A method, device and equipment for estimating the total mass and aerodynamic resistance of a passenger car

Abstract

The application provides a passenger car whole vehicle mass and air resistance combined estimation method, device and equipment, and relates to the technical field of vehicle dynamics parameter estimation.The application comprises the following steps: collecting a current vehicle speed in real time, obtaining a passenger car standard windward area and an initial wind resistance coefficient, and calculating an air resistance benchmark value; when the vehicle speed is lower than a first threshold value, locking the whole vehicle mass by using a recursive least square method based on a rolling resistance model; when the vehicle speed is higher than a second threshold value, substituting the locked whole vehicle mass into a longitudinal dynamics equation as a known constant, and solving a real-time wind resistance coefficient; performing online rationality verification based on a national standard benchmark coefficient; and when a vehicle door is detected to be closed, re-executing the mass locking step.The application can realize accurate estimation of the mass and the wind resistance coefficient respectively, shorten the algorithm convergence time, respond to passenger changes in real time, and improve long-term operation reliability.

Description

Technical Field

[0001] This invention relates to the field of vehicle dynamics parameter estimation technology, and more specifically, to a method, apparatus, equipment, and medium for jointly estimating the total mass and aerodynamic drag of a bus. Background Technology

[0002] With the rapid development of new energy vehicle technology, buses, as an important component of public transportation, require optimized energy consumption and control strategies for their operation to improve operational efficiency and extend driving range. In the bus control system, vehicle mass and aerodynamic drag (real-time dynamic air resistance) are two key parameters that directly affect the accuracy of power demand calculations, energy recovery strategies, and driving range estimations. Vehicle mass determines the inertial loss during acceleration and hill climbing, while aerodynamic drag is closely related to the aerodynamic characteristics of the vehicle at high speeds. Accurately obtaining these two parameters is crucial for achieving efficient vehicle energy management.

[0003] In existing technologies, vehicle mass is typically obtained through static weighing or empirical estimation based on historical data. However, the passenger capacity of buses exhibits a large dynamic range, with significant differences in mass between empty and fully loaded states. Traditional methods struggle to accurately obtain the actual vehicle mass in real time. Furthermore, the determination of the aerodynamic drag coefficient usually relies on fixed parameters obtained from wind tunnel or bench tests, failing to reflect dynamic changes caused by factors such as vehicle body dirt, externally mounted items, or high-speed crosswinds during actual operation. More critically, in traditional parameter estimation methods, vehicle mass and drag coefficient are often severely coupled, requiring the simultaneous estimation of multiple unknown parameters. This not only increases the computational complexity of the algorithm but also leads to long convergence times and low estimation accuracy.

[0004] Furthermore, existing methods for estimating driving resistance typically use a fixed rolling resistance coefficient for verification. However, in actual road driving, gradient factors can significantly interfere with resistance calculations, causing verification methods based on a single rolling resistance coefficient to fail to accurately reflect the actual driving resistance state of the vehicle. Simultaneously, during long-term vehicle operation, sensor drift and accumulated errors gradually affect the reliability of the estimation results, and existing technologies lack effective online verification and adaptive adjustment mechanisms. For commercial vehicles like buses, frequent passenger boarding and alighting mean a significant change in the overall vehicle mass with each door opening, making it difficult for existing technologies to quickly respond to such dynamic mass change scenarios.

[0005] In view of the above, this application is hereby submitted. Summary of the Invention

[0006] This invention aims to provide a method, device, equipment, and medium for jointly estimating the overall mass and aerodynamic drag of a bus, in order to solve the problems of existing bus mass and aerodynamic drag estimation techniques, which mostly rely on fixed drag coefficient lookup tables or separate identification of mass and drag, resulting in serious parameter coupling, inaccurate low-speed mass identification, inability to dynamically correct high-speed drag, inability to adapt to sudden changes in passenger load, and easy drift during long-term operation.

