Hybrid commercial vehicle combined ramp road adaptive shifting method

Abstract

The application provides a combined road adaptive shifting method for a hybrid commercial vehicle on a curved slope, which is used for improving shifting reliability and curved slope stability under uncertain road look-ahead information. The method obtains vehicle state and front road slope, curvature, superhigh rate and speed limit information, calculates road look-ahead information confidence according to map error, positioning error and sensor residual error, and determines curved slope safety vehicle speed and stability margin in combination with vehicle mass, road adhesion coefficient and structure parameters; generates a candidate gear sequence, recursively predicts vehicle speed, solves engine and drive motor power distribution and calculates comprehensive cost; adaptively determines shifting benefit threshold and shortest gear retention time according to comprehensive confidence, current stability margin and candidate sequence stability margin improvement, and outputs target gear and power source torque instruction; and switches to feedback shifting control under low confidence. The method can reduce false early shifting and frequent shifting, and takes into account stability, power and fuel economy.

Description

Technical Field

[0001] This invention relates to the fields of vehicle control for hybrid commercial vehicles and automatic transmission shift control technology, specifically to an adaptive shifting method for hybrid commercial vehicles on combined curves and slopes. Background Technology

[0002] During the operation of hybrid commercial vehicles, the transmission gear not only affects the operating speed, output torque, and power system efficiency of the engine and drive motor, but also influences power interruption during gear shifts, changes in vehicle longitudinal speed, and power demand from the battery. Therefore, gear selection is significantly coupled with the power distribution among the engine, drive motor, and battery. For hybrid commercial vehicles with high mass, a high center of gravity, and significant load variations, when driving on roads with both curves and inclines, gear selection affects not only overall vehicle energy consumption but also power reserves, engine braking capability, regenerative braking capability, and longitudinal and lateral stability during power interruptions during gear shifts. Road gradient alters vehicle drag and power demand, road curvature changes lateral acceleration demand, and road superelevation and road surface adhesion coefficient further alter tire adhesion utilization and vehicle rollover limits. If fixed shift thresholds and fixed gear holding times calibrated for straight roads are still used, problems such as delayed shifts before uphill climbs, frequent shifts on curved and sloping sections, insufficient braking capacity on downhill sections, or the power system's operating point deviating from its high-efficiency zone after gear shifts may occur.

[0003] With the development of high-precision maps, satellite positioning, electronic horizons, and vehicle-mounted sensing technologies, vehicles can obtain information on road gradient, curvature, superelevation, and speed limits a certain distance ahead, enabling predictive speed planning, energy management, and shift control. However, the reliability of this road-look-ahead information is affected by factors such as map accuracy, map version, vehicle longitudinal positioning error, road parameter spatial gradient, and residuals from onboard sensors. When a vehicle is in an area where road parameters change rapidly or when positioning errors are large, the same map error may cause significant deviations in gradient, curvature, or superelevation predictions. If shift control always uses road-look-ahead information with a fixed level of reliability, unnecessary early upshifts or downshifts may occur due to errors in far-end road information; conversely, completely abandoning road-look-ahead information fails to leverage the benefits of predictive shifting in terms of power, economy, and stability. Therefore, it is necessary to adjust the triggering conditions for early shifting based on the actual reliability of the road-look-ahead information.

[0004] Patent CN114148325B proposes a predictive energy management method for heavy-duty hybrid commercial vehicles. This method divides the road into sections based on the road slope ahead, obtains the planned vehicle speed and gear trajectory through dynamic programming, and determines the energy distribution rules between hybrid components using a rolling optimization method based on future road shape changes. This scheme can improve the speed planning and energy distribution of heavy-duty hybrid commercial vehicles by utilizing road slope, with its control focus on reducing powertrain energy consumption based on changes in road slope and road shape. For roads with combined curves and slopes, where road curvature, superelevation, and positioning errors coexist, gear shifting decisions also need to consider lateral adhesion, rollover boundaries, and the reliability of forward-looking road information; otherwise, the planned gear obtained based on deterministic road information may be difficult to adapt to real-time changes in road information deviation and stability risks. Patent CN107097791B proposes a speed optimization control method for four-wheel-drive electric vehicles based on road slope and curvature. This method utilizes road slope and curvature to establish a longitudinal dynamics model, an energy consumption objective function, and lateral safe vehicle speed constraints, and obtains the target vehicle speed and target torque through discrete dynamic programming. This scheme comprehensively considers road energy consumption and cornering safety during speed trajectory optimization, with its stability constraints primarily used to limit the vehicle's speed trajectory. For hybrid commercial vehicles with engines, drive motors, and stepped automatic transmissions, gear changes also alter the operating points of power components, shift power interruption, and braking capabilities, thus requiring further coordination of the relationships between candidate gear sequences, hybrid power distribution, stability margin, and shift triggering conditions. Patent CN109058450B proposes a method for cornering identification and shifting control of a mechanical automatic transmission in commercial vehicles. This method estimates the vehicle's turning direction and road curvature based on wheel speed differences, and limits the estimation process of operating parameters such as vehicle load and dynamic slope based on the current road curvature, thereby correcting the cornering shifting control. This scheme helps reduce the impact of state parameter estimation errors during vehicle cornering on shifting control, with its control focus on correcting the shifting strategy based on the vehicle's current cornering state. For hybrid commercial vehicles that need to complete gear preparation before entering a combination of curves and slopes, future road information, road information uncertainty, vehicle stability boundaries, and hybrid power distribution also need to be incorporated into a unified candidate gear evaluation process.

[0005] Overall, existing solutions still have the following problems: First, traditional shifting rules mainly determine the target gear based on the vehicle's current state and fixed shifting boundaries, making it difficult to adapt to road conditions with varying slopes, curvatures, and superelevation. Second, existing predictive shifting or predictive energy management methods typically use road look-ahead information as deterministic input, lacking mechanisms to evaluate the reliability of this information based on map errors, positioning errors, and sensor residuals. Third, existing vehicle stability constraints are mainly used for speed planning or current curve control, and further solutions are needed to address how curve stability margins can be incorporated into the evaluation of candidate gear sequences for hybrid commercial vehicles. Fourth, fixed shifting benefit thresholds and fixed gear holding times cannot simultaneously meet the requirements of suppressing erroneous early shifts under low confidence conditions and timely and safe shifting under low stability margin conditions.

[0006] Therefore, it is necessary to propose an adaptive shifting method for hybrid commercial vehicles on curved and sloping roads. This method adjusts the shifting benefit threshold and the shortest gear holding time of the candidate gear sequence online based on the confidence level of road prospective information and the stability margin of curved and sloping roads. When the confidence level of road prospective information is insufficient or the vehicle stability margin is too low, corresponding degradation control or safety priority control is implemented to coordinate the vehicle's energy consumption, shifting frequency, power system efficiency and driving stability on curved and sloping roads. Summary of the Invention

[0007] The main technical problem addressed by this invention is to overcome the difficulty of existing hybrid commercial vehicle shift control methods in simultaneously adapting to the stability risks and road look-ahead information errors in combined curve and slope road conditions. This invention proposes an adaptive shifting method for hybrid commercial vehicles on such roads. This method calculates the road look-ahead information confidence level and curve / slope stability margin separately, evaluates the powertrain feasibility, energy consumption, shifting costs, and stability of multiple candidate gear sequences, and adjusts the shift triggering conditions online based on changes in road information confidence and stability margin.

