Turbocharger control method and device based on driving habit prediction, vehicle and medium

Abstract

The invention discloses a turbocharger control method and device based on driving habits, a vehicle and a medium, and relates to the technical field of vehicle control. The method comprises the steps that accelerator operation parameters are collected, and a personalized coefficient and a gas path coupling coefficient are obtained based on the historical driving characteristics of a driver and the operation working condition of an engine; the accelerator operation parameters are input into the short-term driving behavior learning model and the long-term driving behavior learning model, the first prediction result and the second prediction result are dynamically fused, and the predicted accelerator opening degree at the future moment is determined; according to the predicted accelerator opening degree, the individuation coefficient, the gas path coupling coefficient and the engine operation parameters, the time compensation amount and the opening degree compensation amount of the waste gas valve are determined; the wastegate valve is controlled on the basis of the time compensation amount and the opening compensation amount. By the adoption of the method, self-adaptive adjustment of the turbocharger can be achieved according to operation habits of different drivers, supercharging pressure is established in advance, turbine lag is reduced, and the response speed is increased.

Description

Technical Field

[0001] This invention relates to the field of technology, and in particular to a turbocharger control method, device, vehicle, and medium based on driving habit prediction. Background Technology

[0002] As a crucial component for improving engine performance, the turbocharger primarily increases engine power output and density by compressing intake air. This increased intake air volume allows for more thorough mixing of fuel and air, optimizing combustion efficiency and contributing to reduced fuel consumption and emissions. Currently, the mainstream types of turbochargers for automotive engines include exhaust gas turbochargers, mechanical superchargers, and electric superchargers. Among these, exhaust gas turbochargers are widely used in various vehicle models due to their mature structure and high efficiency. However, exhaust gas turbochargers suffer from a lag in power response at low speeds or during rapid acceleration, a condition known as turbo lag. This is mainly because the amount of exhaust gas is insufficient to drive the turbine, resulting in a time delay in exhaust energy transfer. Factors such as turbine rotor inertia, intake manifold length, and turbocharging system design also increase the pressure build-up time. Turbo lag causes the engine to respond less promptly to sudden changes in power demand, thus affecting the driving experience.

[0003] To address turbo lag, related technologies attempt to optimize both the mechanical structure and the control system. Using a low-inertia turbine can reduce rotor inertia and improve response speed, but due to material limitations, boost pressure may still decrease at high speeds. Technologies such as variable cross-sectional area turbines or electric superchargers can mitigate turbo lag to some extent, but require more precise sensors and more complex control strategies, increasing system cost and control complexity.

[0004] In terms of control systems, turbochargers typically employ closed-loop turbocharging control strategies, such as... Figure 1 As shown, the boost pressure is adjusted based on the deviation between the target boost pressure and the actual boost pressure, while the boost operation is corrected by referring to the engine torque deviation. This control process forms a closed loop with real-time monitoring and overpressure protection, enabling on-demand adjustment and abnormal protection to ensure the safe operation of the engine and turbocharger under various operating conditions.

[0005] Although the power response of turbocharged systems has improved, issues such as power lag or unstable boost pressure build-up may still occur during rapid acceleration, frequent gear changes, or different driving conditions, affecting engine power and driving experience. Furthermore, turbocharger control parameters are typically calibrated uniformly for each vehicle model, making it difficult to meet the driving needs of different drivers. Summary of the Invention

[0006] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, the object of this invention is to propose a turbocharger control method, device, vehicle, and medium based on driving habit prediction, to improve the dynamic response speed of the turbocharger, reduce the impact of turbo lag on power output, and achieve adaptive control for different driving styles.

[0007] To achieve the above objectives, a first aspect of the present invention proposes a turbocharger control method based on driving habit prediction, comprising: Collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics and engine operating conditions; The throttle operating parameters are input into the short-term driving behavior learning model and the long-term driving behavior learning model respectively, and the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model are dynamically fused to determine the predicted throttle opening at future times. Based on the predicted throttle opening, the personalized coefficient, the air circuit coupling coefficient, and the engine operating parameters, determine the time compensation amount and opening compensation amount of the exhaust valve; The exhaust valve is controlled based on the time compensation amount and the opening compensation amount.

