A method and system for active suspension control of an electric bus

Abstract

The application discloses a kind of electric bus active suspension control method and system, method includes according to bus parameter and motor distribution establishes the half vehicle dynamics model of bus, constructs the state space equation of model, obtains the active suspension model suitable for bus;Considering the mass distribution change and centroid position change caused by bus passengers getting on and off and walking in the car, a social force model is established;And through the social force model, the relationship between the number of passengers on the bus and the front and rear distribution of passengers is obtained according to the particle swarm optimization algorithm, the passenger distribution law is obtained, and the active suspension model suitable for mass distribution change and centroid position change is established after the range of uncertain parameters is determined;For the direct interference of rear motor vibration, PID control is designed to adapt to the vibration of motor under different working conditions;By deriving the theorem of H2 / H ∞ control, the active suspension feedback control law when the mass distribution changes and the centroid position changes is obtained.The application can effectively improve the comfort of bus.

Description

Technical Field

[0001] This invention belongs to the field of active suspension control technology for vehicle chassis, and specifically relates to an active suspension control method and system for electric buses. Background Technology

[0002] As a shock-absorbing component of the vehicle chassis, the suspension is a crucial part of the vehicle's dynamics system. The suspension system primarily connects the vehicle body and tires, bearing the weight of the vehicle and reducing vibrations transmitted from the ground through the tires, thus improving ride comfort and handling stability. Passive suspension mainly consists of shock-absorbing springs and dampers. Passive suspension can only passively respond to uneven road surfaces, and its parameters cannot change with road surface conditions or operating conditions, resulting in poor vertical performance. Therefore, research has proposed active suspension, which uses an additional actuator to generate active driving force to actively suppress vibrations caused by road unevenness. Active suspension requires appropriate driving force; otherwise, it may deteriorate suspension performance. Therefore, researchers have conducted extensive research on the control system of active suspension and proposed many control strategies: optimal control, robust control, sliding mode variable structure control, and the recently emerging intelligent control methods. These methods have achieved relatively good control results.

[0003] Although significant progress has been made in active suspension control methods, current methods primarily target smaller vehicles like sedans and SUVs. Buses, lacking seat cushioning, experience more severe vertical vibrations compared to smaller vehicles. Buses differ greatly from smaller vehicles in terms of mass, suspension parameters, vehicle size, and operating conditions. Furthermore, the frequent passenger boarding and alighting and movement within the bus cause frequent changes in mass distribution. With a large number of passengers, this distribution fluctuates considerably, but currently, there are no effective methods to identify these changes, leading to deteriorated suspension performance. Additionally, most buses use a rear-engine, rear-wheel-drive configuration, meaning the vibrations of the rear-mounted engine or motor directly affect the vehicle body, exacerbating rear-end vibrations and causing discomfort. Current suspension control methods rely on random road vibrations and do not consider the impact of engine or motor vibrations, thus failing to effectively control these high-frequency, low-amplitude vibrations. Summary of the Invention

[0004] The purpose of this invention is to address the problems in the prior art by providing an active suspension control method and system for electric buses, which can adapt to changes in vehicle mass distribution caused by the bus's excessive length, frequent passenger boarding and alighting, and movement within the vehicle, and eliminate high-frequency vibrations caused by the rear-mounted motor, thereby effectively improving the comfort of the bus.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An active suspension control method for an electric bus includes the following steps:

[0007] A half-vehicle dynamics model of the bus is established based on the bus parameters and motor distribution. The state space equation of the model is constructed to obtain an active suspension model suitable for the bus.

[0008] For the active suspension model, considering the changes in mass distribution and center of mass position caused by passengers getting on and off the bus and moving around inside the vehicle, a social force model is established; and the relationship between the number of passengers on the bus and the front and back distribution of passengers is obtained through the social force model using the particle swarm optimization algorithm, so as to obtain the passenger distribution law. After determining the range of uncertain parameters, an active suspension model that adapts to changes in mass distribution and center of mass position is established.

[0009] Using an active suspension model that adapts to changes in mass distribution and center of gravity position as the control object, PID control is used to suppress the vibration of the rear motor during operation.

[0010] By deriving H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived for changes in mass distribution and center of mass position, and the hybrid H2 / H ... ∞ The optimal performance-preserving control law is used for active suspension control.

[0011] Preferably, the step of establishing a half-vehicle dynamics model of the bus based on the bus parameters and motor distribution, constructing the state-space equations of the model, and obtaining a suitable active suspension model for the bus includes:

[0012] Based on Newton's second law, the bus parameters and motor distribution establish the following half-vehicle dynamics model of the bus:

[0013]

[0014] In the formula, m m m s m uf m ur These are the masses of the motor, body, front suspension, and rear suspension, respectively; I y k is the moment of inertia. sf k sr k tf k tr k m For the stiffness of the front and rear suspension, tires, and motor mounts; C sf C sr C mThese are the damping coefficients for the front and rear suspensions and the motor mount; denoted as vertical and pitch acceleration of the vehicle body; a and b are the distances from the center of the front and rear suspensions to the center of mass of the vehicle body. x uf、 x ur The vertical acceleration, velocity, and displacement of the front and rear suspensions; x sf x sr x sm x m For the front and rear axles of the vehicle body, the vertical velocity and displacement of the vehicle body and motor at the motor location; x gf x gr Input for the front and rear road surfaces; u f u r F provides power to the front and rear suspensions. m This is the excitation force of the motor;

[0015] When constructing the state-space equations of the model, state variables X and road disturbance W are selected, and the half-vehicle dynamics model of the bus is rewritten into the following space equations based on the state variables X and the road disturbance W:

[0016]

[0017] In the formula:

[0018]

[0019]

[0020] U = [u f u r ] T ,

[0021] Preferably, in the step of establishing the social force model, the relationship between the number of passengers and the front and back distribution of passengers is obtained in real time. The combined effect of multiple force vectors on pedestrians during movement is considered. The social forces experienced by passengers on the bus include the repulsive force between passengers, the repulsive force between passengers and obstacles on the bus, the attractive force between passengers and seats, the repulsive force between passengers and the rear of the bus, as well as random and non-systematic forces generated by attraction behavior, aggregation behavior and disturbance behavior.