[0007] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0008] A method for jointly estimating the overall mass and aerodynamic drag of a passenger bus includes: S1: Real-time data collection of current vehicle speed, acquisition of standard frontal area and initial drag coefficient of bus, and calculation of air resistance baseline value; S2, when the current vehicle speed is detected to be lower than the set first threshold, the air resistance term is ignored, and the vehicle mass is estimated based on the simplified rolling resistance model and the recursive least squares method; S3, when the current vehicle speed is detected to be higher than the second threshold, the estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant, the initial drag coefficient and the air resistance reference value are used as the initial reference for iteration, and the real-time drag coefficient under the current working condition is calculated by recursive least squares method. S4 generates a theoretical resistance envelope under the current operating conditions based on the estimated vehicle mass, performs online rationality verification by combining the measured resistance, the air resistance benchmark value and the real-time drag coefficient, and triggers the observer to reset when the deviation exceeds the limit.

[0009] Preferably, it also includes: S5, real-time acquisition of door signals, when any door of the bus is detected to switch from open to closed, re-execute steps S2-S4 to update the vehicle mass and dynamic air resistance under the current operating conditions.

[0010] The present invention also provides a device for jointly estimating the total mass and aerodynamic drag of a bus, comprising: The semi-physical initialization unit is used to collect the current vehicle speed in real time, obtain the standard frontal area and initial drag coefficient of the bus, and calculate the air resistance benchmark value. The low-speed mass locking unit is used to estimate the vehicle mass based on a simplified rolling resistance model and recursive least squares method when the current vehicle speed is detected to be lower than a set first threshold, ignoring the air resistance term. The high-speed drag calculation unit is used to calculate the real-time drag coefficient under the current operating condition when the current vehicle speed is detected to be higher than the second threshold. The estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant. The initial drag coefficient and the air resistance reference value are used as the initial reference for iteration. The recursive least squares method is used to calculate the real-time drag coefficient under the current operating condition. The closed-loop verification unit is used to generate the theoretical resistance envelope under the current operating conditions based on the estimated vehicle mass, and to perform online rationality verification by combining the measured resistance, the air resistance benchmark value and the real-time drag coefficient, and to trigger the observer to reset when the deviation exceeds the limit.

[0011] Preferably, it further includes: a door trigger re-evaluation unit, used to collect door signals in real time, and when any door of the bus is detected to switch from the open state to the closed state, steps S2-S4 are re-executed to update the vehicle mass and dynamic air resistance under the current operating conditions.

[0012] The present invention also provides a device for jointly estimating the overall mass and aerodynamic drag of a bus, including a processor and a memory. The memory stores a computer program that can be executed by the processor to realize the method for jointly estimating the overall mass and aerodynamic drag of a bus as described above.

[0013] In summary, this invention adopts a general technical approach of using physical models to lock mass at low speeds, using observers to solve wind resistance at high speeds, and using semi-physical parameters throughout the entire process to reduce computational power. By introducing the national standard drag coefficient as a semi-physical reference envelope and combining it with a phased dynamic decoupling strategy, it achieves independent and accurate estimation of the vehicle mass and drag coefficient, and can automatically update the mass estimation results based on the door opening and closing status, thereby meeting the dynamic parameter estimation needs in actual bus operation. Compared with the prior art, this invention has the following beneficial effects: First, the standard windward area of ​​the original factory-calibrated bus is used as a known parameter to construct a semi-physical benchmark model, which significantly shortens the algorithm convergence time.

[0014] Secondly, by employing a segmented strategy of low-speed mass locking and high-speed drag decoupling, complete decoupling of mass and aerodynamic drag is achieved, effectively avoiding algorithm divergence caused by multi-parameter coupling and significantly improving the stability of vehicle mass and drag coefficient estimation. Under low-speed conditions, drag interference is effectively eliminated, ensuring the accuracy of mass estimation; under high-speed conditions, mass parameters are substituted into the equations as known quantities, improving the accuracy of drag coefficient calculation.

[0015] Third, through the door opening and closing trigger-based reassessment strategy, the system can respond in a timely manner to sudden changes in the overall vehicle mass caused by passengers getting on and off the bus, and achieve rapid adaptive updates of mass and aerodynamic drag parameters under load change conditions, accurately adapting to the specific operating conditions of the bus.

[0016] Fourth, by introducing online verification logic for national standard benchmark coefficients, the system can effectively suppress the impact of slope interference and sensor drift on the estimation results, thereby improving the estimation reliability during long-term operation.