[0008] In addition, the turbocharger control method based on driving habit prediction in the above embodiments of the present invention may also have the following additional technical features: According to one embodiment of the present invention, the throttle operating parameters include throttle opening and throttle change rate.

[0009] According to an embodiment of the present invention, the dynamic fusion of the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model includes: The weighting coefficient of the first prediction result is determined based on the throttle change rate; Based on the weighting coefficients, the first prediction result and the second prediction result are weighted and fused.

[0010] According to an embodiment of the present invention, determining the time compensation amount and opening compensation amount of the exhaust valve based on the predicted throttle opening, the individualization coefficient, the air circuit coupling coefficient, and engine operating parameters includes: The time compensation amount is determined based on the personalization coefficient, the inherent delay parameter of the turbocharger, the sensitivity adjustment coefficient, and the throttle change rate. The opening compensation amount is determined based on the air path coupling coefficient, the predicted throttle opening, and the engine speed normalization factor.

[0011] According to one embodiment of the present invention, the method includes: Monitor the actual throttle opening; If the deviation between the actual throttle opening and the predicted throttle opening continues to exceed a preset threshold, then the exhaust valve is pressurized based on the actual throttle opening.

[0012] According to one embodiment of the present invention, the method further includes: If the deviation between the actual throttle opening and the predicted throttle opening is lower than a preset threshold, then the short-term driving behavior learning model and the long-term driving behavior learning model are updated based on the actual throttle opening.

[0013] According to one embodiment of the present invention, the time compensation amount is provided with a maximum time compensation threshold to limit the time compensation amount from not exceeding the maximum time compensation threshold.

[0014] To achieve the above objectives, a second aspect of the present invention provides a turbocharger control device based on driving habit prediction, comprising: The data processing module is used to collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics and engine operating conditions. The model prediction module is used to input the throttle operation parameters into a short-term driving behavior learning model and a long-term driving behavior learning model respectively, and dynamically fuse the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model to determine the predicted throttle opening at future times. The compensation amount determination module is used to determine the time compensation amount and opening compensation amount of the exhaust valve based on the predicted throttle opening, the personalized coefficient, the air circuit coupling coefficient and the engine operating parameters. The control module is used to control the exhaust valve based on the time compensation amount and the opening compensation amount.

[0015] To achieve the above objectives, a third aspect of the present invention provides a vehicle including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the turbocharger control method based on driving habit prediction described above.

[0016] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the turbocharger control method based on driving habit prediction described above.

[0017] The turbocharger control method, device, vehicle, and medium based on driving habit prediction in this invention introduce a predictive control mechanism based on driving habits. This mechanism can predict the driver's power demand before throttle operation and dynamically adjust the exhaust valve by combining a personalization coefficient and an air-circuit coupling coefficient, thereby effectively reducing turbo lag. This method not only improves the turbocharger's pressure build-up speed and stability under different driving conditions but also makes the engine's power output smoother during acceleration. Furthermore, by fusing short-term and long-term driving behavior prediction models, adaptive learning and updating of driving behavior can be achieved, making boost control more aligned with individual driver habits and further improving the vehicle's acceleration response and driving experience. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the closed-loop boost control process in related technologies; Figure 2 This is a flowchart illustrating a turbocharger control method based on driving habits in one embodiment; Figure 3 This is a schematic diagram illustrating the specific process of dynamic fusion in one embodiment; Figure 4 Here is a system timing interaction diagram in one embodiment; Figure 5 This is a control flowchart of a turbocharger control method based on driving habits in one embodiment; Figure 6 This is a schematic diagram illustrating the operation of a turbocharger control method based on driving habits under rapid acceleration conditions in one application embodiment. Figure 7 This is a structural block diagram of a turbocharger control device based on driving habits in one embodiment. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with 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.