[0022] Preferably, the following dynamic equation is established for the social forces experienced by the passenger on the bus:

[0023]

[0024] In the formula, It is the resultant force of all social forces acting on passenger i during its movement; The repulsive force between passenger i and passenger j; The repulsive force between passenger i and the obstacle; The attraction between passenger i and the nearest seat; The repulsive force between passenger i and the rear of the bus; The random and non-systematic forces generated for passenger i;

[0025] The socio-psychological force exerted on passenger i by passenger j during the journey is represented by the following dynamic equation:

[0026]

[0027]

[0028]

[0029] In the formula, A i r is the influence intensity coefficient of passenger i; ij d is the sum of the radii of passenger i and passenger j. ij B is the distance between the centroids of passenger i and passenger j; i The influence range coefficient for passenger i; w is the unit direction vector pointing from the centroid of passenger j to the centroid of passenger i; ij λ is the orientation correction coefficient for the sociopsychological forces between passenger i and passenger j. i is the orientation weighting coefficient for passenger i, used to measure the passenger's sensitivity to the direction behind them. Its value is between 0 and 1, and the larger the value, the closer the sensitivity to the direction behind them is to the direction in front of them. For the direction that passenger i is facing and The angle between vectors;

[0030] The repulsive force between the passenger and the bus obstacle is represented by the following dynamic equation:

[0031]

[0032]

[0033]

[0034] In the formula, d ij This represents the shortest distance from passenger i's center of mass to the obstacle; Let represent the unit vector from the obstacle to the centroid of passenger i; the attraction force between the passenger and the seat is represented by the following dynamic equation:

[0035]

[0036] In the formula, F is the fixed value of the attraction of the nearest empty seat; d ij The distance from the passenger's center of mass to the seat; Let represent the unit vector from the seat to the centroid of passenger i;

[0037] The repulsive force between the passengers and the rear of the bus is represented by the following dynamic equation:

[0038]

[0039] In the formula, F is a fixed value of the repulsive force exerted by the rear vehicle body on the passenger.

[0040] Preferably, the step of obtaining the relationship between the number of passengers on the bus and their distribution before and after using the particle swarm optimization algorithm to obtain the passenger distribution pattern includes:

[0041] Mathematical modeling of the seating area and vehicle body of the bus;

[0042] Divide the space on the bus into grids based on the minimum space occupied by each person;

[0043] Randomly generate particles representing the number of passengers, randomly distribute the generated particles in the divided bus grid, and calculate the sum of the social forces of all passengers in each case.

[0044] Let the particles update their positions to the new net force according to their respective resultant force directions and magnitudes. The update formula is as follows:

[0045] Speed ​​update formula:

[0046]

[0047] The displacement update formula is:

[0048]

[0049]

[0050] In the formula, θ is the particle velocity and the angle between the direction of the resultant force on the particle and the direction of travel; [*] represents the rounding function in units of grid division;

[0051] After multiple updates, the scenario with the greatest sum of social forces among all passengers, that is, the scenario where all passengers are most satisfied and feel most comfortable, is the most likely passenger distribution under the current number of passengers.

[0052] By varying the number of passengers and repeating the above steps, the distribution pattern of passengers on the bus can be obtained.

[0053] Preferably, the uncertain parameters include vehicle mass, front and rear vehicle mass distribution, and distance from the front and rear axles to the center of gravity; the step of establishing an active suspension model that adapts to changes in mass distribution and center of gravity position after determining the range of uncertain parameters includes, given the real-time number and distribution of passengers, considering the weight distribution of each person, setting the minimum weight to 45kg and the maximum weight to 90kg, then the uncertain range of vehicle mass, front and rear vehicle mass, and distance from the front and rear axles to the center of gravity is:

[0054]

[0055] In the formula, m s0 Let n be the empty load mass of the bus, n be the number of passengers, and λ be the proportion of the first passengers to the total number of passengers, obtained from the passenger distribution pattern.

[0056] The parameter uncertainties of the above parameters can be described by their nominal values ​​and possible perturbation ranges:

[0057]

[0058] in: These represent the nominal values ​​of the bus body mass, front and rear body mass, and distance between the centers of gravity of the front and rear wheelbases, respectively; that is, the mean values ​​of the corresponding interval ranges. s m sf m sr a and b represent the perturbation values ​​of the system parameters, respectively;

[0059] δ i (i = ms, msf, msr, a, b) represents the time-varying function δ i (t) is used to describe the perturbation range of both and |δ i (t)|≤1;

[0060] d ms ,d mu ,d l The perturbation range for five perturbation quantities.