[0008] The adaptive shifting refers to the shifting benefit threshold and the shortest gear holding time changing online with the comprehensive confidence of the prediction spatial domain, the current minimum stability margin, and the improvement of the minimum stability margin of the candidate gear sequence. When the comprehensive confidence of the prediction spatial domain is lower than the preset confidence lower limit, early shifting based on remote road information is restricted. When the current minimum stability margin is lower than the preset safety mandatory threshold, the conventional gear holding restriction is lifted and safety priority control is implemented.

[0009] Specifically, the following steps are included:

[0010] Step 1: Acquire the vehicle's current position, speed, longitudinal acceleration, current gear, duration of current gear, state of charge of the power battery, operating status of the engine and drive motor, and road slope, curvature, superelevation, and speed limit information within the predicted distance ahead; Calculate the road look-ahead information confidence level at each sampling point in the prediction spatial domain based on the road information measurement uncertainty, vehicle longitudinal positioning uncertainty, and the residual between the onboard sensors and the road look-ahead information; Combine the vehicle mass estimate, road surface adhesion coefficient estimate, and vehicle structural parameters to determine the curve slope safe speed envelope and curve slope stability margin in the prediction spatial domain.

[0011] Step 2: Using the current gear as the initial gear, generate multiple candidate gear sequences according to adjacent gear shifts, gear duration, engine and drive motor speed and torque boundaries, power battery power boundaries, and gear shift count boundaries. For each candidate gear sequence, recursively deduce the vehicle kinetic energy and predicted vehicle speed along the prediction space domain, and solve the engine power and drive motor power allocation under the conditions of satisfying the vehicle power demand and power system constraints to obtain the equivalent fuel consumption, shift additional cost, and prediction stability margin corresponding to each candidate gear sequence.

[0012] Step 3: Using the candidate gear sequence that maintains the current gear as the baseline sequence, calculate the comprehensive cost improvement and minimum stability margin improvement of the remaining candidate gear sequences relative to the baseline sequence; based on the comprehensive confidence of the prediction spatial domain, the improvement of the minimum stability margin, and the current minimum stability margin, adaptively determine the shift benefit threshold and the shortest gear holding time for each candidate gear sequence, wherein, when the comprehensive confidence of the prediction spatial domain decreases, the shift benefit threshold is increased and the shortest gear holding time is extended; when the current minimum stability margin decreases, the shortest gear holding time is shortened; when the improvement of the minimum stability margin increases, the shift benefit threshold is decreased; determine the target gear sequence from the candidate gear sequences that satisfy the shift benefit threshold, the shortest gear holding time, and the powertrain constraints;

[0013] Step four: Execute the first gear command of the target gear sequence and the corresponding engine target torque and drive motor target torque; in the next control cycle, update the road look-ahead information confidence, curve slope stability margin, power battery state of charge prediction deviation and power distribution model parameters according to the actual vehicle response, and repeat steps one to four; when the comprehensive confidence of the prediction spatial domain is lower than the preset confidence lower limit, restrict or stop early gear shifting based on road look-ahead information, and switch to feedback gear control based on the real-time vehicle status.

[0014] Furthermore, the confidence level of the road forward information is calculated according to the following formula:

[0015]

[0016] In the formula, Indicates the current control cycle; Indicates the sampling point number within the prediction spatial domain; Indicates the road parameter index; Represents the set of road parameters, where, Indicates the road slope angle. Indicates the curvature of a signed road. This indicates the cross slope angle corresponding to the road superelevation; Represents the longitudinal spatial coordinates of the road; Represents road parameters In the The overall prediction variance at each sampling point; This indicates the variance of the road parameters provided by the map; This represents the gradient of road parameters relative to spatial coordinates; This represents the variance of the vehicle's longitudinal positioning error; This represents the residual between the vehicle sensor estimates and the road look-ahead information. This represents the confidence weight corresponding to the road parameters; Indicates the variance normalization scale; Indicates the first The confidence level of road prospective information at each sampling point, and .

[0017]

[0018] In the formula, This represents the overall confidence level of the prediction spatial domain for the current control cycle; This indicates the number of sampling points in the prediction spatial domain; Represents the distance attenuation factor, and ; Indicates the first The distance weights of each sampling point. Road sampling points closer to the vehicle's current location have a weight no less than that of road sampling points further away.

[0019] Furthermore, the safe speed envelope for curves and slopes simultaneously satisfies both tire adhesion constraints and vehicle rollover constraints, wherein the tire adhesion constraints are:

[0020]

[0021] In the formula, This represents the normalized vertical attachment utilization rate; Indicates vehicle speed as Normalized lateral attachment utilization rate at that time; This indicates the tire adhesion constraint index; This indicates the predicted longitudinal acceleration; This represents the vehicle speed variable used to calculate the constraint boundaries; This represents the estimated value of the road surface adhesion coefficient for the current control period; It represents the acceleration due to gravity.

[0022] The vehicle rollover constraint is:

[0023]

[0024] In the formula, Indicates vehicle speed as Vehicle rollover constraint indicators at that time; Indicates the wheelbase of the vehicle; Indicates the height of the vehicle's center of gravity. When... At that time, the vehicle meets the set quasi-static rollover constraint.

[0025]

[0026] In the formula, Indicates the first Safe vehicle speed at each sampling point on a bend; This indicates the road speed limit at the corresponding sampling point; This indicates the upper limit of the vehicle speed that satisfies both tire adhesion constraints and vehicle rollover constraints. Indicates the candidate gear sequence number; Indicates the first The candidate gear sequence in the first Predicted vehicle speed at each of the predicted sampling points; This represents the corresponding normalized bend slope stability margin. This represents a positive number to prevent the denominator from being zero; its dimensions are the same as the square of the vehicle speed. When When the predicted speed exceeds the safe speed envelope for curves and slopes, it indicates that the vehicle speed is expected to exceed the safe speed envelope for curves and slopes.

[0027] Furthermore, the candidate gear sequence satisfies:

[0028]

[0029] In the formula, Indicates the first A candidate gear sequence; Indicates the first The candidate gear sequence in the first The range of each predicted sampling point; Indicates the current actual gear; This represents the gear change between adjacent predicted sampling points, where, , and These respectively indicate downshifting, maintaining the current gear, and upshifting. Indicates the first The number of gear shifts in each candidate gear sequence within the prediction spatial domain; This indicates the maximum number of gear shifts allowed.

[0030]

[0031] In the formula, and These represent the equivalent kinetic energy of the vehicle at adjacent predicted sampling points; Indicates the driving force at the wheel end; This indicates an estimated vehicle weight. Indicates the rolling resistance coefficient; Indicates air density; Indicates the air drag coefficient; Indicates the vehicle's frontal area; Indicates the rotational mass conversion factor; Indicates the spatial sampling interval; Indicates the rotational speed at the shaft end of the power transmission system; Indicates gear position The corresponding transmission gear ratio; Indicates the speed ratio of the main reducer; This indicates the wheel rolling radius. Candidate gear sequences that do not meet the speed-torque boundaries of the engine, drive motor, and transmission are eliminated based on the axle end speed.