[0020] The implementation details of the technical solutions in the embodiments of this application are described in detail below.

[0021] In one embodiment, such as Figure 2 The diagram shows a flowchart of a turbocharger control method based on driving habits, which may include the following steps: Step S101: Collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics.

[0022] Throttle operating parameters are used to reflect the driver's current acceleration intention and can be collected in real time by the throttle pedal sensor.

[0023] Stored historical driving characteristic data can reflect a driver's long-term driving habits, typically including indicators such as frequency of rapid acceleration and cruise stability. Statistical analysis of this historical driving characteristic data can extract typical throttle opening patterns under different vehicle speeds, loads, and operating conditions, thereby identifying the driver's driving style. One feasible approach is to categorize driving styles into normal driving modes and aggressive driving modes. Normal driving mode corresponds to smooth acceleration and high cruise stability; aggressive driving mode corresponds to frequent rapid acceleration and a strong demand for power response.

[0024] A personalization coefficient is calculated based on the driver's driving style, which describes the driver's acceleration sensitivity and power response preferences. Furthermore, by analyzing the dynamic relationship between the driver's throttle changes under different acceleration intensities and engine operating data, the coupling characteristics of driving behavior on the airflow pressure build-up process can be determined, thereby obtaining the airflow coupling coefficient to characterize the impact of engine airflow dynamics on boost response.

[0025] Through the above process, while maintaining real-time data collection, a parameter base that takes into account individual driver differences and dynamic characteristics of the air circuit can be established, providing a basis for exhaust valve compensation control.

[0026] In one embodiment, throttle operating parameters include throttle opening and throttle change rate. Throttle opening characterizes the driver's acceleration demand at the current moment, and its value directly reflects the engine load target. Throttle change rate characterizes the rate of change of throttle opening over time, reflecting the intensity of the driver's operation and the abrupt changes in acceleration intention. By simultaneously acquiring throttle opening and throttle change rate, the dynamic characteristics of driving behavior can be captured in the time dimension, enabling subsequent driving habit recognition and prediction models to analyze not only based on the current operating state.

[0027] Step S102: Input the throttle operation parameters into the short-term driving behavior learning model and the long-term driving behavior learning model respectively, and dynamically fuse the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model to determine the predicted throttle opening at future times.

[0028] The collected throttle operation parameters are input into both short-term and long-term driving behavior learning models to predict throttle opening based on driving behavior characteristics at different time scales. The short-term driving behavior learning model is constructed using a Long Short-Term Memory (LSTM) network algorithm, focusing on reflecting the driver's immediate throttle operation characteristics under current conditions. It can perform deep learning analysis on the time-series features of throttle operation parameters. This model has high sensitivity in capturing the driver's operational trends and sudden acceleration intentions within a short period, and can promptly reflect the real-time power demand of the driving element, thereby achieving high-precision prediction of dynamic throttle changes.

[0029] The long-term driving behavior learning model employs the Autoregressive Integrated Moving Average (ARIMA) algorithm. By modeling the statistical characteristics of historical driving data, it identifies fixed operating patterns of drivers in scenarios such as commuting routes and highway cruising, thereby obtaining the driver's throttle control patterns under different vehicle speeds, loads, and road conditions. The long-term driving behavior learning model can characterize the driver's long-established operating habits and reflect the driver's throttle control trends under steady-state driving conditions.

[0030] The first prediction output by the short-term driving behavior learning model characterizes the driver's immediate throttle change trend, while the second prediction output by the long-term driving behavior learning model reflects the inertial characteristics of the driver's long-term behavior. By dynamically fusing the first and second predictions, the future... The predicted throttle opening at any given moment. This prediction takes into account both immediate response and long-term trends, ensuring both rapid throttle opening prediction and stable overall response. The resulting predicted throttle opening for future moments more accurately reflects the driver's actual acceleration intention.