[0061] Through linear fractional transformation, m in the above equation s m sf m sr a and b are represented as follows:

[0062]

[0063]

[0064]

[0065]

[0066]

[0067] The state-space equation for the suspension system with uncertain system parameters is then:

[0068]

[0069] p i =δ(t)q i

[0070] This is transformed into state-space control equations, as follows:

[0071]

[0072] In the formula, ΔA(t) and ΔB(t) are uncertain time-varying function matrices of appropriate dimension, representing the predetermined parameter perturbations of the system model, as follows:

[0073]

[0074]

[0075]

[0076]

[0077] σ is selected as the parameter characterizing the vehicle body mass and the range of front and rear vehicle body mass perturbations; d l The parameters selected to characterize the vehicle body mass and the range of front and rear vehicle body mass perturbations;

[0078] To separate the variables of the uncertain matrices ΔA and ΔB containing δ(t), ΔA and ΔB are expressed in the following norm-bounded form:

[0079] [ΔA(t) ΔB(t)]=Hδ(t)[E1 E2]

[0080] Where H = B p E1 = C q E2 = D zp ;

[0081] Let the state feedback control law be:

[0082] u(t)=Kx(t)

[0083] Applying the control law, the following closed-loop system is obtained:

[0084]

[0085] in:

[0086]

[0087] Preferably, in the step of suppressing vibration during the operation of the rear motor using PID control, the differential equation of the PID control is:

[0088]

[0089] In the formula, K p K represents the proportional gain in a PID controller. i K represents the integral gain in a PID controller. d U is the derivative gain in the PID controller; U is the control signal transmitted by the PID controller to the controlled object, that is, the additional control force of the rear axle suspension actuator in the active suspension for motor vibration.

[0090] Preferably, the step of deriving H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived for changes in mass distribution and center of mass position, and the hybrid H2 / H ... ∞ The steps of active suspension control using the optimal performance-preserving control law include:

[0091] Design a state feedback control law for the closed-loop system, considering the uncertainties in the stiffness coefficient and damping coefficient of the suspension system, so that the closed-loop system meets the following design specifications:

[0092] a. Closed-loop systems are asymptotically stable;

[0093] b. When w(t) is treated as a disturbance input signal with finite energy, the transfer function of the closed-loop control system from the disturbance input w(t) to the control output z(t) satisfies:

[0094]

[0095] in, σ max γ represents the maximum singular value of the matrix, and γ represents the degree of perturbation suppression.

[0096] c. When w(t) is considered as a unit-intensity zero-mean white noise signal, the closed-loop transfer function ||T| from the interference input w(t) to the output z2(t) is... z2w The performance index of ||2 is denoted as J(K), which satisfies:

[0097]

[0098] if If it is asymptotically stable, then J(K) can be expressed as:

[0099]

[0100] in, The continuous-time Lyapunov equations satisfying the closed-loop system form are:

[0101] Where J(K) represents the upper bound of the performance index H2 of the closed-loop system, and the control law that meets the above design requirements is called the H2 / H of the closed-loop system. ∞ The performance-preserving control law that minimizes J(K) is the hybrid H2 / H ∞ Optimal performance-preserving control law;

[0102] For a system with an uncertain matrix δ(t), taking a constant γ > 0, assume the following linear convex optimization problem exists:

[0103]

[0104]

[0105]

[0106] Then there exists a feasible solution α. ms α msf α msr α a α b Given β, X, V, and N, the state feedback control law is u(t) = VX. -1 x(t) is the hybrid H2 / H of the active suspension system. ∞ The optimal performance-preserving control law, with the upper bound of performance preservation for H2 being...

[0107] in:

[0108]

[0109] Preferably, it also includes building H2 / H in MATLAB or Simulink. ∞ A bus suspension model using PID control and PID control is presented, and simulation verification is performed using relevant parameters. The following three modes are discussed and analyzed:

[0110] Case 1: Passive suspension affected by motor vibration, the system only has the passive spring force and damping force of the passive suspension;

[0111] Case 2: H2 / H affected by motor vibration ∞ Control the active suspension, the system at H2 / H ∞ Under the control of the controller;

[0112] Case 3: H2 / H affected by motor vibration ∞ Hybrid PID control active suspension, the system at H2 / H∞ Under the control of a hybrid PID controller.

[0113] An active suspension control system for an electric bus includes:

[0114] The active suspension model building module is used to establish a half-vehicle dynamics model of the bus based on the bus parameters and motor distribution, construct the state space equation of the model, and obtain an active suspension model suitable for the bus.

[0115] The active suspension model adjustment module is used to establish a social force model for the active suspension model, taking into account the changes in mass distribution and center of gravity position caused by passengers getting on and off the bus and moving around inside the vehicle; and to obtain the relationship between the number of passengers on the bus and the front and rear distribution of passengers through the social force model using the particle swarm optimization algorithm, thereby obtaining the passenger distribution law, and establishing an active suspension model that adapts to changes in mass distribution and center of gravity position after determining the range of uncertain parameters.

[0116] The PID control module is used to suppress the vibration of the rear motor during operation by using PID control on an active suspension model that adapts to changes in mass distribution and center of gravity position.

[0117] H2 / H ∞ The control module is used to derive H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived for changes in mass distribution and center of mass position, and the hybrid H2 / H ... ∞ The optimal performance-preserving control law is used for active suspension control.

[0118] Compared with the prior art, the present invention has at least the following beneficial effects:

[0119] The semi-vehicle active suspension model proposed in this invention, which considers the changes in the mass distribution of a bus, can effectively account for the parameter changes caused by passengers getting on and off the bus and by passengers moving around inside the vehicle. The H2 / H ... ∞ The control effectively considers these parameter perturbations, greatly improving the suspension system's parameter insensitivity. This makes it applicable to bus suspension control, significantly enhancing ride comfort. The invention proposes adding a separate PID controller for motor vibration to the rear axle where the motor is located, effectively mitigating high-frequency vibrations caused by the motor and improving passenger comfort in the bus's position above the motor. The H2 / H control proposed in this invention... ∞ The hybrid PID control method can improve the ride comfort of bus suspension. Simulation verification has confirmed the effectiveness and accuracy of the method of this invention, which improves the application of vehicle suspension on buses and is of great significance for the application of active suspension systems on buses. Summary of the Invention