[0017] Fifth, the semi-physical benchmark model replaces the traditional single rolling resistance coefficient verification method, which greatly reduces the computational complexity of the algorithm and makes the present invention easy to deploy on an automotive embedded platform. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of a method for jointly estimating the total mass and aerodynamic drag of a passenger vehicle, as provided in Example 1.

[0020] Figure 2 This is a schematic diagram of a combined estimation device for the overall mass and aerodynamic drag of a passenger vehicle provided in Embodiment 2.

[0021] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0023] Example 1 Embodiment 1 of the present invention provides a method for jointly estimating the mass and aerodynamic drag of a bus, which can be implemented by a device for jointly estimating the mass and aerodynamic drag of a bus (hereinafter referred to as the estimation device), and in particular, executed by one or more processors within the estimation device.

[0024] In this embodiment, the estimation device may be an electronic device equipped with a processor, which carries a computer program for the joint estimation method of the bus mass and aerodynamic drag, and the computer program can be executed, such as a bus, a workstation, etc., which are not limited here.

[0025] This invention proposes a joint estimation method for bus mass and aerodynamic drag based on semi-physical fusion and dynamic decoupling. By employing layered identification of operating conditions, step-by-step decoupling of parameters, benchmark constraint correction, online residual verification, and event-triggered re-estimation mechanism, this method addresses the shortcomings of existing bus mass and aerodynamic drag estimation techniques, which often rely on fixed drag coefficient lookup tables or separate identification of mass and drag. These techniques suffer from severe parameter coupling, inaccurate low-speed mass identification, inability to dynamically correct high-speed drag, inability to adapt to sudden changes in passenger load, and susceptibility to drift during long-term operation.

[0026] like Figure 1 As shown, a method for jointly estimating the overall mass and aerodynamic drag of a bus includes steps S1 to S5.

[0027] S1 collects the current vehicle speed in real time, obtains the standard frontal area and initial drag coefficient of the bus, and calculates the benchmark value of air resistance.

[0028] This step involves real-time acquisition of the bus's current speed, retrieving the bus's inherent standard frontal area and initial drag coefficient calibration parameters, and combining this with aerodynamic formulas to calculate the baseline values ​​of air resistance at different speeds. The calculation formula is as follows: ; in, This is the baseline value for air resistance; Standard air density; The standard frontal area for passenger vehicles; This is the initial reference drag coefficient; This refers to the vehicle's real-time speed.

[0029] The standard frontal area parameters, initial drag coefficient parameters, and the standard running resistance reference coefficients from GB / T 27840-2021 "Limits for Fuel Consumption of Heavy Commercial Vehicles" used in subsequent steps for buses can be stored in the non-volatile memory of the vehicle controller for easy retrieval. These standard running resistance reference coefficients include coefficients related to vehicle running resistance. , , Different types of buses correspond to different baseline coefficient values. For example, for city buses, The value range is usually between 0.8 and 1.2. The value ranges from 0.01 to 0.03. The value ranges from 0.0003 to 0.0005.

[0030] For example, taking a 12-meter-long city bus as an example, its standard frontal area Approximately 7.5 square meters, initial drag coefficient Approximately 0.65, these parameters are determined by the vehicle manufacturer during type approval testing. Since the bus body structure is relatively fixed, its frontal area varies little during actual use, and therefore can be treated as a known fixed value.

[0031] The standard frontal area, initial drag coefficient, and air resistance baseline values ​​obtained in this step serve as the baseline parameters for the vehicle's aerodynamic drag, providing a reference for subsequent S3 real-time drag calculation and S4 drag deviation verification.

[0032] In subsequent iterations of the recursive least squares algorithm, the aforementioned air resistance baseline model is used as a fixed input to the observer, rather than being treated as a parameter to be estimated. Since the frontal area parameter of the bus is a known fixed value, the algorithm does not need to iteratively estimate this parameter; it only needs to calculate the deviation between the actual resistance and the baseline resistance, thereby significantly reducing the iterative complexity of the algorithm and shortening the convergence time.

[0033] S2, when the current vehicle speed is detected to be lower than the set first threshold, ignore the air resistance term and estimate the vehicle mass based on the simplified rolling resistance model and the recursive least squares method.