[0031] In one embodiment, Figure 3 The diagram illustrates the specific process of dynamic fusion, which may include the following steps: Step S201: Determine the weighting coefficient of the first prediction result based on the throttle change rate.

[0032] Throttle change rate reflects the intensity of the driver's acceleration intention and the intensity of the operation at the current moment. When the throttle change rate is large, it indicates that the driver has a clear acceleration demand, and the proportion of short-term prediction results in the overall prediction needs to be increased to ensure the speed of control response. When the throttle change rate is small, it indicates that the driver's operation is relatively stable. At this time, the weight of long-term prediction results should be increased to maintain the stability of the turbocharger system control.

[0033] Based on this, the weight coefficients of the first prediction result output by the short-term driving behavior learning model are adjusted by the throttle change rate, so that the weights can adaptively adjust with changes in driving behavior. Specifically, the relationship between the throttle change rate and the weight coefficients of the first prediction result can be expressed as follows:

[0034] In the above formula, This represents the weighting coefficient of the first prediction result. Represents the personalization coefficient. This represents the rate of change of throttle. Through a nonlinear mapping relationship, short-term predictions can be given higher weight during aggressive driving and lower weight during mild driving, thereby improving the adaptability of the fusion model to differences in driving behavior.

[0035] Step S202: Based on the weighting coefficients, the first prediction result and the second prediction result are weighted and fused.

[0036] In the weighted fusion process, using determined weight coefficients as adjustment factors, the first and second prediction results are weighted, achieving a balance between real-time response speed and prediction stability. The predicted throttle opening can be expressed as:

[0037] In the above formula, This indicates the predicted throttle opening. This represents the first prediction result output by the short-term driving behavior learning model, whose input sequence is the current throttle operation parameters; This represents the second prediction result output by the long-term driving behavior learning model. Its input sequence is the current throttle operation parameter state, and it is generated based on the statistical characteristics of the model's long-term historical throttle operation data. It is used to characterize the steady-state operation trend of the driver under long-term driving conditions.

[0038] By employing weighted fusion, a balance can be struck between response speed and predictive stability. When driving behavior changes abruptly, the fusion result can quickly respond to short-term predicted trends; when driving behavior is stable, the fusion result relies more on the long-term predicted trends. The resulting fusion prediction can more accurately reflect future throttle opening changes, providing highly timely and reliable input for subsequent exhaust valve advance control.

[0039] Step S103: Determine the time compensation amount and opening compensation amount of the exhaust valve based on the predicted throttle opening, individual coefficient, air circuit coupling coefficient and engine operating parameters.

[0040] Predicting throttle opening characterizes the driver's power demand trend in the near future, reflecting their throttle response tendency before the driver actually operates the throttle. A personalization coefficient can be used to reflect the driver's throttle response preferences. In compensation calculations, combining predicted throttle opening and the personalization coefficient allows for adaptive adjustment of the compensation amount according to different drivers' driving styles, ensuring the compensation amount meets the driver's driving needs.

[0041] Meanwhile, the airflow coupling coefficient reflects the coupling relationship between engine intake pressure, exhaust pressure, and dynamic changes in throttle position, and can be used to describe the airflow transmission delay characteristics of the turbocharger system under current operating conditions. A larger airflow coupling coefficient indicates a more significant response delay of the turbocharger system to changes in throttle opening. Therefore, in compensation calculations, by considering the airflow coupling coefficient and engine operating parameters, the compensation amount can be further corrected according to actual operating conditions.

[0042] Based on this, the predicted throttle opening, individual coefficient, air circuit coupling coefficient and engine operating parameters are used as inputs to estimate the adjustment requirements of the exhaust gas valve. The time compensation amount and opening compensation amount of the exhaust gas valve are calculated respectively to perform feedforward compensation for the opening adjustment of the exhaust gas valve.