[0032] Furthermore, for each candidate gear sequence, a quadratic regression model of engine fuel consumption rate and a quadratic regression model of the rate of change of state of charge of the power battery are applied:

[0033]

[0034] In the formula, Indicates engine fuel consumption rate; Indicates engine power; Indicates the power of the drive motor; , and Represents the regression coefficients of the quadratic regression model of engine fuel consumption rate; Indicates the state of charge of the power battery; Indicates the rate of change of the state of charge of the power battery; , , , , and The regression coefficients represent the regression coefficients of the quadratic regression model for the rate of change of the state of charge of the power battery.

[0035] The engine power is determined based on the difference between the vehicle's required power and the drive motor's power, and the Hamiltonian function is constructed as follows:

[0036]

[0037] In the formula, Indicates the power required by the power system; Indicates the first The candidate gear sequence in the first Hamiltonian function for each predicted sampling point; This represents the costate variable corresponding to the state of charge of the power battery.

[0038] Taking the partial derivative of the Hamiltonian function with respect to the drive motor power, we obtain the stationary points, and then combine these stationary points with the boundary of the feasible power interval of the drive motor to form a finite set of candidate points.

[0039]

[0040]

[0041] In the formula, This represents the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor; Indicates in The stationary point obtained at that time; and These represent the lower and upper limits of the feasible power of the drive motor, respectively. Indicates the variable Limit to the lower limit and upper limit The truncation function between; This represents the set of candidate points with finite drive motor power. Indicates the candidate variable for drive motor power; Indicates the optimal drive motor power; This represents the independent variable that minimizes the Hamiltonian function.

[0042] When the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor is greater than zero, the truncated stationary point is added to the finite candidate point set; when the coefficient of the quadratic term is not greater than zero, only the two boundary points of the feasible power range of the drive motor are retained; the optimal drive motor power is determined by comparing the Hamiltonian function values ​​of each point in the finite candidate point set, and the optimal engine power is determined by the difference between the vehicle's required power and the optimal drive motor power.

[0043] Furthermore, the power of the battery is converted into an equivalent fuel consumption rate, and an additional cost for shifting gears is established, including the synchronous speed difference, required torque, and power interruption time.

[0044]

[0045] In the formula, Indicates the equivalent fuel consumption rate; This indicates the power of the battery, with discharge as the positive value. It represents the energy equivalence coefficient determined based on the charging or discharging state; Indicates the charging and discharging efficiency of the power battery; This indicates the fuel has a low calorific value; This indicates the additional cost of shifting gears; Indicates the change in gear position; Indicates characteristic functions; This indicates the difference in shaft end speed before and after gear shift synchronization; Indicates the required torque; Indicates the expected duration of power outage; , and This represents the additional cost coefficient for shifting gears.

[0046] The stability risk penalty and overall cost of the candidate gear sequence are as follows:

[0047]

[0048] In the formula, Indicates a penalty for stability risk on a bend or slope; Indicates the target stability margin; This indicates a low-confidence shift penalty; This indicates the weight of the low-confidence shift penalty; Indicates the overall cost of the stage; Indicates the predicted sampling time interval; Indicates the reference equivalent kinetic energy; This represents a positive number that prevents the denominator from becoming zero when the kinetic energy is normalized. Indicates the first The combined cost of a candidate gear sequence; This indicates the predicted terminal state of charge in the spatial domain; Indicates the target state of charge; , , and These represent the weights corresponding to kinetic energy tracking error, stability risk, shifting additional cost, and terminal state of charge deviation, respectively. Each weight is used to uniformly convert the corresponding cost item to the comprehensive cost dimension.

[0049] Furthermore, taking the candidate gear sequence that maintains the current gear position as the 0th candidate gear sequence, the overall cost improvement and minimum stability margin improvement of the ith candidate gear sequence are calculated according to the following formula:

[0050]

[0051] In the formula, This represents the overall cost of maintaining the baseline candidate gear sequence at the current gear position; Indicates the first The overall cost improvement of each candidate gear sequence relative to the baseline candidate gear sequence; Indicates the first The minimum stability margin of each candidate gear sequence in the prediction spatial domain; This represents the minimum stability margin of the baseline candidate gear sequence in the prediction space domain; Indicates the first The minimum stability margin improvement of each candidate gear sequence relative to the baseline candidate gear sequence.

[0052] The shift gain threshold for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin improvement.

[0053]

[0054] In the formula, This indicates the threshold for unlimited gear shifting revenue; This represents the basic shift benefit threshold; This represents the confidence correction coefficient for road forward information; This represents the correction factor for improving stability margin; This indicates the threshold for shift benefit after limiting the range; Indicates the variable Limit to the lower limit and upper limit Amplitude limiting function between; and These represent the lower and upper limits of the shift benefit threshold, respectively.

[0055] The shortest gear holding time for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin when maintaining the current gear:

[0056]

[0057] In the formula, Indicates the shortest duration of holding gear without limit; Indicates the base gear holding time; This represents the correction factor for the gear holding time based on the confidence level of road forward information. This represents the correction factor for the current stability margin on the gear holding time; Indicates the shortest duration of gear hold after limiting; and These represent the lower and upper limits of the shortest gear holding time, respectively. The lower the confidence level of the road prospect information, the larger the shift benefit threshold and the longer the shortest gear holding time; the greater the stability margin improvement brought by the candidate gear sequence, the smaller the shift benefit threshold; the lower the minimum stability margin of the baseline candidate gear sequence, the shorter the shortest gear holding time.

[0058] The candidate gear sequence is allowed to be selected when the overall cost improvement of the candidate gear sequence is greater than the corresponding shift benefit threshold and the duration of the current gear is not less than the corresponding shortest gear holding time; when multiple candidate gear sequences meet the conditions, the candidate gear sequence with the lowest overall cost is selected.

[0059] Furthermore, when the overall confidence level of the predicted spatial domain is lower than the preset confidence level lower limit, the early upshift or downshift triggered by the remote road information is canceled, and the candidate gear range is limited to the current gear and the feasible gears adjacent to the current gear. Feedback gear control is implemented based on real-time vehicle speed, real-time demand torque, powertrain shaft end speed, and current curve stability margin. When the current minimum stability margin is lower than the preset safety mandatory threshold, the minimum gear holding time constraint is released. Among the candidate gear sequences that meet the constraints of engine anti-drag speed, drive motor braking power, and tire adhesion, the candidate gear sequence that can improve the minimum stability margin without causing powertrain overspeed is selected first, and regenerative braking torque and friction braking compensation torque are output simultaneously.

[0060] Furthermore, after executing the first control variable of the target gear sequence in each control cycle, the co-state variable is corrected based on the deviation between the predicted terminal power battery state of charge and the target state of charge, and the road parameter prediction variance is updated based on the residual between the on-board sensors and the road look-ahead information.

[0061]

[0062] In the formula, This represents the unlimited costate variable for the next control cycle; The Hamiltonian function representing the current control cycle; Indicates the terminal state-of-charge deviation feedback gain; This indicates the predicted terminal state of charge in the spatial domain; and These represent the lower and upper bounds of the costate variable, respectively. Represents road parameters The corresponding prediction variance forgetting factor, and .

