Model-based transmission gear shifting strategy optimization method and heavy truck

Abstract

The invention discloses a model-based transmission gear shifting strategy optimization method and a heavy truck, and the method comprises the steps: building a future working condition prediction model, building a dynamics prediction model, building an energy consumption prediction model, defining a cost function J, and employing a data search method to obtain an optimal gear shifting strategy; according to the method, the influence of power interruption in the gear shifting process is considered in calculation of the output torque Te (t) of the engine, the gear shifting frequency is effectively reduced, and the fuel economy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of predictive optimization of power chains, and in particular to a model-based transmission shift strategy optimization method and a heavy truck. Background Art

[0002] In the field of automatic manual transmission (AMT) for heavy trucks, gear shift control has a significant impact on the vehicle's power performance, fuel economy, and ride comfort. According to Wu (1988), different gear shift strategies may result in up to 30% difference in fuel consumption under the same driving conditions. This significant variation is mainly due to the nonlinear characteristics of engine fuel consumption, as described by Chen (2021). Therefore, it is crucial to select the appropriate gear to adjust the engine operating conditions to meet the same power demand.

[0003] However, existing shift strategy optimization faces multiple challenges. First, the uncertainty of the vehicle's future torque and power requirements makes the power constraint inaccurate, limiting the improvement of fuel economy. Second, the disengagement and engagement of the clutch during the shift process causes a significant power interruption, which may last for about one second and significantly affects vehicle speed tracking, as described by Li (2011) and Zaimin et al. (2011). Therefore, the complex nature of engine fuel consumption, the unpredictability of future driving conditions, and the interruption of power delivery pose significant challenges to shift optimization.

[0004] At present, shift strategies are mainly divided into MAP-based strategies, data-driven methods, and model-based methods. MAP-based strategies are the most widely used methods. With the development of on-board storage capacity, the number of inputs of shift MAP has grown from 2 to 3, and even more than 4, in order to maximize the adaptability of shift MAP to different conditions. Dynamic programming (DP) and genetic algorithms are often used for offline optimization of shift MAP parameters. However, multi-parameter shift MAPs lead to a huge calibration workload, and offline calibrated MAPs still cannot adapt to diverse and changing actual application conditions.

[0005] Data-driven strategies use the powerful fitting and generalization capabilities of neural networks to solve the above problems. By using optimization algorithms to find the optimal shift sequence as the input of the neural network, the neural network can automatically learn the pattern and reduce the workload during the shift MAP calibration process. The use of neural networks can also make it easier to consider driver operating habits. However, the adaptability of data-driven methods depends on the diversity of collected training data, and online applications require a lot of computing resources, which poses a considerable challenge for practical use.

[0006] Model-based shift strategies use algorithms such as model predictive control (MPC) and dynamic programming (DP) to calculate the overall optimal control strategy based on the acquired future power or road constraints. By predicting future conditions online and solving for the optimal control gear in real time, model-based strategies eliminate the need to calibrate complex shift maps, have no strict requirements for data collection, and can adapt to changing conditions. Summary of the invention

[0007] The purpose of the present invention is to provide a model-based transmission shift strategy optimization method to address the technical defects of frequent gear shifting and poor fuel economy of heavy truck automatic or manual transmission (AMT) in the prior art. The shift strategy specifically considers the influence of power interruption and prediction time domain during the gear shifting process.

[0008] Another object of the present invention is to provide a heavy truck.

[0009] The technical solution adopted to achieve the purpose of the present invention is:

[0010] A method for optimizing a transmission shift strategy based on a model comprises the following steps:

[0011] Step 1: Establish a future operating condition prediction model and calculate t from t 0 to f The accelerator pedal sequence α(t) in the prediction time domain;

[0012] Step 2: Establish a dynamic prediction model, calculate the vehicle speed v(t) at time t under different gear shift paths, and iterate the accelerator pedal sequence α(t) obtained in step 1 into the dynamic prediction model in the prediction time domain to obtain the engine output torque T e (t) and engine speed n e (t);

[0013] Step 3: Establish an energy consumption prediction model and transform the engine output torque T obtained in step 2 into e (t) and engine speed n e (t) continuously iteratively substituting the energy consumption prediction model into the prediction time domain to obtain the vehicle fuel consumption per unit driving distance under different gear shifting paths;

[0014] Step 4, define a cost function J, and calculate the cost function value according to the vehicle speed v(t) predicted in step 2 and the vehicle fuel consumption per unit driving distance predicted in step 3;

[0015] Step 5: Using a data search method, search for a shift path corresponding to the minimum value of the cost function value, which is the optimal shift strategy.