[0034] The operating condition is determined based on the real-time vehicle speed collected by S1. When the vehicle speed is lower than a preset first threshold (e.g., 30 km / h), it is judged as a low-speed operating condition. At this time, the actual air resistance and the air resistance benchmark value obtained by S1 are much smaller than the rolling resistance, so the air resistance term is ignored. Based on the simplified longitudinal rolling resistance model of the vehicle, the recursive least squares method is used to identify the vehicle mass online and estimate the vehicle mass under the current passenger-carrying operating condition.

[0035] The core idea of ​​the recursive least squares algorithm is to minimize the sum of squared residuals between the observed values ​​and the model predictions by continuously updating the parameter estimates.

[0036] Under low-speed conditions, the simplified rolling resistance model is as follows: ; in, for The total force at the wheel end at any given moment is calculated by converting the motor drive torque / braking pressure. for Overall vehicle quality at all times; It is the acceleration due to gravity; This is the rolling resistance coefficient, typically taken as an empirical value between 0.006 and 0.008. This represents the current discrete sampling time.

[0037] Then, when estimating the vehicle mass using the recursive least squares method, a least squares method with a forgetting factor is used for calculation. After stable convergence, the estimated vehicle mass is obtained. The estimation process is as follows: make The observed output is the regression input. The parameters to be identified are The simplified rolling resistance model is then converted into a standard linear identification form using the least squares method. Calculate the iterative gain: ; Update vehicle weight: ; Update the covariance matrix: ; in, for The iterative gain of the least squares method at each time step is used to control the magnitude of parameter correction. This is a forgetting factor used to reduce the weight of old data; for Time-regression input vector; for The time-varying covariance matrix.

[0038] The forgetting factor is typically between 0.95 and 0.99, and the specific value needs to be adjusted according to the actual application scenario.

[0039] Taking a practical application scenario as an example, when a bus is traveling at a speed of 25 kilometers per hour on urban roads, it obtains the current driving force or braking force signal from the CAN bus, and at the same time obtains the current vehicle speed. At this speed range, since air resistance contributes very little to the total resistance, the recursive least squares algorithm performs extremely stably and is not affected by external interference factors such as crosswinds and uneven load distribution.

[0040] Experiments show that the algorithm can converge and obtain an accurate estimate of the vehicle mass within a time window of 3 to 5 seconds.

[0041] The vehicle mass estimated in real time in this step provides a known constant for solving the wind resistance in step S3 and generating the resistance envelope in step S4.

[0042] S3, when the current vehicle speed is detected to be higher than the second threshold, the estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant, the initial drag coefficient and the air resistance reference value are used as the initial reference for iteration, and the real-time drag coefficient under the current working condition is calculated by recursive least squares method.

[0043] The operating condition is determined based on the real-time vehicle speed collected by S1. When the vehicle speed exceeds a preset second threshold, it is considered a high-speed operating condition, and air resistance cannot be ignored. The vehicle mass estimated by S2 is used as a known constant. At the same time, the standard frontal area and air resistance benchmark framework of S1 are introduced to construct the longitudinal dynamic equation of the vehicle.

[0044] Using the initial drag coefficient S1 and the air resistance benchmark value as the initial benchmark for iteration, the deviation is identified online using the recursive least squares method, and the real-time drag coefficient that fits the current vehicle body state and ambient wind speed is calculated.

[0045] Specifically, the process of calculating the real-time drag coefficient using the recursive least squares method is as follows: First, the estimated vehicle mass is substituted as a known constant into the longitudinal dynamics equation, and its expression is: ; in, This represents the current discrete sampling time. for The equivalent driving force of the entire vehicle drive unit at all times; For the estimated total vehicle weight; for The real-time longitudinal acceleration of the vehicle is calculated from the rate of change of vehicle speed. Standard air density; The standard frontal area for passenger vehicles; for Real-time drag coefficient at any given moment; for Real-time vehicle speed.

[0046] Then let To observe the output quantity, The regression input is the parameter to be identified. The least squares method with a forgetting factor is used for calculation. After stable convergence, the real-time drag coefficient under the current operating conditions is obtained. This allows us to obtain the real-time aerodynamic drag (real-time dynamic air drag) under the current operating conditions.