[0063] Furthermore, the hybrid commercial vehicle is a P2 coaxial parallel hybrid commercial vehicle in which the engine, disengagement clutch, drive motor, and mechanical automatic transmission are connected in sequence; the feasibility constraints of the candidate gear sequence further include the engagement or disengagement state of the disengagement clutch, the start or stop state of the engine, the synchronization capability of the drive motor speed during gear shifting, and the maximum charging and discharging power of the power battery; the first gear command, the target torque of the engine, the target torque of the drive motor, the target state of the disengagement clutch, and the friction braking compensation torque are output by the vehicle controller.

[0064] Compared with the prior art, the beneficial effects of the present invention are:

[0065] 1. The present invention discloses an adaptive shifting method for hybrid commercial vehicles on curved and sloping roads. Based on map data errors, vehicle longitudinal positioning errors, road parameter spatial gradients, and the residuals between onboard sensors and road forward-looking information, the method calculates the confidence level of road forward-looking information at each road sampling point. A comprehensive confidence level in the prediction spatial domain is then formed according to the distance between sampling points. This enables shifting control to distinguish between reliable road forward-looking information and road forward-looking information with significant errors. When the comprehensive confidence level decreases, the shifting benefit threshold is increased and the minimum gear holding time is extended, thereby suppressing unnecessary early upshifts, early downshifts, and frequent shifts caused by map errors, positioning errors, and perception biases. This improves the adaptability of predictive shifting control to road information uncertainties.

[0066] 2. The adaptive gear shifting method for hybrid commercial vehicles on curved and sloping roads described in this invention integrates road slope, road curvature, road superelevation, road surface adhesion coefficient, vehicle mass, vehicle wheelbase, and center of gravity height to establish longitudinal-lateral adhesion constraints and vehicle rollover constraints, thereby determining the safe speed envelope for curved and sloping roads and the stability margin corresponding to the candidate gear sequence. By introducing stability risk penalties and minimum stability margin improvement during the evaluation process of the candidate gear sequence, the gear decision considers not only the energy consumption of the power system but also the impact of the candidate gear on the driving stability of curved and sloping roads, thereby reducing the risk of adhesion constraint overstepping and rollover of high-load commercial vehicles on curved and sloping roads.

[0067] 3. The adaptive shifting method for hybrid commercial vehicles on curved and hilly roads described in this invention recursively calculates the vehicle's equivalent kinetic energy and predicted speed for each candidate gear sequence, solves the power distribution between the engine and drive motor, and evaluates the candidate gear sequences comprehensively based on equivalent fuel consumption, shifting costs, low-confidence shifting penalties, and stability risk penalties. Furthermore, based on the comprehensive confidence level in the predicted spatial domain, the current minimum stability margin, and the improvement in the minimum stability margin of the candidate gear sequences, the shifting benefit threshold and the shortest gear holding time are adjusted online. This increases the shifting threshold when the road information credibility decreases, shortens the waiting time for safe shifting when the vehicle stability margin decreases, and lowers the shifting threshold when a candidate gear can improve stability, thereby coordinating shifting reliability, curved and hilly driving stability, and the energy utilization efficiency of the hybrid system. Attached Figure Description

[0068] The invention will now be further described with reference to the accompanying drawings:

[0069] Figure 1 This is a flowchart illustrating the adaptive shifting method for hybrid commercial vehicles on curved and sloping roads as described in this invention.

[0070] Figure 2 This is a diagram showing the calculation of road look-ahead information confidence and curve / slope stability margin in a hybrid commercial vehicle curve / slope combined road adaptive shifting method according to the present invention.

[0071] Figure 3 This is an adaptive evaluation and decision graph of candidate gear sequence in the adaptive gear shifting method for hybrid commercial vehicles on curved and sloping roads as described in this invention. Detailed Implementation

[0072] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0073] This embodiment takes a P2 coaxial parallel hybrid commercial vehicle equipped with an engine, a disengagement clutch, a drive motor, and a mechanical automatic transmission as the control object.

[0074] like Figure 1As shown in this embodiment, an adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads includes the following steps: First, acquiring the real-time operating status of the vehicle and the road prospective information in the predicted spatial domain, calculating the road prospective information confidence, the curve-slope safe speed envelope, and the curve-slope stability margin; Second, generating multiple candidate gear sequences with the current gear as the initial gear, and performing vehicle state recursion, power system feasibility judgment, engine and drive motor power allocation, and comprehensive cost calculation on each candidate gear sequence; Third, adaptively determining the shifting benefit threshold and the shortest gear holding time based on the road prospective information confidence, the current minimum stability margin, and the stability margin improvement amount corresponding to the candidate gear sequence, and determining the target gear sequence from the candidate gear sequences that meet the constraints; Finally, executing the first gear command and corresponding torque command of the target gear sequence, updating the road information confidence and control model parameters based on the actual vehicle response, and entering the next control cycle. When the confidence level of the road look-ahead information is lower than the preset confidence level limit, the controller restricts or stops early gear shifting based on the road look-ahead information and switches to feedback gear control based on the real-time vehicle status.

[0075] Step 1: Acquire the vehicle's current position, speed, longitudinal acceleration, current gear, duration of current gear, state of charge of the power battery, operating status of the engine and drive motor, and road slope, curvature, superelevation, and speed limit information within the predicted distance ahead; Calculate the road look-ahead information confidence level at each sampling point in the prediction spatial domain based on the road information measurement uncertainty, vehicle longitudinal positioning uncertainty, and the residual between the onboard sensors and the road look-ahead information; Combine the vehicle mass estimate, road surface adhesion coefficient estimate, and vehicle structural parameters to determine the curve slope safe speed envelope and curve slope stability margin in the prediction spatial domain.

[0076] Step 2: Using the current gear as the initial gear, generate multiple candidate gear sequences according to adjacent gear shifts, gear duration, engine and drive motor speed and torque boundaries, power battery power boundaries, and gear shift count boundaries. For each candidate gear sequence, recursively deduce the vehicle kinetic energy and predicted vehicle speed along the prediction space domain, and solve the engine power and drive motor power allocation under the conditions of satisfying the vehicle power demand and power system constraints to obtain the equivalent fuel consumption, shift additional cost, and prediction stability margin corresponding to each candidate gear sequence.

[0077] Step 3: Using the candidate gear sequence that maintains the current gear as the baseline sequence, calculate the comprehensive cost improvement and minimum stability margin improvement of the remaining candidate gear sequences relative to the baseline sequence; based on the comprehensive confidence of the prediction spatial domain, the improvement of the minimum stability margin, and the current minimum stability margin, adaptively determine the shift benefit threshold and the shortest gear holding time for each candidate gear sequence, wherein, when the comprehensive confidence of the prediction spatial domain decreases, the shift benefit threshold is increased and the shortest gear holding time is extended; when the current minimum stability margin decreases, the shortest gear holding time is shortened; when the improvement of the minimum stability margin increases, the shift benefit threshold is decreased; determine the target gear sequence from the candidate gear sequences that satisfy the shift benefit threshold, the shortest gear holding time, and the powertrain constraints;

[0078] Step four: Execute the first gear command of the target gear sequence and the corresponding engine target torque and drive motor target torque; in the next control cycle, update the road look-ahead information confidence, curve slope stability margin, power battery state of charge prediction deviation and power distribution model parameters according to the actual vehicle response, and repeat steps one to four; when the comprehensive confidence of the prediction spatial domain is lower than the preset confidence lower limit, restrict or stop early gear shifting based on road look-ahead information, and switch to feedback gear control based on the real-time vehicle status.