[0016] In the above technical solution, in step 1, the future operating condition prediction model is:

[0017]

[0018] Among them, α(t) is the accelerator pedal opening, α(t 0 ) is t 0 The accelerator pedal opening at the time, the unit is %, k is the decay rate, which is at the initial time t 0 and the end time t f The definition of between is:

[0019]

[0020] Among them, k d1 , k d2 , k d3 are the correction coefficients related to the throttle opening change rate respectively.

[0021] In the above technical solution, in step 2, the kinetic prediction model is:

[0022]

[0023]

[0024] Among them, v(t) is the vehicle speed at time t in the future, dt is the iteration step in the dynamic prediction model, and i g is the transmission ratio corresponding to the selected gear in the prediction time domain, i 0 is the main reduction ratio, m is the mass of the vehicle, g is the acceleration of gravity, θ is the road slope, f is the rolling resistance coefficient, ρ a Air density, A f is the frontal area, C d is the air resistance coefficient, δ is the rotational mass coefficient, l(t) is the unit distance traveled, F d (t) is the wheel-end driving force at time t in the future, T e (t) is the engine output torque, t shift Indicates the last shift time, T shift represents the shift duration, r is the wheel rotation radius, η T is the transmission efficiency, T table is the engine torque lookup function, n e (t) is the engine speed.

[0025] In the above technical solution, in step 3, the energy consumption prediction model is:

[0026]

[0027] Where Q(t) is the vehicle fuel consumption per unit driving distance, in g.

[0028] In the above technical solution, in step 4, the cost function is defined as:

[0029]

[0030] in, is the power performance index, is the economic index, J shift is the gear shift penalty index, for The weight coefficient of for The weight coefficient of For J shift The weight coefficient of .

[0031] In the above technical solution, the power performance index for:

[0032]

[0033] J Dyn =-v t,dyn

[0034] Among them, J Dyn,min represents the minimum value of the dynamic performance index in all gear shift paths, and t is t 0 ~t f,dyn The dynamic time domain, v t,dyn Represents the terminal vehicle speed of the dynamic time domain, and the length of the dynamic time domain is equal to the prediction time domain length of the prediction model.

[0035] In the above technical solution, the economic index for:

[0036]

[0037] Among them, t 0 ~t f,eco represents the economic time domain, the length of the economic time domain is shorter than the length of the dynamic time domain, J Eco,min Represents the minimum value of the economic index among all gear shift paths.

[0038] In the above technical solution, the gear shift penalty index J Shift for:

[0039]

[0040] Among them, Gear t is the gear position corresponding to time t in the prediction time domain.

[0041] Another aspect of the present invention also includes a heavy truck, including a control unit and an automatic manual transmission communicating with the control unit, wherein the control unit is used to execute the shift strategy.

[0042] In the above technical solution, the control unit performs the following operations:

[0043] Predict the future accelerator pedal sequence, and based on the predicted future accelerator pedal sequence, use the energy consumption prediction model and the dynamic prediction model to calculate the vehicle speed and fuel consumption performance per unit driving distance under different shifting paths, determine the shifting strategy that minimizes the cost function according to the calculated performance, and output the shifting strategy to control the automatic or manual transmission to shift gears.

[0044] In the above technical solution, the control unit includes a processor for executing iterative calculations of the prediction model, selecting a shift strategy that minimizes a cost function, and generating a shift control signal.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The engine output torque T of the present invention e The calculation of (t) takes into account the impact of power interruption during the gear shifting process, effectively reducing the gear shifting frequency and improving fuel economy.

[0047] 2. The present invention utilizes the dynamic time domain and the economic time domain to achieve flexible adjustment of the balance between vehicle speed and fuel consumption.

[0048] 3. The present invention has been verified through simulation, which proves the effectiveness and superiority of the proposed shifting strategy under simulated actual driving conditions, indicating that by considering the impact of power interruption, the shifting frequency can be reduced by 12%.

[0049] 4. The shift strategy of the present invention reduces the reliance on complex shift MAP calibration, reduces the calibration workload, and improves the adaptability to variable driving conditions.

[0050] 5. Under C-WTVC conditions, the proposed shift strategy can reduce fuel consumption by 2.78% and reduce the shift frequency by 13.78% compared with the MAP-based strategy. These improvements verify the effectiveness of the proposed algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a general architecture diagram of Example 1 of the present invention. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0053] Example 1

[0054] A model-based shift strategy for a heavy-duty truck automated manual transmission includes the following steps:

[0055] like Figure 1 As shown, a model-based shift strategy for a heavy-duty truck automatic or manual transmission includes the following steps:

[0056] Step 1: Establish a future operating condition prediction model and calculate t from t 0 to f The accelerator pedal sequence α(t) in the prediction time domain, the future working condition prediction model is:

[0057]

[0058] Among them, α(t) is the accelerator pedal opening, the unit is %, α(t 0 ) is t 0 The accelerator pedal opening at the time, the unit is %, k is the decay rate, which is at the initial time t 0 and the end time t f The definition of between is:

[0059]

[0060] Among them, k d1 , k d2 , k d3 are the correction coefficients related to the throttle opening change rate respectively.

