New energy buses and their speed control methods and systems

Abstract

This invention discloses a new energy bus and its speed control method and system. Considering the effects of starting, driving, and braking conditions, different confidence factors are set. Through these confidence factors, the vehicle speed estimator is adaptively adjusted, solving the key problem of inaccurate longitudinal speed acquisition under extreme conditions. This invention not only requires less computation but also achieves high accuracy.

Description

Technical Field

[0001] This invention relates to the field of vehicle intelligent control technology, and in particular to a new energy bus and its speed control method and system. Background Technology

[0002] Wheel-side drive systems eliminate the mechanical differential of traditional central direct-drive systems. Instead, they connect to the drive wheels via wheel-side motors and reduction mechanisms. Under extreme conditions, such as wheel lock-up or slippage, the wheel speed can deviate significantly from the vehicle speed. Since longitudinal vehicle speed is a crucial input parameter for vehicle stability control in wheel-side drive systems, accurately estimating the current vehicle speed based on vehicle status and wheel speed information is one of the key challenges that wheel-side drive system control urgently needs to address.

[0003] In new energy buses with wheel-side drive structures, the rear axle is driven by a wheel-side motor while the front axle is a non-drive wheel. Therefore, the vehicle's current speed cannot be determined solely by the motor's rotational speed or the wheel speed feedback from the ABS. Because the motors are independently driven and their torque can be precisely controlled, distributed drive electric vehicles can fully utilize their unique advantages to achieve numerous safety controls.

[0004] Longitudinal vehicle speed is a crucial parameter for vehicle stability control and forms the foundation for its implementation. Accurately estimating longitudinal vehicle speed, taking into account vehicle structure and driving conditions, is both a key focus and a challenge in achieving distributed drive system control.

[0005] Currently, the main methods for estimating longitudinal vehicle speed both domestically and internationally include: 1. Kinematic model-based methods, which are robust and the estimation results are almost unaffected by model parameters, but require high accuracy of sensor information. 2. Dynamic model-based observers, which have relatively lower requirements for sensors, but are more sensitive to model parameters.

[0006] The invention patent application "Calculation method for vehicle speed and road slope of electric vehicle" (publication number: CN 114547782 A) is based on the vehicle dynamics model and uses fuzzy control rules to determine the Kalman filter factor, and then estimates the vehicle speed and slope. It has high requirements for the sensitivity of model parameters and the accuracy of sensors. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a new energy bus and its speed control method and system to accurately estimate the vehicle speed in the absence of a longitudinal acceleration sensor, in order to address the shortcomings of the existing technology.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a speed control method for new energy buses, comprising the following steps:

[0009] Estimate the vehicle speed v(k+1) at time k+1 using the following formula:

[0010] in, P - (k)=P(k-1)+Q, P(k)=(IK(k)H)P - (k), and P k (k) represents the covariance between the predicted value and the true value at time k, and the covariance between the optimal estimate and the true value, respectively. R - This is the corrected observation error covariance matrix. R is the observation error covariance matrix, α is the confidence factor, α = [α fl α fr α rl α rr ] T α fl α fr α rl α rr Let Q be the confidence factor of the observation error covariance of the front left, front right, rear left, and rear right wheels, respectively; Q be the error covariance; K(k) be the Kalman gain at time k; a(k) be the estimated acceleration at time k; and v be the weight of the wheel. m (k+1) and Let v(k+1) be the wheel speed at time k+1 and the estimated wheel speed, respectively. Let H represent the vehicle speed and estimated speed at time k+1, respectively. H = [1 1 1 1] T Ts is the sampling period;

[0011] The process of determining the credibility factor α includes:

[0012] Under driving conditions, reduce the reliability factor of the slipping wheels;

[0013] Under braking conditions, reduce the reliability factor of wheel lock-up.