[0079] like Figure 2 As shown, this embodiment performs confidence evaluation and curve stability evaluation on the road look-ahead information. The confidence calculation branch calculates the confidence of each road sampling point in the prediction spatial domain based on the map road parameter variance, vehicle longitudinal positioning uncertainty, spatial gradient of road parameters along the driving direction, and the residual between the onboard sensor estimates and the road look-ahead information, and further obtains the overall confidence of the prediction spatial domain. The curve stability calculation branch calculates tire adhesion constraints and vehicle rollover constraints based on the road slope angle, road curvature, road cross slope angle converted from road superelevation, estimated road surface adhesion coefficient, estimated vehicle mass, vehicle wheelbase, and vehicle center of gravity height, thereby determining the curve safety speed and curve stability margin corresponding to the candidate gear sequence for each road sampling point. The overall confidence of the prediction spatial domain and the curve stability margin together serve as inputs for subsequent candidate gear sequence evaluation and adaptive gear shifting decisions.

[0080] The confidence level of the road forward information is calculated according to the following formula:

[0081]

[0082] In the formula, Indicates the current control cycle; Indicates the sampling point number within the prediction spatial domain; Indicates the road parameter index; Represents the set of road parameters, where, Indicates the road slope angle. Indicates the curvature of a signed road. This indicates the cross slope angle corresponding to the road superelevation; Represents the longitudinal spatial coordinates of the road; Represents road parameters In the The overall prediction variance at each sampling point; This indicates the variance of the road parameters provided by the map; This represents the gradient of road parameters relative to spatial coordinates; This represents the variance of the vehicle's longitudinal positioning error; This represents the residual between the vehicle sensor estimates and the road look-ahead information. This represents the confidence weight corresponding to the road parameters; Indicates the variance normalization scale; Indicates the first The confidence level of road prospective information at each sampling point, and .

[0083]

[0084] In the formula, This represents the overall confidence level of the prediction spatial domain for the current control cycle; This indicates the number of sampling points in the prediction spatial domain; Represents the distance attenuation factor, and ; Indicates the first The distance weights of each sampling point. Road sampling points closer to the vehicle's current location have a weight no less than that of road sampling points further away.

[0085] The safe speed envelope for curves and slopes simultaneously satisfies both tire adhesion constraints and vehicle rollover constraints. The tire adhesion constraint is as follows:

[0086]

[0087] In the formula, This represents the normalized vertical attachment utilization rate; Indicates vehicle speed as Normalized lateral attachment utilization rate at that time; This indicates the tire adhesion constraint index; This indicates the predicted longitudinal acceleration; This represents the vehicle speed variable used to calculate the constraint boundaries; This represents the estimated value of the road surface adhesion coefficient for the current control period; It represents the acceleration due to gravity.

[0088] The vehicle rollover constraint is:

[0089]

[0090] In the formula, Indicates vehicle speed as Vehicle rollover constraint indicators at that time; Indicates the wheelbase of the vehicle; Indicates the height of the vehicle's center of gravity. When... At that time, the vehicle meets the set quasi-static rollover constraint.

[0091]

[0092] In the formula, Indicates the first Safe vehicle speed at each sampling point on a bend; This indicates the road speed limit at the corresponding sampling point; This indicates the upper limit of the vehicle speed that satisfies both tire adhesion constraints and vehicle rollover constraints. Indicates the candidate gear sequence number; Indicates the first The candidate gear sequence in the first Predicted vehicle speed at each of the predicted sampling points; This represents the corresponding normalized bend slope stability margin. This represents a positive number to prevent the denominator from being zero; its dimensions are the same as the square of the vehicle speed. When When the predicted speed exceeds the safe speed envelope for curves and slopes, it indicates that the vehicle speed is expected to exceed the safe speed envelope for curves and slopes.

[0093] like Figure 3 As shown, the controller uses the current gear as the initial gear and generates multiple candidate gear sequences under constraints such as adjacent shift rules, maximum number of shifts, gear duration, and power system operating boundaries. The candidate gear sequence that maintains the current gear is set as the baseline sequence. For each candidate gear sequence, the vehicle kinetic energy and predicted vehicle speed are recursively derived along the prediction space domain to determine the shaft end speed of the powertrain system and the vehicle's required power. Under the conditions of satisfying the constraints of the engine, drive motor, power battery, and transmission, the engine power and drive motor power allocation are solved to obtain the corresponding equivalent fuel consumption, shift additional cost, low-confidence shift penalty, and stability risk penalty.

[0094] The controller further calculates the overall cost improvement and minimum stability margin improvement of each candidate gear sequence relative to the baseline sequence, and adaptively adjusts the shift benefit threshold and minimum gear holding time based on the predicted spatial domain overall confidence, the current minimum stability margin, and the minimum stability margin improvement. When a candidate gear sequence simultaneously satisfies the overall cost improvement, gear holding time, and powertrain feasibility conditions, it is allowed to enter the target sequence selection set. When multiple candidate gear sequences meet the conditions, the candidate gear sequence with the lowest overall cost is selected as the target gear sequence. The controller executes the first gear command of the target gear sequence and simultaneously outputs the target engine torque and the target drive motor torque.

[0095] The candidate gear sequence satisfies:

[0096]

[0097] In the formula, Indicates the first A candidate gear sequence; Indicates the first The candidate gear sequence in the first The range of each predicted sampling point; Indicates the current actual gear; This represents the gear change between adjacent predicted sampling points, where, , and These respectively indicate downshifting, maintaining the current gear, and upshifting. Indicates the first The number of gear shifts in each candidate gear sequence within the prediction spatial domain; Indicates the maximum number of gear shifts allowed.

[0098]

[0099] In the formula, and These represent the equivalent kinetic energy of the vehicle at adjacent predicted sampling points; Indicates the driving force at the wheel end; This indicates an estimated vehicle weight. Indicates the rolling resistance coefficient; Indicates air density; Indicates the air drag coefficient; Indicates the vehicle's frontal area; Indicates the rotational mass conversion factor; Indicates the spatial sampling interval; Indicates the rotational speed at the shaft end of the power transmission system; Indicates gear position The corresponding transmission gear ratio; Indicates the speed ratio of the main reducer; This indicates the wheel rolling radius. Candidate gear sequences that do not meet the speed-torque boundaries of the engine, drive motor, and transmission are eliminated based on the axle end speed.

[0100] For each candidate gear sequence, a quadratic regression model of engine fuel consumption rate and a quadratic regression model of the rate of change of state of charge of the power battery are used:

[0101]

[0102] In the formula, Indicates engine fuel consumption rate; Indicates engine power; Indicates the power of the drive motor; , and Represents the regression coefficients of the quadratic regression model of engine fuel consumption rate; Indicates the state of charge of the power battery; Indicates the rate of change of the state of charge of the power battery; , , , , and The regression coefficients represent the regression coefficients of the quadratic regression model for the rate of change of the state of charge of the power battery.