[0047] This step uses the S1 air resistance benchmark as the initial constraint for iteration to avoid divergence in wind resistance identification and achieve dynamic and accurate correction of aerodynamic drag.

[0048] This technical solution enables the system to detect changes in aerodynamic performance in real time due to factors such as the addition of roof racks, dirt on the vehicle body surface, or high-speed crosswinds, and outputs a high-precision estimate of the drag coefficient.

[0049] Taking a typical high-speed driving condition as an example, when a bus is traveling at 80 km / h on a highway, the current driving force signal is obtained from the CAN bus, along with the current vehicle speed and acceleration. Then, the current drag coefficient is calculated using the aforementioned equation. Assuming that there is some dirt on the vehicle body surface, causing the drag coefficient to increase from the initial 0.65 to 0.68, this step can accurately capture this change and output the updated drag coefficient to the vehicle controller for optimizing energy management and power distribution strategies.

[0050] S4 generates a theoretical resistance envelope under the current operating conditions based on the estimated vehicle mass, performs online rationality verification by combining the measured resistance, the air resistance benchmark value and the real-time drag coefficient, and triggers the observer to reset when the deviation exceeds the limit.

[0051] This step utilizes the vehicle mass output from S2 to generate a theoretical drag envelope curve corresponding to the current vehicle speed and load conditions, matching the vehicle's baseline drag parameters. Further, it combines the air resistance baseline value from S1 and the updated real-time drag coefficient from S3 to construct a composite drag verification model that includes rolling resistance and dynamic aerodynamic drag. The measured drag calculated in real-time is compared with the theoretical drag envelope, simultaneously verifying the reasonableness of the deviation of the real-time drag coefficient relative to the S1 baseline drag, thus achieving joint online verification of vehicle mass and aerodynamic drag. When the continuous sampling deviation exceeds a preset threshold, it is determined that the recursive least squares observer has accumulated error, triggering an observer reset to ensure the accuracy of subsequent vehicle mass and drag coefficient estimations.

[0052] Specifically, when the vehicle is in a coasting condition, the national standard reference constant term is adjusted using the currently estimated total vehicle mass. Perform quality scaling, one item Quadratic terms The theoretical resistance envelope under the current operating condition is calculated without scaling, as it is weakly correlated with mass. Its expression is: ; in, Current vehicle speed Theoretical resistance below; This refers to the vehicle's maximum design gross weight. For the estimated total vehicle weight; For national standard reference constants, For national standard primary items, This is the second term of the national standard benchmark.

[0053] It should be noted that a single term and quadratic terms Aerodynamic and mechanical properties, typically considered weakly correlated with mass, are not subject to mass scaling in this invention. Taking a city bus with a maximum design gross weight of 18 tons as an example, when estimating the mass... When the constant term is 15 tons, Multiply by the ratio of 18 to 15 to scale, generating the theoretical resistance envelope under the current operating conditions.

[0054] The measured resistance is calculated based on the estimated vehicle mass and real-time longitudinal acceleration, and the expression is: ; in, This is the measured resistance. This refers to the vehicle's real-time longitudinal acceleration.

[0055] Next, the theoretical resistance envelope is calculated. Compared with the measured resistance The standardized residuals, and the formula for calculating the standardized residuals: ; in, The standardized residual is used to characterize the relative deviation between the measured resistance and the theoretical resistance.

[0056] Verify real-time drag coefficient Relative air resistance reference value Reasonableness of the deviation: If the standardized residuals are sampled continuously over multiple periods If the error exceeds the set threshold (e.g., 10%), it is determined that the recursive least squares observer has accumulated error, triggering the observer reset.

[0057] Furthermore, when the system detects that the vehicle is in a low-speed coasting condition (i.e., the throttle opening is zero and the vehicle speed is decreasing), it will use the currently estimated mass. The rolling resistance coefficient is calculated by reverse calculation. Specifically, this is done in conjunction with the air resistance baseline value. With real-time drag coefficient Compare the real-time dynamic aerodynamic drag with the baseline value of air resistance at the same vehicle speed. The difference is used to determine whether the current real-time dynamic aerodynamic drag deviates from a reasonable range. If the deviation exceeds 5%, the current mass estimate is considered to have a large error, and then... Fine-tune and reset the observer covariance matrix to bring the system back into the quality locking phase.