[0043] The time compensation amount is used to adjust the timing of the exhaust valve's operation in advance, so that the turbocharger can build up boost pressure before the driver actually accelerates, thus offsetting the delay caused by the lag in boost pressure build-up; the opening compensation amount is used to correct the amplitude of the exhaust valve opening, so that the boost pressure build-up speed matches the predicted power demand, thereby improving rapid acceleration response and reducing turbo lag.

[0044] In one embodiment, the specific process for determining the time compensation amount and the opening compensation amount is explained.

[0045] The time compensation amount is determined based on the personalization coefficient, the inherent delay parameter of the turbocharger, the sensitivity adjustment coefficient, and the throttle change rate. Specifically, the time compensation amount can be expressed as:

[0046] In the above formula, Indicates the amount of time compensation; This is a personalization coefficient; This is the inherent delay parameter of the turbine, used to characterize the physical delay required for the turbine to build up boost at the current speed; This is the sensitivity adjustment coefficient, reflecting the throttle response speed.

[0047] The throttle opening compensation amount is determined based on the air-path coupling coefficient, the predicted throttle opening, and the engine speed normalization factor. Specifically, the throttle opening compensation amount can be expressed as:

[0048] In the above formula, Indicates the opening compensation amount; Indicates the gas path coupling coefficient; This represents the engine speed normalization factor, used to standardize the response characteristics at different engine speeds.

[0049] In one embodiment, the setting of the time compensation amount is further limited. Since the time compensation amount is used to adjust the timing of the exhaust valve's operation, if the compensation amount is too large, it may cause excessive premature release of boost pressure, resulting in power response fluctuations or unnecessary exhaust valve operation. Therefore, by setting a maximum time compensation threshold, the calculated time compensation amount is constrained to ensure that the time compensation amount is within a reasonable range. Based on this, the time compensation amount can be expressed as:

[0050] in, The maximum time compensation threshold can be determined based on engine type, turbocharger response characteristics, etc. When determining the actual time compensation amount, if the calculated value... Greater than the maximum time compensation threshold Then the time compensation amount Restricted to If the calculated Less than or equal to If so, then the original calculation result will remain.

[0051] Step S104: Control the exhaust valve based on the time compensation amount and the opening compensation amount.

[0052] After obtaining the time compensation and opening compensation of the exhaust gas valve, the compensation can be applied to adjust the actual opening of the exhaust gas valve, allowing the boost pressure to build up before the predicted power demand. Specifically, when performing exhaust gas valve adjustment, the timing of the exhaust gas valve's action is advanced based on the time compensation, causing the exhaust gas valve to initiate the corresponding adjustment action before the actual throttle signal arrives. Subsequently, the target opening of the exhaust gas valve is determined based on the opening compensation, causing the exhaust gas valve to advance... Time begins The angle guides the engine exhaust gas into the turbine so that the exhaust gas flow matches the predicted boost demand.

[0053] In one feasible approach, the time compensation amount With opening compensation amount The signal is converted into a corresponding pulse width modulation (PWM) control command and output to the actuator of the exhaust valve. Upon receiving the PWM control command, the exhaust valve... Adjust the lead time of the action and follow the specified lead time. Adjusting the opening allows the exhaust gas to flow to the turbine impeller area as expected, thereby increasing the turbine speed and reducing the delay in boost pressure build-up.

[0054] By using the above method, the exhaust valve opening can be reduced in advance, so that the turbocharger is activated when the driver actually presses the accelerator. The turbocharger reaches the target RPM before the target time and establishes pre-boost. In this case, the turbocharger's response delay is completely eliminated, and the driver can obtain sufficient power output the moment he presses the accelerator, improving the power lag phenomenon during acceleration.