[0103] The engine power is determined based on the difference between the vehicle's required power and the drive motor's power, and the Hamiltonian function is constructed as follows:

[0104]

[0105] In the formula, Indicates the power required by the power system; Indicates the first The candidate gear sequence in the first Hamiltonian function for each predicted sampling point; This represents the costate variable corresponding to the state of charge of the power battery.

[0106] Taking the partial derivative of the Hamiltonian function with respect to the drive motor power, we obtain the stationary points, and then combine these stationary points with the boundary of the feasible power interval of the drive motor to form a finite set of candidate points.

[0107]

[0108]

[0109] In the formula, This represents the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor; Indicates in The stationary point obtained at that time; and These represent the lower and upper limits of the feasible power of the drive motor, respectively. Indicates the variable Limit to the lower limit and upper limit The truncation function between; This represents the set of candidate points with finite drive motor power. Indicates the candidate variable for drive motor power; Indicates the optimal drive motor power; This represents the independent variable that minimizes the Hamiltonian function.

[0110] When the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor is greater than zero, the truncated stationary point is added to the finite candidate point set; when the coefficient of the quadratic term is not greater than zero, only the two boundary points of the feasible power range of the drive motor are retained; the optimal drive motor power is determined by comparing the Hamiltonian function values ​​of each point in the finite candidate point set, and the optimal engine power is determined by the difference between the vehicle's required power and the optimal drive motor power.

[0111] The power of the battery is converted into an equivalent fuel consumption rate, and an additional cost for shifting gears is established, including the synchronous speed difference, required torque, and power interruption time.

[0112]

[0113] In the formula, Indicates the equivalent fuel consumption rate; This indicates the power of the battery, with discharge as the positive value. It represents the energy equivalence coefficient determined based on the charging or discharging state; Indicates the charging and discharging efficiency of the power battery; This indicates the fuel has a low calorific value; This indicates the additional cost of shifting gears; Indicates the change in gear position; Indicates characteristic functions; This indicates the difference in shaft end speed before and after gear shift synchronization; Indicates the required torque; Indicates the expected duration of power outage; , and This represents the additional cost coefficient for shifting gears.

[0114] The stability risk penalty and overall cost of the candidate gear sequence are as follows:

[0115]

[0116] In the formula, Indicates a penalty for stability risk on a bend or slope; Indicates the target stability margin; This indicates a low-confidence shift penalty; This indicates the weight of the low-confidence shift penalty; Indicates the overall cost of the stage; Indicates the predicted sampling time interval; Indicates the reference equivalent kinetic energy; This represents a positive number that prevents the denominator from becoming zero when the kinetic energy is normalized. Indicates the first The combined cost of a candidate gear sequence; This indicates the predicted terminal state of charge in the spatial domain; Indicates the target state of charge; , , and These represent the weights corresponding to kinetic energy tracking error, stability risk, shifting additional cost, and terminal state of charge deviation, respectively. Each weight is used to uniformly convert the corresponding cost item to the comprehensive cost dimension.

[0117] Using the candidate gear sequence that maintains the current gear position as the 0th candidate gear sequence, the overall cost improvement and minimum stability margin improvement of the ith candidate gear sequence are calculated according to the following formula:

[0118]

[0119] In the formula, This represents the overall cost of maintaining the baseline candidate gear sequence at the current gear position; Indicates the first The overall cost improvement of each candidate gear sequence relative to the baseline candidate gear sequence; Indicates the first The minimum stability margin of each candidate gear sequence in the prediction spatial domain; This represents the minimum stability margin of the baseline candidate gear sequence in the prediction space domain; Indicates the first The minimum stability margin improvement of each candidate gear sequence relative to the baseline candidate gear sequence.

[0120] The shift gain threshold for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin improvement.

[0121]

[0122] In the formula, This indicates the threshold for unlimited gear shifting revenue; This represents the basic shift benefit threshold; This represents the confidence correction coefficient for road forward information; This represents the correction factor for improving stability margin; This indicates the threshold for shift benefit after limiting the range; Indicates the variable Limit to the lower limit and upper limit Amplitude limiting function between; and These represent the lower and upper limits of the shift benefit threshold, respectively.

[0123] The shortest gear holding time for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin when maintaining the current gear:

[0124]

[0125] In the formula, Indicates the shortest duration of holding gear without limit; Indicates the base gear holding time; This represents the correction factor for the gear holding time based on the confidence level of road forward information. This represents the correction factor for the current stability margin on the gear holding time; Indicates the shortest duration of gear hold after limiting; and These represent the lower and upper limits of the shortest gear holding time, respectively. The lower the confidence level of the road prospect information, the larger the shift benefit threshold and the longer the shortest gear holding time; the greater the stability margin improvement brought by the candidate gear sequence, the smaller the shift benefit threshold; the lower the minimum stability margin of the baseline candidate gear sequence, the shorter the shortest gear holding time.

[0126] The candidate gear sequence is allowed to be selected when the overall cost improvement of the candidate gear sequence is greater than the corresponding shift benefit threshold and the duration of the current gear is not less than the corresponding shortest gear holding time; when multiple candidate gear sequences meet the conditions, the candidate gear sequence with the lowest overall cost is selected.

[0127] When the overall confidence level of the predicted spatial domain is lower than the preset confidence level lower limit, the early upshift or downshift triggered by the remote road information is canceled, and the candidate gear range is limited to the current gear and the feasible gears adjacent to the current gear. Feedback gear control is implemented based on real-time vehicle speed, real-time demand torque, powertrain shaft end speed and current curve stability margin. When the current minimum stability margin is lower than the preset safety mandatory threshold, the minimum gear holding time constraint is released. Among the candidate gear sequences that meet the constraints of engine anti-drag speed, drive motor braking power and tire adhesion, the candidate gear sequence that can improve the minimum stability margin without causing powertrain overspeed is selected first, and regenerative braking torque and friction braking compensation torque are output simultaneously.

[0128] After executing the first control input of the target gear sequence in each control cycle, the co-state variables are corrected based on the deviation between the predicted terminal power battery state of charge and the target state of charge, and the road parameter prediction variance is updated based on the residual between the on-board sensors and the road look-ahead information.

[0129]

[0130] In the formula, This represents the unlimited costate variable for the next control cycle; The Hamiltonian function representing the current control cycle; Indicates the terminal state-of-charge deviation feedback gain; This indicates the predicted terminal state of charge in the spatial domain; and These represent the lower and upper bounds of the costate variable, respectively. Represents road parameters The corresponding prediction variance forgetting factor, and .

[0131] The hybrid commercial vehicle is a P2 coaxial parallel hybrid commercial vehicle in which the engine, disengagement clutch, drive motor and mechanical automatic transmission are connected in sequence; the feasibility constraints of the candidate gear sequence further include the engagement or disengagement state of the disengagement clutch, the start or stop state of the engine, the speed synchronization capability of the drive motor during gear shifting and the maximum charging and discharging power of the power battery; the first gear command, the target torque of the engine, the target torque of the drive motor, the target state of the disengagement clutch and the friction braking compensation torque are output by the vehicle controller.