[0058] The expression for the current real-time dynamic aerodynamic drag is: ; in, For real-time dynamic aerodynamic drag; Standard air density; The standard frontal area for passenger vehicles; This refers to the vehicle's real-time speed.

[0059] In addition, it also includes: S5, real-time acquisition of door signals, when any door of the bus is detected to switch from open to closed, steps S2-S4 are re-executed to update the vehicle mass and dynamic air resistance under the current operating conditions.

[0060] This step identifies vehicle load changes caused by passenger boarding and alighting by real-time acquisition of the opening and closing status signals of each bus door. When any door is detected to switch from open to closed, it is determined that the passenger boarding or alighting action has been completed, and the vehicle will enter a new passenger-carrying condition. The overall vehicle mass changes, and the original mass parameters become invalid. The low-speed mass locking phase of S2 is immediately restarted, and the current overall vehicle mass is re-identified using the recursive least squares method. The overall vehicle mass under the current passenger-carrying condition is updated, and the basic parameters for subsequent real-time drag coefficient calculation in S3 and drag joint verification in S4 are updated, realizing the dynamic joint update and estimation of overall vehicle mass and aerodynamic drag.

[0061] This technical solution is designed to fully consider the frequent changes in passenger capacity of buses. Taking city buses as an example, passengers frequently get on and off at stops, and the number of passengers in the carriage may change each time the doors close, resulting in a change in the overall vehicle mass. Traditional mass estimation methods cannot respond to such changes in a timely manner, while this invention, through a door-triggered re-estimation strategy, can complete the mass re-identification shortly after the doors close, enabling the system to update the overall vehicle mass estimate in a timely manner.

[0062] In practice, the door signal acquisition module is connected to the door switch signal line of the vehicle's electrical system, including the opening and closing states of the front, middle, and rear doors. When any door changes from open to closed, the system immediately switches to a low-speed mass lock mode, re-estimating the vehicle's mass using currently available observation data. This re-estimation process typically completes within 5 to 10 seconds, providing accurate mass parameters for subsequent high-speed drag calculations.

[0063] During the overall system operation, the estimation task is executed cyclically according to the logical sequence of the five stages mentioned above. When the vehicle is first started, the system reads pre-stored parameters and builds a baseline model. Then, based on the current vehicle speed, it determines whether to enter the low-speed mass locking stage or the high-speed wind resistance calculation stage. While executing stage S3, the system executes stage S4 (closed-loop verification) in parallel, continuously monitoring the estimation results. When a change in the door's open / closed state is detected, the system prioritizes executing stage S5 (door-triggered re-estimation), completing the mass update before continuing the normal estimation process.

[0064] The technical solution of this invention also considers strategies for handling various practical operating scenarios. In the slope driving scenario, since the slope causes changes in the gravity component in the longitudinal dynamic equation, the residual detection mechanism in the S4 closed-loop verification stage can identify slope interference and trigger observer reset, avoiding slope contamination of the estimation results. In the high-speed crosswind scenario, the S3 high-speed drag calculation stage can calculate the changed drag coefficient in real time, providing accurate input parameters for the vehicle control strategy. In the long-term operation scenario, the adaptive forgetting mechanism periodically resets the observer to prevent error accumulation from causing a decrease in estimation accuracy.

[0065] In summary, this invention achieves joint estimation of the overall mass and aerodynamic drag of a bus through five core steps: semi-physical initialization, low-speed mass locking, high-speed wind resistance calculation, closed-loop verification, and door-triggered reestimation. Compared with existing technologies, this invention has the following advantages: (1) Achieve parameter decoupling and identification, completely solve the problem of multi-parameter coupling and divergence, and improve estimation stability.

[0066] Existing technologies often simultaneously identify vehicle mass and drag coefficient. The coupling of multiple unknown parameters leads to oscillations, slow convergence, and unstable identification results in the least squares (RLS) observer. This invention addresses this by implementing layered control under high and low speed conditions: in low-speed conditions, aerodynamic drag is ignored based on the S1 air resistance baseline characteristics, and only the vehicle mass parameter is identified; in high-speed conditions, the converged vehicle mass is solidified as a known quantity, and only the real-time drag coefficient is identified. This single-parameter identification under different conditions achieves complete decoupling of vehicle mass and aerodynamic drag, effectively avoiding algorithm divergence caused by multi-parameter coupling and significantly improving the stability of vehicle mass and drag coefficient estimation.