[0055] In one embodiment, to ensure the stability and safety of predictive boost control, the actual throttle opening can be monitored in real time during the exhaust valve compensation adjustment process. If the deviation between the actual throttle opening and the predicted throttle opening continuously exceeds a preset threshold for multiple consecutive sampling periods, it can be determined that the prediction result has significantly deviated from the driver's actual operation. At this time, to avoid control instability caused by prediction errors, the feedforward compensation adjustment based on the predicted throttle opening can be stopped, and the process can be re-executed based on the actual throttle opening. Figure 1 The traditional boost control method shown.

[0056] For example, when the above deviation continues to exceed a preset threshold (e.g., 10%), the opening of the exhaust valve is adjusted by PID control so that the boost pressure is adjusted according to the actual throttle demand, which can ensure that the turbocharger still has a stable and reliable pressure response capability under abnormal conditions.

[0057] In one embodiment, when the deviation between the actual throttle opening and the predicted throttle opening is lower than a preset threshold, the currently collected actual throttle opening can be used to update the short-term and long-term driving behavior learning models. Specifically, the latest throttle operation data can be used as incremental samples to input into the short-term and long-term driving behavior learning models to update the model parameters online, thereby enabling the models to improve the prediction accuracy of throttle opening and maintain the long-term performance stability of predictive control.

[0058] In practical applications, the turbocharger control method based on driving habits in this embodiment can achieve predictive control of the turbocharger by working together through sensor data acquisition, driving behavior analysis, throttle prediction, compensation calculation and exhaust valve adjustment. Figure 4 The system timing interaction diagram illustrates the data flow and time sequence between the sensors, electronic control unit (ECU), driving behavior learning module, LSTM prediction module, ARMA prediction module, dynamic fusion unit, compensation controller, and exhaust valve actuator. Figure 5This is a control flowchart for a turbocharger control method based on driving habits. It describes the specific process from system initialization, driving mode determination, throttle prediction and strategy generation, control scheme comparison, compensation execution and online evaluation, to strategy switching when there are abnormalities or prediction errors exceed the limit.

[0059] While the vehicle is running, sensors continuously collect throttle operation parameters and transmit the data to the ECU. The ECU sends a query request to the driving behavior learning module to retrieve the driver's historical driving characteristics. Based on these historical driving characteristics, the driver's driving mode can be determined, thereby outputting a personalization coefficient and an air-circuit coupling coefficient.

[0060] The ECU further inputs throttle operating parameters into the LSTM and ARMA predictors simultaneously, which run in parallel, outputting a first prediction result and a second prediction result respectively. The dynamic fusion unit calculates the comprehensive dynamic weight of the LSTM based on the current throttle change rate, and weights and fuses the first prediction result and the second prediction result to generate the predicted throttle opening at future time.

[0061] The compensation controller further calculates the time compensation and opening compensation by predicting throttle opening, individual coefficient, air-circuit coupling coefficient, and engine operating parameters. The ECU converts the calculated time compensation and opening compensation into PWM control signals and sends them to the exhaust gas valve, enabling the exhaust gas valve to adjust according to the advance timing and target opening, thereby improving turbo lag and increasing response speed.

[0062] During vehicle operation, the system continuously monitors the actual throttle opening and compares the real-time throttle opening with the predicted throttle opening to evaluate the model's predictive performance. When the deviation between the actual throttle opening and the predicted value continuously exceeds a preset threshold (e.g., 10%), it can switch to traditional PID boost control to ensure the stability of the boost response; while when the deviation is below the preset threshold, real-time throttle data can be used to incrementally update the short-term and long-term predictive models.

[0063] In one application embodiment Figure 6 This is a schematic diagram illustrating the operation of a turbocharger control method based on driving habits under rapid acceleration conditions. The following is combined with... Figure 6 This paper explains the application of a turbocharger control method based on driving habits under specific operating conditions. Taking a rapid acceleration scenario as an example, assuming the vehicle is currently cruising at a constant speed of 80 km / h, the engine speed is 2500 rpm, and the engine's rated speed is 6000 rpm, when the vehicle encounters a slower vehicle ahead and needs to accelerate rapidly to overtake, the sensor detects the driver's rapid acceleration signal, at which point the throttle change rate is 580 deg / s.