[0121] To more clearly illustrate the technical solutions in this application, the drawings used in the application description will be briefly introduced below. The drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0122] Figure 1 Flowchart of the active suspension control method for electric buses according to an embodiment of the present invention;

[0123] Figure 2 A schematic diagram of the active suspension system model according to an embodiment of the present invention;

[0124] Figure 3 A schematic diagram of an active suspension model adapted to changes in mass distribution and center of mass position in an embodiment of the present invention;

[0125] Figure 4 The principle block diagram of obtaining passenger distribution patterns based on the social force model in this embodiment of the invention;

[0126] Figure 5 The H2 / H proposed in the embodiments of the present invention ∞ Block diagram of the hybrid PID control method;

[0127] Figure 6 Rear axle vehicle body acceleration curve in parking state according to an embodiment of the present invention;

[0128] Figure 7 Spectrum analysis diagram of rear axle vehicle body acceleration in parking state according to an embodiment of the present invention;

[0129] Figure 8 Rear axle vehicle body acceleration curve in driving state according to an embodiment of the present invention;

[0130] Figure 9 Rear axle vehicle body acceleration spectrum analysis diagram in driving state according to an embodiment of the present invention;

[0131] Figure 10 Bar chart showing the improvement in vehicle acceleration of the suspension control method under different mass distributions in embodiments of the present invention;

[0132] Figure 11 A bar chart showing the improvement in pitch angle acceleration of the suspension control method under different mass distributions in embodiments of the present invention. Detailed Implementation

[0133] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0134] This invention proposes an active suspension control method for buses driven by a rear-mounted, rear-drive motor to address existing bus comfort issues. Based on a half-model of the bus, this method designs an H2 / H ratio that can adapt to changes in vehicle mass distribution caused by the bus's excessive length. ∞ In addition to the control law, a PID controller was designed to power the rear axle active suspension in response to the high-frequency vibration caused by the rear-mounted motor, which greatly eliminated the high-frequency vibration caused by the motor.

[0135] Please see Figure 1 The active suspension control method for electric buses according to embodiments of the present invention includes the following steps:

[0136] A half-vehicle dynamics model of the bus is established based on the bus parameters and motor distribution. The state space equation of the model is constructed to obtain an active suspension model suitable for the bus.

[0137] For the active suspension model, considering the changes in mass distribution and center of mass position caused by passengers getting on and off the bus and moving around inside the vehicle, a social force model is established; and the relationship between the number of passengers on the bus and the front and back distribution of passengers is obtained through the social force model using the particle swarm optimization algorithm, so as to obtain the passenger distribution law. After determining the range of uncertain parameters, an active suspension model that adapts to changes in mass distribution and center of mass position is established.

[0138] Using an active suspension model that adapts to changes in mass distribution and center of gravity position as the control object, PID control is used to suppress the vibration of the rear motor during operation.

[0139] By deriving H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived for changes in mass distribution and center of mass position, and the hybrid H2 / H ... ∞ The optimal performance-preserving control law is used for active suspension control.

[0140] In one possible implementation, a half-vehicle dynamics model of the bus is established based on the bus parameters and motor distribution, and the state-space equations of the model are constructed to obtain an active suspension model suitable for the bus, specifically including:

[0141] See Figure 2According to Newton's second law, the dynamic equation of the half-model of the electric bus with added motor vibration is:

[0142]

[0143] In the formula, m m m s m uf m ur These are the masses of the motor, body, front suspension, and rear suspension, respectively; I y k is the moment of inertia. sf k sr k tf k tr k m For the stiffness of the front and rear suspension, tires, and motor mounts; C sf C sr C m These are the damping coefficients for the front and rear suspensions and the motor mount; denoted as vertical and pitch acceleration of the vehicle body; a and b are the distances from the center of the front and rear suspensions to the center of mass of the vehicle body. x uf x ur The vertical acceleration, velocity, and displacement of the front and rear suspensions; x sf x sr x sm x m For the front and rear axles of the vehicle body, the vertical velocity and displacement of the vehicle body and motor at the motor location; x gf x gr Input for the front and rear road surfaces; u f u r F provides power to the front and rear suspensions. m This is the excitation force of the motor.

[0144] Select state variable X and road surface disturbance W, and rewrite the dynamic model of the bus active suspension into a space equation based on state variable X and road surface disturbance W:

[0145]

[0146] In the formula:

[0147]

[0148]

[0149] U = [u f u r ] T ,

[0150] Furthermore, for the active suspension model, considering the significant changes in mass distribution and center of mass position caused by frequent boarding and alighting of bus passengers and their movement inside the vehicle, a social force model is established, and the relationship between the number of passengers on the bus and their front-to-back distribution is obtained through a particle swarm optimization algorithm. The passenger distribution pattern is obtained, and after determining the approximate range of uncertain parameters, an active suspension model that adapts to changes in mass distribution and center of mass position is established.

[0151] This invention takes into account the significant changes in mass distribution and center of gravity caused by frequent boarding and alighting of bus passengers and their movement inside the vehicle. Therefore, the vehicle body mass, front and rear vehicle body mass distribution, and distance from the front and rear axles to the center of gravity are considered as uncertain parameters. Since the number of bus passengers is large, the range of changes in vehicle body mass is large. Therefore, it is necessary to determine the range of changes in mass to ensure the control effect.

[0152] In this embodiment of the invention, the number of passengers on a bus is relatively easy to obtain in real time. Therefore, the relationship between the number of passengers and their distribution is established. First, a social force model on the bus is established. The original social force model believes that pedestrians will be subjected to the combined action of multiple force vectors when they are moving. These mainly include the driving force of pedestrians moving towards the target, the interaction force between pedestrians, and the interaction force between pedestrians and obstacles. In addition, it also includes random and non-systematic forces generated by attraction behavior, aggregation behavior, and disturbance behavior.

