate adaptive weight and output total actuator combined with each frequency band actuator; The total actuation force comprises a feedback control actuation force and a feedforward regulation control force is expressed as: ； In the formula, , , respectively represent low, medium, and high frequency band weights.

7. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 6, characterized in that: The specific process of step S3.1 is: Design low frequency H∞robust controller closed loop system: For the rigid body vibration of the vehicle body, based on the state space expression and the disturbance observation value, a low-frequency H∞ robust controller closed-loop system output index is defined , represents the vertical acceleration of the vehicle body, and is the second-order derivative of the vertical displacement of the vehicle body , represents the pitch angle acceleration of the vehicle body, and is the second-order derivative of the pitch angle of the vehicle body ; a low-frequency H∞ robust controller closed-loop system is constructed Design H∞robust control structure: Adopting partial state The feedback low-frequency H∞ robust controller, the feedback gain matrix of the low-frequency H∞ robust controller is ; Solving , to make the low-frequency H∞ robust controller closed-loop system satisfy the H∞ norm constraint: , represents the H∞ norm; represents the preset upper threshold value of the H∞ norm; By constructing Lyapunov function, the linear matrix inequality satisfying H∞norm constraint is obtained; Solving linear matrix inequalities to obtain feedback gains , output low frequency band actuator .

8. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 7, characterized in that: The specific process of step S3.2 is: Build generalized controlled system: For wheel vibration, the vehicle mass parameter is modeled as a bounded uncertain form: ； wherein denotes a vehicle mass parameter; denotes a nominal mass; denotes an uncertainty weight; denotes a normalized uncertainty block; Defining the mid-frequency band mu comprehensive robust controller closed-loop system output index ; Determine the μ synthesis robust controller closed loop system of medium frequency band; The integrated system is a generalized controlled system: , represents the generalized controlled system matrix; Design μ synthesis robust control structure: Dynamic output feedback controller parameters for a mid-band mu comprehensive robust controller ; Determining robust performance condition: satisfying structured singular value constraint , denotes structured singular value, denotes generalized plant matrix with feedback controller parameters closed-loop transfer function matrix; Solving the D-K iteration algorithm , output mid-frequency band actuator .

9. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 8, characterized in that: The specific process of step S3.3 is: Estimating motor vibration frequency: observing motor unbalance magnetic pull disturbance Performing Fourier transform, finding peak frequency of transform result, obtaining motor peak vibration frequency ; Designing an adaptive notch filter: determining a target notch frequency ; The transfer function of the notch filter is derived using the zero-pole configuration method ; Designing resonant controller: for the residual vibration after the notch, the resonant controller is designed based on the internal model principle; parallel resonant controllers with different frequencies are obtained to get the comprehensive resonant controller , represents the complex frequency variable; Solving high frequency band actuator power: adaptive notch filter and resonant controller in parallel to calculate high frequency band actuator power .

10. A method for vertical vibration control of an integrated electrically driven vehicle according to claim 9, characterized in that: The specific process of step S3.4 is: Design feedforward control force based on disturbance estimation: use disturbance observation Compute real-time feedforward compensation force ; Design disturbance prediction compensation: predicting the delay of the disturbance observation: , denotes the rate of change of the disturbance observation obtaining a predicted feedforward compensation force .

Images