[0061] Step 2: Establish a dynamic prediction model, calculate the vehicle speed v(t) at time t under different shift paths, and iterate the accelerator pedal sequence α(t) obtained in step 1 into the dynamic prediction model in the prediction time domain to obtain the engine output torque T e (t) and engine speed n e (t), the kinetic prediction model is:

[0062]

[0063] Among them, v(t) is the vehicle speed at time t in the future, dt is the iteration step in the dynamic prediction model, and i g is the transmission ratio corresponding to the selected gear in the prediction time domain, i 0 is the main reduction ratio, m is the mass of the vehicle, g is the acceleration of gravity, θ is the road slope, f is the rolling resistance coefficient, ρ a Air density, Af is the frontal area, C d is the air resistance coefficient, δ is the rotational mass coefficient, l(t) is the unit distance traveled, F d (t) is the wheel-end driving force at time t in the future, T e (t) is the engine output torque, t shift Indicates the last gear shifting moment, T shift represents the shift duration, r is the wheel rotation radius, η T is the transmission efficiency, T table is the engine torque lookup function, n e (t) is the engine speed.

[0064] Step 3: Establish an energy consumption prediction model and transform the engine output torque T obtained in step 2 into e (t) and engine speed n e (t) is substituted into the energy consumption prediction model to obtain the vehicle fuel consumption per unit driving distance under different gear shift paths. The energy consumption prediction model is:

[0065]

[0066] Where Q(t) is the vehicle fuel consumption per unit driving distance, in g.

[0067] Step 4, define a cost function J, and calculate the cost function value according to the vehicle speed v(t) predicted in step 2 and the vehicle fuel consumption per unit driving distance predicted in step 3. The cost function is defined as:

[0068]

[0069] J Dyn =-v t,dyn

[0070] in, is the power performance index, is the economic index, J Shift is the gear shift penalty index, for The weight coefficient of for The weight coefficient of For J shift The weight coefficient of t is t 0 ~t f,dym The dynamic time domain, v t,dyn Represents the terminal vehicle speed of the dynamic time domain, the length of the dynamic time domain is equal to the prediction time domain length of the prediction model;

[0071]

[0072] Among them, t 0 ~t f,eco Represents the economic time domain, the length of the economic time domain is shorter than that of the dynamic time domain;

[0073]

[0074] Among them, J Dyn,min Represents the minimum value of the dynamic performance index in all gear shift paths;

[0075]

[0076] Among them, J Eco,min Represents the minimum value of the economic index among all gear shift paths;

[0077]

[0078] Among them, Gear t is the gear position corresponding to time t in the prediction time domain.

[0079] Adjust J according to the optimization goal Dym , and J shift The weight coefficients of the three are as follows: the optimization goal is to reduce the vehicle fuel consumption while ensuring the vehicle speed, while reducing the shifting frequency to achieve a compromise among multiple objectives such as power, economy, and comfort.

[0080] Step 5: Using a data search method, search for a gear shift path corresponding to the minimum value of the cost function value, which is the optimal gear shift strategy. The process of determining the optimal gear shift strategy involves evaluating each possible gear shift path.

[0081] In this embodiment, the shifting path is defined, and it is stipulated that in the prediction time domain, only the first two steps can be used for the shifting operation, and each shifting path allows at most one shifting operation, and the range of the shifting operation is from upshifting 2 gears to downshifting 2 gears. Therefore, there are 4 paths for shifting at the beginning of the first step, 4 paths for shifting at the beginning of the second step, and 1 path for not shifting, for a total of 9 paths.

[0082] Example 2

[0083] On the basis of embodiment 1, in order to further optimize the shifting strategy, the following steps are also included:

[0084] Step 1, under C-WTVC (China World Transient Vehicle Cycle) conditions, the optimal shift strategy is verified using a high-precision simulation platform, and the performance of the proposed optimal shift strategy is compared with the existing MAP-based shift strategy in terms of fuel consumption and shift frequency.

[0085] Step 2: Based on the simulation verification results, make necessary adjustments to the shift strategy to ensure optimal performance under different driving conditions and vehicle configurations, and apply the optimized shift strategy to the actual vehicle control system.

[0086] Example 3

[0087] A heavy truck includes a control unit and an automatic manual transmission communicating with the control unit, wherein the control unit is used to execute the shift strategy described in Embodiment 1 or 2, and the control unit performs the following operations:

[0088] Predict the future accelerator pedal sequence, and based on the predicted future accelerator pedal sequence, use the energy consumption prediction model and the dynamic prediction model to calculate the vehicle speed and fuel consumption performance per unit driving distance under different shifting paths, determine the shifting strategy that minimizes the cost function according to the calculated performance, and output the shifting strategy to control the automatic or manual transmission to shift gears.