[0014] This invention considers the varying degrees of influence of wheel speeds on vehicle speed estimation under different operating conditions. Under driving conditions, the front axle wheels are essentially in a pure rolling state, therefore the initial value of the observation error covariance of the front axle wheels is smaller than that of the rear axle wheels. Under braking conditions, since all wheels are at risk of locking up, the initial values ​​of the observation error covariance of the front and rear axle wheel speeds are the same. This invention solves the key problem of inaccurately obtaining longitudinal vehicle speed under extreme conditions. In the absence of a longitudinal acceleration sensor, a tracking differentiator for acceleration estimation and an adaptive Kalman filter-based vehicle speed estimator are designed, addressing the high requirements of existing technologies for model parameters and sensor accuracy. Simulation and real-vehicle testing show that the method of this invention not only has low computational complexity but also high accuracy.

[0015] In this invention, when the vehicle starts (i.e., in the starting condition), the wheel speed is calculated based on the motor speed, and the wheel speed is used as the estimated wheel speed at the current moment, which is then substituted into the vehicle speed estimation formula to calculate the current vehicle speed.

[0016] Under starting conditions, the measured wheel speed is: v fl v fr v rl v rr These are the wheel speed measurements for the front left wheel, front right wheel, rear left wheel, and rear right wheel, respectively. r is the tire radius, i is the speed ratio of the wheel-side drive system reducer, and n... l n r These represent the rotational speeds of the left and right motors, respectively.

[0017] Under the initial operating condition, the corrected observation error covariance matrix is: R represents the corrected covariance of observation errors for the front left wheel, front right wheel, rear left wheel, and rear right wheel, respectively. rl,0 R rr,0 These represent the initial values ​​of the observation error covariance for the left and right rear wheels, respectively.

[0018] In this invention, under driving conditions, if |v m，ij (k)-v(k)|>v set If v, then it is determined that the wheel slips; where v m，ij (k) represents the wheel speed of the four wheels, i represents the left or right wheel, j represents the front or rear axle wheel, and v(k) is the estimated current vehicle speed. set The threshold value for the speed difference.

[0019] In this invention, under braking conditions, if |v m，ij (k)-v(k)|>v set If v is the wheel lockup, then the wheel is determined to be locked; where v m，ij (k) represents the wheel speed of the four wheels, and v(k) is the estimated current vehicle speed. set The threshold value for the speed difference.

[0020] To ensure the accuracy of vehicle speed estimation and the stability of vehicle operation, the confidence factor is reduced by no more than 0.2 in each calculation.

[0021] As an inventive concept, the present invention also provides a speed control system for a new energy bus, which includes a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the steps of the method described above.

[0022] As an inventive concept, the present invention also provides a new energy bus that adopts the above-mentioned control system.

[0023] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention considers the effects of starting, driving, and braking conditions, sets different confidence factors, and adaptively adjusts the vehicle speed estimator through these confidence factors, thereby improving the accuracy of the vehicle speed estimator. This invention also has low computational complexity and high accuracy. Attached Figure Description

[0024] Figure 1 This is a diagram of the wheel-side drive system architecture.

[0025] Figure 2 This is a schematic diagram of the vehicle speed estimation system according to an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of the acceleration estimation system according to an embodiment of the present invention;

[0027] Figure 4 This is a flowchart illustrating the credibility factor adjustment process in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Example 1

[0030] Embodiment 1 of the present invention provides a speed control method for a new energy bus, wherein the wheel-side drive system structure applicable to the new energy bus is as follows: Figure 1As shown, wheel-side drive assemblies are installed at the left and right rear wheels, respectively. These assemblies mainly consist of wheel-side motors and reducers, capable of driving the vehicle and also performing regenerative braking. At the front axle, the steering system assembly is the primary component, and the front axle is a non-drive axle. At the perception layer, the system is equipped with ABS as standard, which provides high-precision wheel speed signals. A steering wheel angle sensor is installed at the steering wheel to sense the driver's intentions and determine whether the vehicle is traveling in a straight line or turning. A yaw rate sensor is installed at the vehicle's center of gravity to monitor the yaw rate in real time and assess the risk of instability. Wheel speed signals, steering wheel angle signals, and yaw rate signals are key input parameters for electronic differential control. At the control layer, the left and right wheel-side drive assemblies share a single control assembly, which integrates inverter modules for two motor controllers, as well as auxiliary power control systems such as oil pumps and air pumps. Furthermore, the differential control algorithm is integrated into the vehicle controller, eliminating the need for an additional differential controller and effectively reducing costs.