[0153] This invention only considers the passenger's position selection on the bus in a static state. Therefore, in this system, the social forces experienced by the passenger on the bus include the repulsive force between passengers, the repulsive force between the passenger and obstacles such as the bus body, the attractive force between the passenger and the seat, the repulsive force between the passenger and the rear of the bus, as well as random and non-systematic forces generated by attraction behavior, aggregation behavior, and disturbance behavior. The dynamic equation is as follows:

[0154]

[0155] In the formula, It is the resultant force of all social forces acting on passenger i during its movement; The repulsive force between passenger i and passenger j; The repulsive force between passenger i and the obstacle; The attraction between passenger i and the nearest seat; The repulsive force between passenger i and the rear of the bus; The random and non-systematic forces generated for passenger i.

[0156] When passengers are on a bus, for various physiological and psychological reasons, they always try to maintain a certain distance from other passengers, especially strangers, in order to gain more space. When other unfamiliar passengers get too close, passengers feel uncomfortable and stressed, thus generating a repulsive force effect and moving away from other passengers. In social force models, this phenomenon is usually reflected by repulsive force or social psychological force. Social psychological force is related to the distance between passengers; generally, the closer the distance, the greater the force, but it will not exceed a certain reasonable maximum value. Furthermore, due to the limitations of the passenger's perspective, differences in direction and relative position can also affect the magnitude of the social psychological force. Passengers tend to be more sensitive to passengers facing them (i.e., those in front of them) and less sensitive to passengers facing them (i.e., those behind them). The social psychological force exerted on passenger i by passenger j during its movement can be represented by the following dynamic equation:

[0157]

[0158]

[0159]

[0160] In the formula, A i r is the influence intensity coefficient of passenger i; ij d is the sum of the radii of passenger i and passenger j. ij B is the distance between the centroids of passenger i and passenger j; i The influence range coefficient for passenger i; w is the unit direction vector pointing from the centroid of passenger j to the centroid of passenger i; ij λ is the orientation correction coefficient for the sociopsychological forces between passenger i and passenger j. i is the orientation weighting coefficient for passenger i, used to measure the passenger's sensitivity to the direction behind them. Its value is between 0 and 1, and the larger the value, the closer the sensitivity to the direction behind them is to the direction in front of them. For the direction that passenger i is facing and The angle between vectors.

[0161] Similar to the forces between passengers, passengers will subconsciously maintain a distance from obstacles on the vehicle body during movement, but they will almost never come into contact with the obstacles to generate squeezing or frictional forces. It can be considered that the force between passengers and obstacles in this system is only a socio-psychological force, and its dynamic equation is as follows:

[0162]

[0163]

[0164]

[0165] In the formula, d ij The shortest distance from passenger i's center of mass to the obstacle; Let be the unit vector from the obstacle to the centroid of passenger i.

[0166] Meanwhile, passengers on buses will prioritize sitting in empty seats. Therefore, the socio-psychological force between passengers and seats manifests as attraction. Since the space on a bus is limited and the status of almost all seats can be seen, this attraction is set here as a fixed value of attraction between the passenger and the nearest empty seat. When there are no empty seats, standing passengers need something to hold onto, such as a seat or overhead handrail, which will also exhibit a certain degree of attraction. The dynamic equations are as follows:

[0167]

[0168] In the formula, F is the fixed value of the attraction of the nearest empty seat; d ij The distance from the passenger's center of mass to the seat; Let represent the unit vector from the seat to the centroid of passenger i.

[0169] Because the rear of the bus is directly above the engine or motor, the seating comfort is not as good as other parts of the bus. Furthermore, it's inconvenient to get off at the rear and feels more cramped than the front. Therefore, under the same conditions, passengers tend to sit at the front. This phenomenon manifests as a repulsive force between the passenger and the rear of the bus, and its dynamic equation is shown below:

[0170]

[0171] In the formula, F is a fixed value of the repulsive force exerted by the rear vehicle body on the passenger.

[0172] like Figure 4 As shown in the diagram, a simple social force diagram of a random passenger on a bus shows that the passenger is subjected to both repulsive and attractive forces from other passengers and other obstacles such as seats inside the bus.

[0173] Therefore, given the number of passengers, a global optimization can be performed using a social force model following the principles of particle swarm optimization, as shown in the attached diagram. Figure 3 The simulation revealed the passenger distribution pattern on the bus, as follows:

[0174] As attached Figure 4 Mathematical modeling of the seating area and vehicle body of the bus;

[0175] The space on the bus was then divided into grids based on the minimum space occupied by each person.

[0176] Then, particles representing the number of passengers are randomly generated and randomly distributed in the divided bus grid. The sum of the social forces of all passengers in each case is calculated.

[0177] Let the particles update their positions to the new net force according to their respective resultant force directions and magnitudes. The update formula is as follows:

[0178] Speed ​​update formula:

[0179]

[0180] The displacement update formula is:

[0181]

[0182]

[0183] In the formula, θ is the particle velocity and the angle between the direction of the resultant force on the particle and the direction of travel; [*] represents the rounding function in units of grid division.

[0184] After multiple updates, the scenario with the greatest sum of social forces among all passengers, that is, the scenario where all passengers are most satisfied and feel most comfortable, is the most likely passenger distribution under the current number of passengers.

[0185] By varying the number of passengers and repeating the above steps, the distribution pattern of passengers on the bus can be obtained.

[0186] Given the real-time passenger count and distribution, and considering the weight distribution of each individual, with the minimum weight set at 45kg and the maximum weight at 90kg, the uncertain ranges for the vehicle's mass, the front and rear vehicle masses, and the distance from the front and rear axles to the center of gravity are as follows:

[0187]

[0188] In the formula, m s0 Let λ be the empty load mass of the bus, n be the number of passengers, and λ be the proportion of the first passengers to the total number of passengers, derived from the passenger distribution pattern.