[0089] Furthermore, the control unit includes a processor for executing iterative calculations of the prediction model, selecting a shift strategy that minimizes a cost function, and generating a shift control signal.

[0090] The above is only a preferred embodiment of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A transmission shift strategy optimization method based on a model, characterized in that: The following steps are involved: Step 1: Establish a future operating condition prediction model and calculate t from t0 to t f The accelerator pedal sequence α(t) in the prediction time domain; Step 2: Establish a dynamic prediction model, calculate the vehicle speed v(t) at time t under different gear shift paths, and iterate the accelerator pedal sequence α(t) obtained in step 1 into the dynamic prediction model in the prediction time domain to obtain the engine output torque T e (t) and engine speed n e (t); Step 3: Establish an energy consumption prediction model and transform the engine output torque T obtained in step 2 into e (t) and engine speed n e (t) continuously iteratively substituting the energy consumption prediction model into the prediction time domain to obtain the vehicle fuel consumption per unit driving distance under different gear shifting paths; Step 4, define a cost function J, and calculate the cost function value according to the vehicle speed v(t) predicted in step 2 and the vehicle fuel consumption per unit driving distance predicted in step 3; Step 5: Using a data search method, search for a shift path corresponding to the minimum value of the cost function value, which is the optimal shift strategy.

2. The optimization method according to claim 1, characterized in that: In step 1, the future operating condition prediction model is: Among them, α(t) is the accelerator pedal opening, α(t0) is the accelerator pedal opening at time t0, the unit is %, and k is the attenuation rate, which is the same between the initial time t0 and the end time t f The definition of between is: Among them, k d1 , k d2 , k d3 are the correction coefficients related to the throttle opening change rate respectively.

3. The optimization method according to claim 1, characterized in that: In step 2, the kinetic prediction model is: Among them, v(t) is the vehicle speed at time t in the future, dt is the iteration step in the dynamic prediction model, and i g is the transmission ratio corresponding to the gear selected in the prediction time domain, i0 is the main reduction ratio, m is the mass of the car, g is the acceleration of gravity, θ is the road slope, f is the rolling resistance coefficient, ρ is the air density, A f is the frontal area, C d is the air resistance coefficient, δ is the rotational mass coefficient, l(t) is the unit distance traveled, F d (t) is the wheel end driving force at time t in the future, T e (t) is the engine output torque, t shift Indicates the last gear shifting moment, T shift represents the shift duration, r is the wheel rotation radius, η T is the transmission efficiency, T table is the engine torque lookup function, n e (t) is the engine speed.

4. The optimization method according to claim 1, characterized in that: In step 3, the energy consumption prediction model is: Where Q(t) is the vehicle fuel consumption per unit driving distance, in g.

5. The optimization method according to claim 1, characterized in that: In step 4, the cost function is defined as: in, is the power performance index, is the economic index, J Shift is the gear shift penalty index, for The weight coefficient of for The weight coefficient of For J shift The weight coefficient of .

6. The optimization method according to claim 5, characterized in that: The power performance index for: J Dyn =-in t,dyn Among them, J Dyn,min Represents the minimum value of the dynamic performance index in all shift paths, t is t0~t f,dyn The dynamic time domain, v t,dyn Represents the terminal vehicle speed of the dynamic time domain, and the length of the dynamic time domain is equal to the prediction time domain length of the prediction model.

7. The optimization method according to claim 6, characterized in that: The economic indicators for: Among them, t0~t f,eco represents the economic time domain, the length of the economic time domain is shorter than the length of the dynamic time domain, J Eco,min Represents the minimum value of the economic index among all gear shift paths.

8. The optimization method according to claim 7, characterized in that: The gear shift penalty index J Shift for: Among them, Gear t is the gear position corresponding to time t in the prediction time domain.

9. A heavy truck, characterized in that: The invention comprises a control unit and an automatic manual transmission communicating with the control unit, wherein the control unit is used to execute the optimal shifting strategy, and the control unit performs the following operations: Predict the future accelerator pedal sequence, and based on the predicted future accelerator pedal sequence, use the energy consumption prediction model and the dynamic prediction model to calculate the vehicle speed and fuel consumption performance per unit driving distance under different shifting paths, determine the shifting strategy that minimizes the cost function according to the calculated performance, and output the shifting strategy to control the automatic or manual transmission to shift gears.

10. The heavy truck according to claim 9, characterized in that: The control unit includes a processor for performing iterative calculations of the prediction model, selecting a shift strategy that minimizes a cost function, and generating a shift control signal.

Images