[0031] Since this wheel-side driven new energy bus does not have an acceleration sensor, and longitudinal acceleration is one of the important parameters for vehicle speed estimation, the longitudinal speed estimation system consists of two parts: a longitudinal acceleration estimator and a longitudinal speed estimator. The output of the longitudinal acceleration estimator serves as the input to the longitudinal speed estimator to achieve longitudinal speed estimation. The longitudinal speed estimator employs an estimation method based on adaptive Kalman filtering. It adaptively calculates the Kalman gain according to the current operating conditions, thereby accurately estimating the current vehicle speed based on information such as the high-precision wheel speed of the ABS and the wheel-side motor speed. Its schematic diagram is shown below. Figure 2 As shown.

[0032] Parameter 'a' represents the estimated acceleration output by the acceleration estimator, Vm represents the high-precision wheel speeds of the four wheels fed back by the ABS system, R and Q represent the observation error covariance and system error covariance, respectively, Ts represents the sampling period in seconds, Vpre and Ppre represent the estimated vehicle speed and the estimation error covariance output by the Kalman filter at the previous time step, respectively, and Vpre is also the input to the acceleration estimator. The vehicle speed estimator, based on an adaptive Kalman filter algorithm, filters the estimated acceleration and the high-precision wheel speeds of the four wheels to obtain the estimated vehicle speed at the current time step.

[0033] A tracking differentiator was designed to estimate the current acceleration in real time, assuming the system does not have an accelerometer installed. The schematic diagram of the tracking differentiator is shown below. Figure 3 As shown. The input is velocity V and sampling period Ts, and the output is acceleration. x1 and x2 are internal state variables, which are updated to Dx1 and Dx2 after this calculation. Among them, Dx1 tracks the input signal V and is the filtered velocity value; theoretically, there are... Therefore, Dx2 can be considered as an approximate differential value of the velocity V. Dx1 and Dx2 pass through... The state value at this moment is returned to the next moment, forming a recursive operation. The initial value of Dx1 is x. 10 It is generally taken as the first value of the velocity V. The initial value of Dx2 is x. 20 It is usually set to 0. In this way, the acceleration corresponding to the velocity can be continuously output.

[0034] The iterative formula for the tracking differentiator is as follows:

[0035]

[0036] The inputs are states x1 and x2, velocity value V, and sampling period Ts. The outputs are the updated state values ​​Dx1 and Dx2. h0, h1, v, and f are internal variables. The adjustable parameter r is the speed factor; the larger r is, the faster x1 tracks the signal, but the more susceptible it is to noise. r is typically set to 30. The adjustable parameter K is the step size factor; the larger K is, the better the filtering effect, but the greater the phase loss of signal x1 tracking signal V. K ranges from 1 to 1.5, and is typically set to 1. By appropriately adjusting r and K, a tracking effect with low time delay and low noise pollution can be achieved.

[0037]

[0038] y, h, and a0 are intermediate variables calculated from x1, x2, r, and h. The core function in the tracking differentiator is Fhan(x1, x2, r, h0). The differentiation function is achieved by tracking the input signal as quickly as possible. Optimal control theory is applied, and to avoid chattering when the system enters steady state, numerical calculation is used to derive the comprehensive function Fhan.

[0039] Based on the above three formulas, Figure 3 The recursive calculation can calculate the acceleration value in real time from the velocity value.