Claims

1. A method for adaptive gear shifting on curved and slope-adaptive roads for hybrid commercial vehicles, characterized in that, Includes the following steps: Step 1: Acquire the vehicle's current position, speed, longitudinal acceleration, current gear, duration of current gear, state of charge of the power battery, operating status of the engine and drive motor, and road slope, curvature, superelevation, and speed limit information within the predicted distance ahead; Calculate the road look-ahead information confidence level at each sampling point in the prediction spatial domain based on the road information measurement uncertainty, vehicle longitudinal positioning uncertainty, and the residual between the onboard sensors and the road look-ahead information; Combine the vehicle mass estimate, road surface adhesion coefficient estimate, and vehicle structural parameters to determine the curve slope safe speed envelope and curve slope stability margin in the prediction spatial domain. Step 2: Using the current gear as the initial gear, generate multiple candidate gear sequences according to adjacent gear shifts, gear duration, engine and drive motor speed and torque boundaries, power battery power boundaries, and gear shift count boundaries. For each candidate gear sequence, recursively deduce the vehicle kinetic energy and predicted vehicle speed along the prediction space domain, and solve the engine power and drive motor power allocation under the conditions of satisfying the vehicle power demand and power system constraints to obtain the equivalent fuel consumption, shift additional cost, and prediction stability margin corresponding to each candidate gear sequence. Step 3: Using the candidate gear sequence that maintains the current gear position as the baseline sequence, calculate the overall cost improvement and minimum stability margin improvement of the remaining candidate gear sequences relative to the baseline sequence; Based on the overall confidence level of the predicted spatial domain, the improvement amount of the minimum stability margin, and the current minimum stability margin, the shift benefit threshold and the shortest gear holding time for each candidate gear sequence are adaptively determined. Specifically, when the overall confidence level of the predicted spatial domain decreases, the shift benefit threshold is increased and the shortest gear holding time is extended; when the current minimum stability margin decreases, the shortest gear holding time is shortened; and when the improvement amount of the minimum stability margin increases, the shift benefit threshold is decreased. The target gear sequence is determined from the candidate gear sequences that satisfy the shift benefit threshold, the shortest gear holding time, and the powertrain constraints. Step four: Execute the first gear command of the target gear sequence and the corresponding engine target torque and drive motor target torque; In the next control cycle, the road look-ahead information confidence level, curve slope stability margin, power battery state of charge prediction deviation and power distribution model parameters are updated according to the actual vehicle response, and steps one to four are repeated; when the comprehensive confidence level of the prediction spatial domain is lower than the preset confidence level lower limit, the advance shifting based on road look-ahead information is restricted or stopped, and the system is switched to feedback gear control based on the real-time vehicle status.

2. The adaptive gear shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 1, characterized in that, The confidence level of the road forward information is calculated according to the following formula: In the formula, Indicates the current control cycle; Indicates the sampling point number within the prediction spatial domain; Indicates the road parameter index; Represents the set of road parameters, where, Indicates the road slope angle. Indicates the curvature of a signed road. This indicates the cross slope angle corresponding to the road superelevation; Represents the longitudinal spatial coordinates of the road; Represents road parameters In the The overall prediction variance at each sampling point; This indicates the variance of the road parameters provided by the map; This represents the gradient of road parameters relative to spatial coordinates; This represents the variance of the vehicle's longitudinal positioning error; This represents the residual between the vehicle sensor estimates and the road look-ahead information. This represents the confidence weight corresponding to the road parameters; Indicates the variance normalization scale; Indicates the first The confidence level of road prospective information at each sampling point, and ; In the formula, This represents the overall confidence level of the prediction spatial domain for the current control cycle; This indicates the number of sampling points in the prediction spatial domain; Represents the distance attenuation factor, and ; Indicates the first The distance weight of each sampling point; the road sampling points closer to the current location of the vehicle have a weight no less than that of the road sampling points further away.

3. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 1, characterized in that, The safe speed envelope for curves and slopes simultaneously satisfies both tire adhesion constraints and vehicle rollover constraints. The tire adhesion constraint is as follows: In the formula, This represents the normalized vertical attachment utilization rate; Indicates vehicle speed as Normalized lateral attachment utilization rate at that time; This indicates the tire adhesion constraint index; This indicates the predicted longitudinal acceleration; This represents the vehicle speed variable used to calculate the constraint boundaries; This represents the estimated value of the road surface adhesion coefficient for the current control period; Represents gravitational acceleration; The vehicle rollover constraint is: In the formula, Indicates vehicle speed as Vehicle rollover constraint indicators at that time; Indicates the wheelbase of the vehicle; Indicates the height of the vehicle's center of gravity; when At that time, the vehicle meets the set quasi-static rollover constraint; In the formula, Indicates the first Safe vehicle speed at each sampling point on a bend; This indicates the road speed limit at the corresponding sampling point; This indicates the upper limit of the vehicle speed that satisfies both tire adhesion constraints and vehicle rollover constraints. Indicates the candidate gear sequence number; Indicates the first The candidate gear sequence in the first Predicted vehicle speed at each of the predicted sampling points; This represents the corresponding normalized bend slope stability margin. This represents a positive number to prevent the denominator from being zero; its dimensions are the same as the square of the vehicle speed. When the predicted speed exceeds the safe speed envelope for curves and slopes, it indicates that the vehicle speed is expected to exceed the safe speed envelope for curves and slopes.

4. The adaptive gear shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 1, characterized in that, The candidate gear sequence satisfies: In the formula, Indicates the first A candidate gear sequence; Indicates the first The candidate gear sequence in the first The range of each predicted sampling point; Indicates the current actual gear; This represents the gear change between adjacent predicted sampling points, where, , and These respectively indicate downshifting, maintaining the current gear, and upshifting. Indicates the first The number of gear shifts in each candidate gear sequence within the prediction spatial domain; Indicates the maximum number of gear shifts allowed; In the formula, and These represent the equivalent kinetic energy of the vehicle at adjacent predicted sampling points; Indicates the driving force at the wheel end; This indicates an estimated vehicle weight. Indicates the rolling resistance coefficient; Indicates air density; Indicates the air drag coefficient; Indicates the vehicle's frontal area; Indicates the rotational mass conversion factor; Indicates the spatial sampling interval; Indicates the rotational speed at the shaft end of the power transmission system; Indicates gear position The corresponding transmission gear ratio; Indicates the speed ratio of the main reducer; This indicates the wheel rolling radius; candidate gear sequences that do not meet the speed and torque boundaries of the engine, drive motor, and transmission are eliminated based on the axle end speed.

5. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 4, characterized in that, For each candidate gear sequence, a quadratic regression model of engine fuel consumption rate and a quadratic regression model of the rate of change of state of charge of the power battery are used: In the formula, Indicates engine fuel consumption rate; Indicates engine power; Indicates the power of the drive motor; , and Represents the regression coefficients of the quadratic regression model of engine fuel consumption rate; Indicates the state of charge of the power battery; Indicates the rate of change of the state of charge of the power battery; , , , , and Represents the regression coefficients of the quadratic regression model for the rate of change of state of charge of the power battery; The engine power is determined based on the difference between the vehicle's required power and the drive motor's power, and the Hamiltonian function is constructed as follows: In the formula, Indicates the power required by the power system; Indicates the first The candidate gear sequence in the first Hamiltonian function for each predicted sampling point; This represents the costate variable corresponding to the state of charge of the power battery; Taking the partial derivative of the Hamiltonian function with respect to the drive motor power, we obtain the stationary points, and then combine these stationary points with the boundary of the feasible power interval of the drive motor to form a finite set of candidate points. In the formula, This represents the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor; Indicates in The stationary point obtained at that time; and These represent the lower and upper limits of the feasible power of the drive motor, respectively. Indicates the variable Limit to the lower limit and upper limit The truncation function between; This represents the set of candidate points with finite drive motor power. Indicates the candidate variable for drive motor power; Indicates the optimal drive motor power; Represents the independent variable that minimizes the Hamiltonian function; When the coefficient of the quadratic term of the Hamiltonian function with respect to the power of the drive motor is greater than zero, the truncated stationary point is added to the finite candidate point set; when the coefficient of the quadratic term is not greater than zero, only the two boundary points of the feasible power range of the drive motor are retained; the optimal drive motor power is determined by comparing the Hamiltonian function values ​​of each point in the finite candidate point set, and the optimal engine power is determined by the difference between the vehicle's required power and the optimal drive motor power.

6. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 5, characterized in that, The power of the battery is converted into an equivalent fuel consumption rate, and an additional cost for shifting gears is established, including the synchronous speed difference, required torque, and power interruption time. In the formula, Indicates the equivalent fuel consumption rate; This indicates the power of the battery, with discharge as the positive value. It represents the energy equivalence coefficient determined based on the charging or discharging state; Indicates the charging and discharging efficiency of the power battery; This indicates the fuel has a low calorific value; This indicates the additional cost of shifting gears; Indicates the change in gear position; Indicates characteristic functions; This indicates the difference in shaft end speed before and after gear shift synchronization; Indicates the required torque; Indicates the expected duration of power outage; , and This represents the additional cost coefficient for shifting gears; The stability risk penalty and overall cost of the candidate gear sequence are as follows: In the formula, Indicates a penalty for stability risk on a bend or slope; Indicates the target stability margin; This indicates a low-confidence shift penalty; This indicates the weight of the low-confidence shift penalty; Indicates the overall cost of the stage; Indicates the predicted sampling time interval; Indicates the reference equivalent kinetic energy; This represents a positive number that prevents the denominator from becoming zero when the kinetic energy is normalized. Indicates the first The combined cost of a candidate gear sequence; This indicates the predicted terminal state of charge in the spatial domain; Indicates the target state of charge; , , and These represent the weights corresponding to kinetic energy tracking error, stability risk, shifting additional cost, and terminal state of charge deviation, respectively. Each weight is used to uniformly convert the corresponding cost item to the comprehensive cost dimension.

7. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 6, characterized in that, Using the candidate gear sequence that maintains the current gear position as the 0th candidate gear sequence, the overall cost improvement and minimum stability margin improvement of the ith candidate gear sequence are calculated according to the following formula: In the formula, This represents the overall cost of maintaining the baseline candidate gear sequence at the current gear position; Indicates the first The overall cost improvement of each candidate gear sequence relative to the baseline candidate gear sequence; Indicates the first The minimum stability margin of each candidate gear sequence in the prediction spatial domain; This represents the minimum stability margin of the baseline candidate gear sequence in the prediction space domain; Indicates the first The minimum stability margin improvement of each candidate gear sequence relative to the baseline candidate gear sequence; The shift gain threshold for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin improvement. In the formula, This indicates the threshold for unlimited gear shifting revenue; This represents the basic shift benefit threshold; This represents the confidence correction coefficient for road forward information; This represents the correction factor for improving stability margin; This indicates the threshold for shift benefit after limiting the range; Indicates the variable Limit to the lower limit and upper limit Amplitude limiting function between; and These represent the lower and upper limits of the shift benefit threshold, respectively; The shortest gear holding time for the i-th candidate gear sequence is determined based on the comprehensive confidence level of the predicted spatial domain and the minimum stability margin when maintaining the current gear: In the formula, Indicates the shortest duration of holding gear without limit; Indicates the base gear holding time; This represents the correction factor for the gear holding time based on the confidence level of road forward information. This represents the correction factor for the current stability margin on the gear holding time; Indicates the shortest duration of gear hold after limiting; and These represent the lower and upper limits of the shortest gear holding time, respectively; the lower the confidence level of the road prospect information, the larger the shift benefit threshold and the longer the shortest gear holding time; the greater the stability margin improvement brought by the candidate gear sequence, the smaller the shift benefit threshold; the lower the minimum stability margin of the baseline candidate gear sequence, the shorter the shortest gear holding time. The candidate gear sequence is allowed to be selected when the overall cost improvement of the candidate gear sequence is greater than the corresponding shift benefit threshold and the duration of the current gear is not less than the corresponding shortest gear holding time; when multiple candidate gear sequences meet the conditions, the candidate gear sequence with the lowest overall cost is selected.

8. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 7, characterized in that, When the overall confidence level of the predicted spatial domain is lower than the preset confidence level lower limit, the early upshift or downshift triggered by the remote road information is canceled, and the candidate gear range is limited to the current gear and the feasible gears adjacent to the current gear. Feedback gear control is implemented based on real-time vehicle speed, real-time demand torque, powertrain shaft end speed and current curve stability margin. When the current minimum stability margin is lower than the preset safety mandatory threshold, the minimum gear holding time constraint is released. Among the candidate gear sequences that meet the constraints of engine anti-drag speed, drive motor braking power and tire adhesion, the candidate gear sequence that can improve the minimum stability margin without causing powertrain overspeed is selected first, and regenerative braking torque and friction braking compensation torque are output simultaneously.

9. The adaptive shifting method for hybrid commercial vehicles on curved and slope-adaptive roads according to claim 7, characterized in that, After executing the first control input of the target gear sequence in each control cycle, the co-state variables are corrected based on the deviation between the predicted terminal power battery state of charge and the target state of charge, and the road parameter prediction variance is updated based on the residual between the on-board sensors and the road look-ahead information. In the formula, This represents the unlimited costate variable for the next control cycle; The Hamiltonian function representing the current control cycle; Indicates the terminal state-of-charge deviation feedback gain; This indicates the predicted terminal state of charge in the spatial domain; and These represent the lower and upper bounds of the costate variable, respectively. Represents road parameters The corresponding prediction variance forgetting factor, and .

10. The adaptive shifting method for hybrid commercial vehicles on curved and slope-combined roads according to any one of claims 1 to 9, characterized in that, The hybrid commercial vehicle is a P2 coaxial parallel hybrid commercial vehicle in which the engine, disengagement clutch, drive motor and mechanical automatic transmission are connected in sequence; the feasibility constraints of the candidate gear sequence further include the engagement or disengagement state of the disengagement clutch, the start or stop state of the engine, the speed synchronization capability of the drive motor during gear shifting and the maximum charging and discharging power of the power battery; the first gear command, the target torque of the engine, the target torque of the drive motor, the target state of the disengagement clutch and the friction braking compensation torque are output by the vehicle controller.

Images