[0067] (2) Based on the air resistance benchmark constraint, the aerodynamic drag is dynamically and adaptively corrected to fit the actual driving conditions.

[0068] (3) Construct a mass-scaled resistance envelope and composite verification model to achieve dual-parameter joint verification and suppress algorithm cumulative error.

[0069] Existing technologies lack online verification and error correction mechanisms. After long-term vehicle operation, factors such as slope, sensor drift, and road surface interference cause errors in mass and wind resistance estimation to accumulate. This invention innovatively employs a theoretical resistance envelope calculation method based on real-time vehicle mass scaling of the national standard resistance benchmark, closely aligning with the dynamic characteristics of bus load changes. Simultaneously, it integrates rolling resistance and dynamic aerodynamic resistance to construct a composite resistance verification model. By combining measured resistance, standardized theoretical resistance residual comparison, and dual verification of real-time wind resistance deviation from the benchmark wind resistance, it can accurately identify accumulated observer errors and automatically trigger a reset, achieving closed-loop algorithm correction and significantly improving the accuracy of full lifecycle estimation.

[0070] (4) Based on the door status event triggering re-evaluation, it is adapted to the frequent passenger-carrying change working conditions of buses, and has stronger working condition adaptability.

[0071] Traditional estimation methods often involve periodic updates or continuous iterations, which cannot adapt to the operational scenarios of urban buses with frequent passenger pick-up and drop-off and sudden changes in vehicle weight. When passenger load changes, the original weight parameters become invalid, leading to continuous distortion in drag estimation. This invention monitors door opening and closing events in real time. Upon door closure, it immediately restarts vehicle weight identification, rapidly updates the vehicle weight benchmark under the current passenger load condition, and simultaneously refreshes the basic parameters for calculating the drag coefficient. This enables rapid adaptive updates of weight and aerodynamic drag parameters under sudden load changes, accurately adapting to the specific operating conditions of urban buses.

[0072] (5) The hierarchical closed-loop logic is clear, the convergence speed is fast, the robustness is high, and the engineering practicality is strong.

[0073] This solution significantly shortens the convergence time of the RLS algorithm by relying on baseline parameter constraints, while ensuring the algorithm's robustness under complex road surfaces, complex wind fields, and variable load conditions through a residual verification mechanism. The entire method relies solely on conventional onboard sensor signals, requiring no additional hardware equipment. It is low-cost, highly portable, and can be directly applied to engineering scenarios such as bus power control, energy consumption estimation, and coasting condition optimization, demonstrating extremely high practical application value.

[0074] (6) Achieve true joint estimation of vehicle mass and aerodynamic drag, breaking through the technical barrier of traditional independent estimation.

[0075] Existing technologies involve independent mass estimation and wind resistance estimation, lacking parameter linkage verification and failing to form a data closed loop. This invention achieves deep parameter reuse throughout the entire process: S1 baseline parameters support wind resistance calculation; S2 mass parameters support the drag envelope and wind resistance solution; S3 real-time wind resistance participates in drag rationality verification; S4 correction results feed back into the identification algorithm; S5 operating conditions dynamically update basic parameters. This truly realizes joint identification, linkage verification, and collaborative correction of mass and aerodynamic drag, perfectly aligning with the invention's theme, and achieving estimation accuracy far superior to traditional independent estimation schemes.

[0076] Example 2 like Figure 2 As shown, the second embodiment of the present invention also provides a device for jointly estimating the overall mass and aerodynamic drag of a bus, comprising: The semi-physical initialization unit is used to collect the current vehicle speed in real time, obtain the standard frontal area and initial drag coefficient of the bus, and calculate the air resistance benchmark value. The low-speed mass locking unit is used to estimate the vehicle mass based on a simplified rolling resistance model and recursive least squares method when the current vehicle speed is detected to be lower than a set first threshold, ignoring the air resistance term. The high-speed drag calculation unit is used to calculate the real-time drag coefficient under the current operating condition when the current vehicle speed is detected to be higher than the second threshold. The estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant. The initial drag coefficient and the air resistance reference value are used as the initial reference for iteration. The recursive least squares method is used to calculate the real-time drag coefficient under the current operating condition. The closed-loop verification unit is used to generate the theoretical resistance envelope under the current operating conditions based on the estimated vehicle mass, and to perform online rationality verification by combining the measured resistance, the air resistance benchmark value and the real-time drag coefficient, and to trigger the observer to reset when the deviation exceeds the limit.