[0040] An adaptive Kalman filter estimates and corrects the statistical characteristics of the model and noise while utilizing measurement data, thereby modifying the filter design and filtering error. For wheel-side drive systems, under normal operating conditions, vehicle speed can be calculated based on the rotational speeds of the left and right motors or the high-precision wheel speed feedback from the ABS. However, for low-traction surfaces or extreme conditions, due to wheel slippage or lock-up, vehicle speed cannot be simply calculated from motor speed or wheel speed. Therefore, an adaptive Kalman filter needs to be designed to adjust the covariance of the system's observation error in real time when vehicle slippage or lock-up is detected, thereby improving the accuracy of vehicle speed estimation.

[0041] The system state equation is:

[0042]

[0043] in, This represents the estimated vehicle speed at the current moment, where Ts is the sampling period and a(k) is the estimated acceleration at the current moment.

[0044] The system measurement equation is:

[0045] y(k)=H(k)x(k)+e(k)................................................(2)

[0046] Where H =

[1111] T y = [v fl v fr v rl v rr ] T v ij Let e(k) represent the wheel speeds of the four wheels respectively, and let e(k) be the system error, which is assumed to be 0.

[0047] The Kalman filter iterative equation for the vehicle speed estimation system is:

[0048]

[0049] in, and P k These represent the covariances between the predicted value and the optimal estimate, respectively, and the true value; R is the observation error covariance; Q is the system error covariance; K is the Kalman gain; a is the estimated acceleration; and v is the system error covariance. m and These are the current wheel speed and the estimated wheel speed, v and v', respectively. These represent the current vehicle speed and the estimated vehicle speed, respectively. Through the iteration of the Kalman filter described above, the current vehicle speed can be effectively estimated under good road conditions.

[0050] However, on low-traction surfaces, rapid acceleration or deceleration can easily lead to drive wheel slippage or wheel lockup. In this situation, the impact of slipping or locked wheels on vehicle speed estimation differs from that of wheels that do not slip or lock up. Setting the same confidence factor α for all wheels would severely reduce the accuracy of speed estimation. Therefore, considering wheel states and different operating conditions, an adaptive confidence factor α is designed to improve speed estimation accuracy. This confidence factor α characterizes the confidence of the state estimate; a larger α indicates a higher confidence in the predicted value, and a smaller α indicates a higher confidence in the measured value.

[0051] Analysis of Kalman gain K:

[0052]

[0053] The larger the observation error covariance R, the smaller the Kalman gain K, and the more the system's state value depends on the predicted value; conversely, the smaller the observation error covariance R, the larger the Kalman gain K, and the more the system's state value depends on the measured value. Therefore, different confidence factors α are designed for the observation error covariance R under different operating conditions.

[0054]

[0055] Among them, R - Given the corrected observation error covariance matrix, the corrected Kalman gain is:

[0056]

[0057] From the above formula, we can see that the larger α is, the smaller R is, the larger K is, the higher the reliability of the observed values, and the more the vehicle speed estimate depends on the observed values; the smaller α is, the smaller R is. - The larger the value of K, the smaller the value of K, the higher the reliability of the predicted value, and the more the vehicle speed estimate depends on the prediction result.

[0058] The adjustment rules for the credibility factor are as follows: Figure 4 .

[0059] When a vehicle starts (starting condition) (vehicle speed < 2 km / h), the wheel speed reported by the ABS is 0. If the vehicle speed is estimated using wheel speed at this time, the estimated speed will also be 0. Instead, the wheel speed can be estimated using the motor speed. The wheel speed calculated from the motor speed is then used as input to estimate the current vehicle speed. For a distributed drive system architecture, only the left and right rear wheels are equipped with drive motors. Therefore, during starting, the current vehicle speed is estimated using only the wheel speeds calculated from the left and right motor speeds on the rear axle.