[0189] The parameter uncertainties of the above parameters can be described by their nominal values ​​and possible perturbation ranges:

[0190]

[0191] in: These represent the nominal values ​​of the bus body mass, front and rear body mass, and distance between the centers of gravity of the front and rear wheelbases, respectively; that is, the average values ​​within the aforementioned range. s m sf m sr a, b represent the perturbation values ​​of the system parameters, respectively; δi (i = ms, msf, msr, a, b) can be expressed as a time-varying function δ i (t) is used to describe the perturbation range of both and |δ i (t)|≤1;d ms ,d mu ,d l The perturbation range for five perturbation quantities.

[0192] Through linear fractional transformation, m in the above equation s m sf m sr a and b can be represented as:

[0193]

[0194]

[0195]

[0196]

[0197]

[0198] The state-space equation for the suspension system with uncertain system parameters is then:

[0199]

[0200] p i =δ(t)q i

[0201] This is transformed into state-space control equations, as follows:

[0202]

[0203] In the formula, ΔA(t) and ΔB(t) are uncertain time-varying function matrices of appropriate dimension, representing the predetermined parameter perturbations of the system model, as follows:

[0204]

[0205]

[0206]

[0207]

[0208] Since the mass variable exists in the denominator of the matrix variables, σ needs to be selected as the parameter to characterize the perturbation range of the vehicle body mass and the front and rear vehicle body mass. For example, when σ is 0.25, the robustness of the vehicle body mass and the front and rear vehicle body mass within the variation range of (0.8m to 1.2m) can be guaranteed; at the same time, d l The parameters selected to characterize the vehicle body mass and the range of front and rear vehicle body mass perturbations, for example, when d l When the value is 0.2, robustness can be guaranteed within the range of (0.8l~1.2l) of the distance from the front and rear axles to the center of mass.

[0209] To separate the variables of the uncertain matrices ΔA and ΔB containing δ(t), the two equations can be combined to write ΔA and ΔB in the following norm-bounded form:

[0210] [ΔA(t) ΔB(t)]=Hδ(t)[E1 E2]

[0211] in,

[0212] H = B p E1 = C q E2 = D zp

[0213] Let the state feedback control law be:

[0214] u(t)=Kx(t)

[0215] Applying the control law, the following closed-loop system is obtained:

[0216]

[0217] in:

[0218]

[0219] In one possible implementation, embodiments of the present invention design PID control to adapt to the vibration conditions of the motor under different operating conditions, in response to the direct interference of the rear motor vibration.

[0220] The working principle of PID control is to perform proportional, integral, and derivative operations on the feedback quantity e(t) to obtain the control quantity, and then transmit the signal to the 1 / 2 vehicle active suspension model, thereby realizing the control of the 1 / 2 vehicle active suspension model. The differential equation of PID control is:

[0221]

[0222] In the formula: K p K is the proportional gain in a PID controller. iK is the integral gain in a PID controller. d U is the derivative gain in the PID controller, and U is the control signal transmitted by the PID controller to the controlled object (1 / 2 vehicle active suspension model), which is the additional control force of the rear axle suspension actuator in the active suspension against the motor vibration.

[0223] This invention adds an additional controller to the rear axle where the motor is located, and uses PID control to suppress the vibration of the motor during operation. Since the motor is a rear-mounted motor, it has a direct impact on the vibration of the rear axle. Therefore, the input is selected as the dynamic stroke of the motor mount, and the output is the active motion adjustment of the rear suspension. The parameters of the PID controller are obtained through empirical adjustment.

[0224] In one possible implementation, embodiments of the present invention derive H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived when the mass distribution and the position of the center of mass change.

[0225] Design a state feedback control law for the closed-loop system, considering the uncertainties in the stiffness coefficient and damping coefficient of the suspension system, so that the closed-loop system meets the following design specifications:

[0226] (a) The closed-loop system is asymptotically stable.

[0227] (b) When w(t) is treated as a disturbance input signal with finite energy, the transfer function of the closed-loop control system from the disturbance input w(t) to the control output z(t) satisfies:

[0228]

[0229] in, σ max γ represents the maximum singular value of the matrix, and γ represents the degree of perturbation suppression.

[0230] (c) When w(t) is considered as a unit-intensity zero-mean white noise signal, the closed-loop transfer function from the interference input w(t) to the output z2(t) is... The performance index can be expressed as J(K), which satisfies:

[0231]

[0232] if If it is asymptotically stable, then J(K) can be expressed as:

[0233]

[0234] in, The continuous-time Lyapunov equations satisfying the closed-loop system form are:

[0235] Where J(K) represents the upper bound of the performance index H2 of the closed-loop system, and the control law that meets the above design requirements is called the H2 / H of the closed-loop system. ∞ Performance-preserving control law. J(K) typically depends on the chosen control law; the control law that minimizes J(K) is called the hybrid H2 / H of the system. ∞ Optimal performance-preserving control law.

[0236] For a system with an uncertain matrix δ(t), taking a constant γ > 0, assume the following linear convex optimization problem exists:

[0237]

[0238]

[0239]

[0240] Then there exists a feasible solution α. ms α msf α msr α a α b Given β, X, V, and N, the state feedback control law is u(t) = VX. -1 x(t) is the hybrid H2 / H of the active suspension system. ∞ Optimal performance-preserving control law, H2 performance-preserving upper bound

[0241] in:

[0242]

[0243] In one possible implementation, embodiments of the present invention also involve establishing an H2 / H-based system in Simulink. ∞ A bus suspension model using PID control and ADC was developed, and then simulation verification was performed using the relevant parameters.