[0064] The ECU queries the driving behavior history database and finds that the driver's frequency of rapid acceleration exceeded 25% in nearly 1,000 kilometers of driving data, which is matched as an aggressive driving mode, thus generating a personalized coefficient. and gas path coupling coefficient Subsequently, the real-time collected throttle operation parameters are input into the Short-Term Driving Behavior Learning Model (LSTM) and the Long-Term Driving Behavior Learning Model (ARMA). The dynamic fusion module predicts that the throttle opening will reach 85% in the next 320ms.

[0065] Turbine inherent delay parameters obtained from a table and throttle sensitivity coefficient By combining the personalization coefficient and the throttle change rate, the time compensation amount of the exhaust valve is calculated. :

[0066] Opening compensation amount :

[0067] The ECU converts the above calculation results into a PWM signal and sends a control command to the exhaust valve, causing the exhaust valve to operate in advance. Turn it on and adjust the opening angle to... This directs more exhaust gas into the turbine. Thus, when the driver actually presses the accelerator to 85%, the turbine has already reached the target speed through predictive feedforward control, resulting in a measured reduction of acceleration delay by approximately 95%, significantly improving rapid acceleration response performance.

[0068] In the above embodiments, the turbocharger control method based on driving habits of the present invention can comprehensively consider the driver's historical driving characteristics, current throttle operation trends, and engine operating conditions to achieve predictive feedforward adjustment of the exhaust gas valve. This method adaptively adjusts the time compensation and opening compensation amounts through personalized coefficients and air-path coupling coefficients, enabling the turbocharger to pre-establish boost pressure before the driver actually operates the throttle, significantly reducing turbo lag and improving the engine's power response speed. Simultaneously, this method can adapt to different driving styles and operating conditions, optimizing rapid acceleration response while maintaining boost efficiency under smooth driving conditions, demonstrating good applicability.

[0069] In one embodiment, a turbocharger control device based on driving habits is provided, with reference to... Figure 7 As shown, the turbocharger control device 300 based on driving habits may include: a data processing module 301, a model prediction module 302, a compensation amount determination module 303, and a control module 304. Among them, The data processing module 301 is used to collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics and engine operating conditions. The model prediction module 302 is used to input the throttle operation parameters into the short-term driving behavior learning model and the long-term driving behavior learning model respectively, and dynamically fuse the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model to determine the predicted throttle opening at future moments. The compensation amount determination module 303 is used to determine the time compensation amount and opening compensation amount of the exhaust valve based on the predicted throttle opening, the individual coefficient, the air circuit coupling coefficient and the engine operating parameters. The control module 304 is used to control the exhaust valve based on the time compensation amount and the opening compensation amount.

[0070] In one embodiment, the throttle operating parameters include throttle opening and throttle change rate.

[0071] In one embodiment, the model prediction module 302 is specifically used for: The weighting coefficient of the first prediction result is determined based on the throttle change rate; Based on the weighting coefficients, the first prediction result and the second prediction result are weighted and fused.

[0072] In one embodiment, the compensation amount determination module 303 is specifically used for: The time compensation amount is determined based on the individualization coefficient, the inherent delay parameter of the turbocharger, the sensitivity adjustment coefficient, and the throttle change rate. The opening compensation amount is determined based on the air path coupling coefficient, the predicted throttle opening, and the engine speed normalization factor.

[0073] In one embodiment, the control module 304 is used for: Monitor the actual throttle opening; If the deviation between the actual throttle opening and the predicted throttle opening continues to exceed a preset threshold, the exhaust valve will be pressurized based on the actual throttle opening.

[0074] In one embodiment, the control module 304 is further configured to: If the deviation between the actual throttle opening and the predicted throttle opening is lower than a preset threshold, the short-term driving behavior learning model and the long-term driving behavior learning model are updated based on the actual throttle opening.