[0060] The measured wheel speed under starting conditions and the corrected observation error covariance matrix are as follows:

[0061]

[0062]

[0063] In the formula, r is the tire radius, i is the speed ratio of the wheel-side drive system reducer, and n l n r R represents the rotational speed of the left and right motors, respectively. rl,0 R rr,0 Let represent the initial values ​​of the observation error covariances for the left and right rear wheels, respectively. These represent the corrected covariances of the observation errors for the front left wheel, front right wheel, rear left wheel, and rear right wheel, respectively. At start-up... The initial value of K is set to a very large value. At this time, K fl K frWhen the speed is close to 0, the estimated vehicle speed depends mainly on the rotational speed of the rear left and rear right wheel motors.

[0064] Under driving conditions, non-driving wheels are essentially in a pure rolling state due to the lack of driving force. Therefore, the confidence factors for the front left wheel and both front and rear wheels should be higher than those for the rear left wheel and rear right wheel under driving conditions. For driving wheels, slippage is highly likely on low-traction surfaces or during rapid acceleration. When driving wheels slip, their wheel speed will be higher than the current vehicle speed. To reduce the impact of driving wheel slippage on the accuracy of vehicle speed estimation, the confidence factor for slipping wheels needs to be further reduced. Under driving conditions, the measured wheel speed and the corrected observation error covariance are as follows:

[0065]

[0066]

[0067] In the formula, v m,fl v m,fr v m,rl v m,rr These are the high-precision wheel speeds of the four wheels fed back by the ABS, R fl,0 R fr,0 R rl,0 R rr,0 Let α be the initial values ​​of the observation error covariances of the front left, front right, rear left, and rear right wheels, respectively. Under driving conditions, the initial values ​​of the front left and front right wheels are equal, the initial values ​​of the rear left and rear right wheels are equal, and the initial value of the front axle wheel is greater than that of the rear axle wheel. fl α fr α rl α rr The confidence factors for the observation error covariance of the front left, front right, rear left, and rear right wheels are given by the following rules:

[0068] Slippage of the drive wheel:

[0069] α fl =α fr =α rl =α rr =1..............................................(11)

[0070] When the drive wheel slips, the confidence factor of each wheel is equal.

[0071] Rear left drive wheel slipping:

[0072]

[0073] Rear right drive wheel slipping:

[0074]

[0075] Both the left and right rear drive wheels slipped:

[0076]

[0077] When drive wheel slippage is detected, the confidence factor of the slipping wheel should be reduced promptly to minimize the impact of the slipping wheel's speed on vehicle speed estimation. The rule for determining whether a wheel is slipping is as follows:

[0078] |v m，ij (k)-v(k)|>v set .............................................(15)

[0079] In the formula, v m,ij (k) represents the wheel speed of each of the four wheels, and v(k) is the estimated current vehicle speed. set This is the threshold for the speed difference. set It can be set to a value below 3.

[0080] Under braking conditions, wheel lock-up is prone to occur on low-traction surfaces or during emergency braking, seriously threatening vehicle safety. When a wheel locks up, its wheel speed is lower than the current vehicle speed. Therefore, when wheel lock-up is detected, the confidence factor of that wheel needs to be adjusted promptly to reduce speed estimation errors.

[0081] During normal braking, when no wheels lock up, there is no need to adaptively adjust the confidence factors of each wheel; the initial values ​​can be maintained. It is worth noting that, unlike the driving condition, the initial values ​​of the covariance matrices of the observation errors for each wheel speed are equal under braking conditions.

[0082] When any wheel locks up, the confidence factor of the locked wheel should be reduced promptly to decrease the impact of the locked wheel speed on vehicle speed estimation during braking. The rules for determining whether a wheel is locked up and the confidence factor adjustment strategy are the same as for driving conditions.

[0083] Example 2

[0084] Embodiment 2 of the present invention provides a control system corresponding to Embodiment 1 above, including a memory, a processor and a computer program stored in the memory; the processor executes the computer program in the memory to implement the steps of the method of Embodiment 1 above.