[0244] This invention aims to verify the constructed H2 / H ∞ The effectiveness of the hybrid PID controller was investigated by establishing an uncertain model of a bus suspension with a mass distribution variation of 1 / 2 in the MATLAB / Simulink environment, and the accuracy of the controller was verified through simulation. Figure 5 As shown, H2 / H ∞ The hybrid PID controller receives state variables from the suspension system and calculates the control input; the two controllers complement each other to improve bus comfort. For example... Figure 6 , Figure 7 As shown in the rear axle vehicle body acceleration curve and spectrum analysis diagram in the parking state, when only the motor vibrates in the parking state, the simple H2 / H ∞The controller and passive suspension have similar effects, indicating that H2 / H ∞ The controller cannot effectively control the high-frequency vibration interference of the motor, while H2 / H ∞ The hybrid PID controller can significantly reduce high-frequency vibrations at this time, and is very effective in dealing with high-frequency interference from the rear axle motor. The rear axle vehicle acceleration curve and spectrum analysis diagram during driving are shown below. Figure 8 , Figure 9 As shown, under normal driving conditions, H2 / H ∞ The controller can significantly reduce vertical acceleration, but it still has little effect on high-frequency vibrations. As can be seen from the spectrum, H2 / H ∞ The hybrid PID controller not only performs excellently at high frequencies but also shows some improvement in low-frequency vibration. The bar charts showing the improvement in vehicle acceleration and pitch angle acceleration under different mass distributions using different suspension control methods are as follows: Figure 10 , Figure 11 As shown, H2 / H ∞ Controller and H2 / H ∞ The hybrid PID controller exhibits good robustness under different mass distribution conditions. The vehicle acceleration has maintained an improvement of about 60-70%, and the pitch acceleration has maintained an improvement of about 20%. This indicates that the proposed controller can adapt well to the large changes in mass distribution and center of mass position caused by frequent boarding and alighting of bus passengers and their movement inside the vehicle.

[0245] Simulation results demonstrate that the H2 / H proposed in this embodiment of the invention... ∞ The hybrid PID control method can improve the ride comfort of bus suspension, thus verifying the effectiveness and accuracy of the embodiments of the present invention, improving the application of vehicle suspension on buses, and having great significance for the application of active suspension systems on buses.

[0246] Another embodiment of the present invention also proposes an active suspension control system for electric buses, comprising:

[0247] The active suspension model building module is used to establish a half-vehicle dynamics model of the bus based on the bus parameters and motor distribution, construct the state space equation of the model, and obtain an active suspension model suitable for the bus.

[0248] The active suspension model adjustment module is used to establish a social force model for the active suspension model, taking into account the changes in mass distribution and center of gravity position caused by passengers getting on and off the bus and moving around inside the vehicle; and to obtain the relationship between the number of passengers on the bus and the front and rear distribution of passengers through the social force model using the particle swarm optimization algorithm, thereby obtaining the passenger distribution law, and establishing an active suspension model that adapts to changes in mass distribution and center of gravity position after determining the range of uncertain parameters.

[0249] The PID control module is used to suppress the vibration of the rear motor during operation by using PID control on an active suspension model that adapts to changes in mass distribution and center of gravity position.

[0250] H2 / H ∞ The control module is used to derive H2 / H ∞ Based on the control theorem, the active suspension feedback control law is derived for changes in mass distribution and center of mass position, and the hybrid H2 / H ... ∞ The optimal performance-preserving control law is used for active suspension control.

[0251] Another embodiment of the present invention provides an electronic device, comprising: a memory storing at least one instruction; and a processor executing the instructions stored in the memory to implement the active suspension control method for electric buses.

[0252] Another embodiment of the present invention provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the active suspension control method for electric buses.

[0253] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the active suspension control method for electric buses described in this invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, which describe the execution process of the computer program on the server.

[0254] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method of controlling an electric bus active suspension, characterized by, It comprises the following steps: According to the bus parameters and motor distribution, a half-car dynamics model of the bus is established, a state space equation of the model is constructed, and a main suspension model suitable for the bus is obtained. For the main suspension model, the mass distribution change and the center of mass position change caused by the passengers getting on and off the bus and walking in the bus are considered, a social force model is established, and the relationship between the number of passengers on the bus and the passenger front and rear distribution is obtained according to the particle swarm optimization algorithm, and the passenger distribution rule is obtained. The active suspension model suitable for the mass distribution change and the center of mass position change is taken as the control object, and the vibration of the rear motor during operation is suppressed by using PID control. By deriving H2 / H ∞ The control law of active suspension feedback control is obtained when the mass distribution and the position of mass center change, and the mixed H2 / H ∞ Optimal guaranteed cost control law is determined for active suspension control; According to the bus parameters and motor distribution, a half-car dynamics model of the bus is established, a state space equation of the model is constructed, and a main suspension model suitable for the bus is obtained. According to Newton's second law, the half-car dynamics model of the bus is established as follows: wherein, is the mass of the motor, is the mass of the body, is the mass of the front suspension, is the mass of the rear suspension; is the moment of inertia; is the front suspension stiffness, is the rear suspension stiffness, is the front tire stiffness, is the rear tire stiffness, is the motor mount stiffness; is the front suspension damping coefficient, is the rear suspension damping coefficient, is the motor mount damping coefficient; is the body vertical acceleration, is the body pitch angular acceleration; is the distance from the front suspension center to the body center of mass, is the distance from the rear suspension center to the body center of mass; is the front suspension vertical acceleration, is the rear suspension vertical acceleration, is the front suspension velocity, is the rear suspension velocity, is the front suspension displacement, is the rear suspension displacement; is the front axle body vertical velocity, is the rear axle body vertical velocity, is the motor at body vertical velocity, is the motor vertical velocity, is the front axle body vertical displacement, is the rear axle body vertical displacement, is the motor at body vertical displacement, is the motor vertical displacement; is the front road input, is the rear road input; is the front suspension actuation force, is the rear suspension actuation force, is the motor excitation force.