[0075] In one embodiment, the time compensation amount is provided with a maximum time compensation threshold to limit the time compensation amount from not exceeding the maximum time compensation threshold.

[0076] Specific limitations regarding the turbocharger control device 300 based on driving habits can be found in the above description of the turbocharger control method based on driving habits, and will not be repeated here. Each module in the aforementioned turbocharger control device 300 based on driving habits can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0077] In one embodiment, a vehicle is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement a turbocharger control method based on driving habits.

[0078] In one embodiment, a computer storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements a turbocharger control method based on driving habits.

[0079] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0080] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0081] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A turbocharger control method based on driving habits, characterized in that, include: Collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics and engine operating conditions; The throttle operating parameters are input into the short-term driving behavior learning model and the long-term driving behavior learning model respectively, and the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model are dynamically fused to determine the predicted throttle opening at future times. Based on the predicted throttle opening, the personalized coefficient, the air circuit coupling coefficient, and the engine operating parameters, determine the time compensation amount and opening compensation amount of the exhaust valve; The exhaust valve is controlled based on the time compensation amount and the opening compensation amount.

2. The turbocharger control method based on driving habits according to claim 1, characterized in that, The throttle operating parameters include throttle opening and throttle change rate.

3. The turbocharger control method based on driving habits according to claim 2, characterized in that, The step of dynamically fusing the first prediction result output by the short-term driving behavior learning model with the second prediction result output by the long-term driving behavior learning model includes: The weighting coefficient of the first prediction result is determined based on the throttle change rate; Based on the weighting coefficients, the first prediction result and the second prediction result are weighted and fused.

4. The turbocharger control method based on driving habits according to claim 2, characterized in that, The step of determining the time compensation and opening compensation of the exhaust valve based on the predicted throttle opening, the individualized coefficient, the air circuit coupling coefficient, and engine operating parameters includes: The time compensation amount is determined based on the personalization coefficient, the inherent delay parameter of the turbocharger, the sensitivity adjustment coefficient, and the throttle change rate. The opening compensation amount is determined based on the air path coupling coefficient, the predicted throttle opening, and the engine speed normalization factor.

5. The turbocharger control method based on driving habits according to claim 1, characterized in that, The method includes: Monitor the actual throttle opening; If the deviation between the actual throttle opening and the predicted throttle opening continues to exceed a preset threshold, then the exhaust valve is pressurized based on the actual throttle opening.

6. The turbocharger control method based on driving habits according to claim 5, characterized in that, The method further includes: If the deviation between the actual throttle opening and the predicted throttle opening is lower than a preset threshold, then the short-term driving behavior learning model and the long-term driving behavior learning model are updated based on the actual throttle opening.

7. The turbocharger control method based on driving habits according to claim 1, characterized in that, The time compensation amount is provided with a maximum time compensation threshold to limit the time compensation amount from not exceeding the maximum time compensation threshold.

8. A turbocharger control device based on driving habits, characterized in that, include: The data processing module is used to collect throttle operation parameters and obtain personalized coefficients and air-circuit coupling coefficients based on the driver's historical driving characteristics and engine operating conditions. The model prediction module is used to input the throttle operation parameters into a short-term driving behavior learning model and a long-term driving behavior learning model respectively, and dynamically fuse the first prediction result output by the short-term driving behavior learning model and the second prediction result output by the long-term driving behavior learning model to determine the predicted throttle opening at future times. The compensation amount determination module is used to determine the time compensation amount and opening compensation amount of the exhaust valve based on the predicted throttle opening, the personalized coefficient, the air circuit coupling coefficient and the engine operating parameters. The control module is used to control the exhaust valve based on the time compensation amount and the opening compensation amount.

9. A vehicle comprising a memory and a processor, said memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the turbocharger control method based on driving habits as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the turbocharger control method based on driving habits as described in any one of claims 1 to 7.

Images