[0085] In some implementations, the memory may be high-speed random access memory (RAM), and may also include non-volatile memory, such as at least one disk storage device.

[0086] In other implementations, the processor can be any type of general-purpose processor, such as a central processing unit (CPU) or a digital signal processor (DSP), and there is no limitation here.

[0087] Example 3

[0088] Embodiment 3 of the present invention provides a new energy bus corresponding to Embodiment 2 above, which adopts the control system of Embodiment 2. This control system can be integrated with the original control system of the new energy bus, or it can be set up independently of the original control system of the new energy bus.

[0089] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0090] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A speed control method for a new energy bus, characterized in that, Includes the following steps: Estimate the vehicle speed v(k+1) at time k+1 using the following formula: in, P - (k)=P(k-1)+Q, P(k)=(IK(k)H)P - (k), and P k (k) represents the covariance between the predicted value and the true value at time k, and the covariance between the optimal estimate and the true value, respectively. R - This is the corrected observation error covariance matrix. R is the observation error covariance matrix, α is the confidence factor, α = [α fl α fr α rl α rr ] T α fl α ffr α rl α rr Let Q be the confidence factor of the observation error covariance of the front left, front right, rear left, and rear right wheels, respectively; Q be the error covariance; K(k) be the Kalman gain at time k; a(k) be the estimated acceleration at time k; and v be the weight of the wheel. m (k+1) and Let v(k+1) be the wheel speed at time k+1 and the estimated wheel speed, respectively. Let H represent the vehicle speed and estimated speed at time k+1, respectively. H = [1 1 1 1] T Ts is the sampling period; The process of determining the credibility factor α includes: Under driving conditions, reduce the reliability factor of the slipping wheels; Under braking conditions, reduce the reliability factor of wheel lock-up.

2. The speed control method for new energy buses according to claim 1, characterized in that, When the vehicle starts, the wheel speed is calculated based on the motor speed. This wheel speed is then used as the estimated wheel speed at the current moment and substituted into the vehicle speed estimation formula to calculate the current vehicle speed.

3. The speed control method for new energy buses according to claim 2, characterized in that, Under starting conditions, the measured wheel speed is: v fl v fr v rl v rr These are the wheel speed measurements for the front left wheel, front right wheel, rear left wheel, and rear right wheel, respectively. r is the tire radius, i is the speed ratio of the wheel-side drive system reducer, and n... l n r These represent the rotational speeds of the left and right motors, respectively.

4. The speed control method for new energy buses according to claim 2, characterized in that, Under the initial operating condition, the corrected observation error covariance matrix is: R represents the corrected covariance of observation errors for the front left wheel, front right wheel, rear left wheel, and rear right wheel, respectively. rl，0 R rr，0 These represent the initial values ​​of the observation error covariance for the left and right rear wheels, respectively.

5. The speed control method for new energy buses according to claim 1, characterized in that, Under driving conditions, if |v m，ij (k)-v(k)|>v set If so, it is determined that the wheel is slipping; Among them, v m，ij (k) represents the wheel speed of the four wheels, i represents the left or right wheel, j represents the front or rear axle wheel, and v(k) is the estimated current vehicle speed. set The threshold value for the speed difference.

6. The speed control method for new energy buses according to claim 1, characterized in that, Under braking conditions, if |v m，ij (k)-v(k)|>v set If so, it is determined that the wheel is locked; Among them, v m，ij (k) represents the wheel speed of the four wheels, and v(k) is the estimated current vehicle speed. set The threshold value for the speed difference.

7. The speed control method for new energy buses according to claim 5 or 6, characterized in that, In each calculation process, the confidence factor decreases by no more than 0.

2.

8. A speed control system for a new energy bus, comprising a memory, a processor, and a computer program stored in the memory; characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A new energy bus, characterized in that, It employs the control system described in claim 8.

Images