2. The electric bus active suspension control method of claim 1, wherein, In the step of establishing the social force model, the relationship between the number of passengers and the passenger front and rear distribution is obtained in real time, the passengers on the bus are considered to be subjected to the combined action of multiple force vectors when walking, and the social force received by the passengers on the bus includes the repulsive force between passengers, the repulsive force between passengers and bus obstacles, the attractive force between passengers and seats, the repulsive force and attractive behavior, gathering behavior and disturbance behavior behind the bus.

3. The method of claim 2, wherein The social force received by the passengers on the bus is established as follows: In the formula, It is the resultant force of all social forces acting on passenger i during movement; For passengers With passengers The repulsive force between them; For passengers Repulsive force between the object and the obstacle; For passengers The attraction between the nearest seat; For passengers The repulsive force from the rear of the bus; For passengers The resulting random and non-systematic forces.

4. The method of claim 1, wherein The step of obtaining the relationship between the number of passengers on the bus and the passenger front and rear distribution according to the particle swarm optimization algorithm includes: Mathematical modeling of the seat area and the body of the bus; Divide the space on the bus into grids by occupying the minimum space for each person; Randomly generate particles of the number of passengers, randomly distribute the generated particles in the divided bus grid, and calculate the sum of the social force of all people in each case; Let the particles update their positions in the new social force according to their respective force directions and sizes; After multiple updates, the case with the maximum sum of the social force of all people is obtained, that is, the most comfortable case for all passengers under the current number of passengers is the most possible passenger distribution; Obtain the passenger distribution rule on the bus.

5. The method of claim 1, wherein The uncertain parameters include the body mass, the front and rear body mass distribution, and the front and rear axle to the center of mass distance; The step of establishing the active suspension model suitable for the mass distribution change and the center of mass position change after determining the range of uncertain parameters includes setting the minimum weight to 45kg and the maximum weight to 90kg considering the weight distribution of each person under the known real-time passenger number and passenger distribution.

6. The electric bus active suspension control method of claim 1, wherein, In the step of suppressing the vibration of the rear motor during operation by using PID control, the differential equation of PID control is: In the formula, is a proportional gain in the PID controller; is an integral gain in the PID controller; is a differential gain in the PID controller; is a control signal delivered to the controlled object by the PID control, i.e., a control force of the rear axle suspension actuator in the active suspension that is additionally directed against the motor vibration; denotes a deviation of the feedback value from the desired value.

7. The method of claim 1, wherein, Also included is the establishment of H2 / H ∞ The bus suspension model of the control and PID control is simulated and verified by combining the corresponding parameters, and the following three modes are discussed and analyzed: case1: passive suspension with motor vibration influence, the system only has passive spring force and damping force of passive suspension; case2: H2 / H with motor vibration influence ∞ Control active suspension, system in H2 / H ∞ Control under control of the controller; case3: H2 / H with motor vibration influence ∞ Hybrid PID controlled active suspension, the system is in H2 / H ∞ under the control of hybrid PID control controller.

8. An electric bus active suspension control system, characterized by, It comprises: The active suspension model construction module is configured to establish a half-car dynamics model of the bus according to bus parameters and motor distribution, construct a state space equation of the model, and obtain an active suspension model suitable for the bus. The active suspension model adjustment module is configured to consider the mass distribution change and the center of mass position change caused by the passengers getting on and off the bus and walking in the bus, establish a social force model, and obtain the relationship between the number of passengers on the bus and the front and rear distribution of the passengers and the passenger distribution law by using the particle swarm optimization algorithm according to the social force model, so as to establish the active suspension model suitable for the mass distribution change and the center of mass position change after the range of the uncertain parameters is determined. The PID control module is configured to use the active suspension model suitable for the mass distribution change and the center of mass position change as a control object and use PID control to suppress the vibration when the rear motor operates. H2 / H ∞ a control module for deriving H2 / H ∞ control theorem to derive the active suspension feedback control law when the mass distribution and the center of mass position change, and to determine the mixed H2 / H ∞ optimal guaranteed cost control law for active suspension control; The active suspension model construction module is configured to establish a half-car dynamics model of the bus according to bus parameters and motor distribution, construct a state space equation of the model, and obtain an active suspension model suitable for the bus. According to Newton's second law, the half-car dynamics model of the bus established according to the bus parameters and the motor distribution is as follows: wherein, is the mass of the motor, is the mass of the body, is the mass of the front suspension, is the mass of the rear suspension; is the moment of inertia; is the front suspension stiffness, is the rear suspension stiffness, is the front tire stiffness, is the rear tire stiffness, is the motor mount stiffness; is the front suspension damping coefficient, is the rear suspension damping coefficient, is the motor mount damping coefficient; is the body vertical acceleration, is the body pitch angular acceleration; is the distance from the front suspension center to the body center of mass, is the distance from the rear suspension center to the body center of mass; is the front suspension vertical acceleration, is the rear suspension vertical acceleration, is the front suspension velocity, is the rear suspension velocity, is the front suspension displacement, is the rear suspension displacement; is the front axle body vertical velocity, is the rear axle body vertical velocity, is the motor at body vertical velocity, is the motor vertical velocity, is the front axle body vertical displacement, is the rear axle body vertical displacement, is the motor at body vertical displacement, is the motor vertical displacement; is the front road input, is the rear road input; is the front suspension actuation force, is the rear suspension actuation force, is the motor excitation force.

Images