[0077] Preferably, it further includes: a door trigger re-evaluation unit, used to collect door signals in real time, and when any door of the bus is detected to switch from the open state to the closed state, steps S2-S4 are re-executed to update the vehicle mass and dynamic air resistance under the current operating conditions.

[0078] Example 3 The third embodiment of the present invention also provides a device for jointly estimating the overall mass and aerodynamic drag of a bus, which includes a memory and a processor. The memory stores a computer program, which can be executed by the processor to realize the method for jointly estimating the overall mass and aerodynamic drag of a bus as described above.

[0079] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle, characterized in that, include: S1: Real-time data collection of current vehicle speed, acquisition of standard frontal area and initial drag coefficient of bus, and calculation of air resistance baseline value; S2, when the current vehicle speed is detected to be lower than the set first threshold, the air resistance term is ignored, and the vehicle mass is estimated based on the simplified rolling resistance model and the recursive least squares method; S3, when the current vehicle speed is detected to be higher than the second threshold, the estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant, the initial drag coefficient and the air resistance reference value are used as the initial reference for iteration, and the real-time drag coefficient under the current working condition is calculated by recursive least squares method. S4. Based on the estimated vehicle mass, a theoretical drag envelope under the current operating conditions is generated. This envelope is then combined with the measured drag, the aforementioned air resistance benchmark value, and the real-time drag coefficient for online rationality verification. If the deviation exceeds the limit, the observer is reset. The theoretical drag envelope utilizes the currently estimated vehicle mass against the national standard benchmark constant term. Perform quality scaling, one item Quadratic terms It is weakly correlated with quality and is calculated without scaling; its expression is: ； in, Current vehicle speed Theoretical resistance below; This refers to the vehicle's maximum design gross weight. For the estimated total vehicle weight; For national standard reference constants, For national standard primary items, This is the second term of the national standard benchmark; The measured resistance is calculated based on the estimated vehicle mass and real-time longitudinal acceleration, and the expression is: ； in, This is the measured resistance. For the vehicle's real-time longitudinal acceleration; Calculate the theoretical resistance envelope Compared with the measured resistance The standardized residuals, and the formula for calculating the standardized residuals: ； in, The standardized residual is used to characterize the relative deviation between the measured resistance and the theoretical resistance. Verify real-time drag coefficient Relative air resistance reference value Reasonableness of the deviation: If the standardized residuals are sampled continuously over multiple periods If the threshold is exceeded, it is determined that the recursive least squares observer has accumulated error, triggering the observer reset.

2. The method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle according to claim 1, characterized in that... It also includes: S5, real-time acquisition of door signals. When any door of the bus is detected to switch from open to closed, steps S2-S4 are re-executed to update the vehicle mass and dynamic air resistance under the current operating conditions.

3. The method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle according to claim 1, characterized in that... The air resistance benchmark value is obtained by real-time acquisition of the bus's current speed, retrieving the bus's inherent standard frontal area and initial drag coefficient calibration parameters, and combining them with aerodynamic formulas to calculate the air resistance benchmark value at different vehicle speeds. The calculation formula is as follows: ； in, This is the baseline value for air resistance; Standard air density; The standard frontal area for passenger vehicles; This is the initial reference drag coefficient; This refers to the vehicle's real-time speed.

4. The method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle according to claim 1, characterized in that... The simplified rolling resistance model is as follows: ； in, for The total force at the wheel end at any given moment is calculated by converting the motor drive torque / braking pressure. for Overall vehicle quality at all times; It is the acceleration due to gravity; This is the rolling resistance coefficient; This represents the current discrete sampling time. When estimating the vehicle mass using recursive least squares, a recursive least squares method with a forgetting factor is used for calculation. After stable convergence, the estimated vehicle mass is obtained. ; The estimation process using recursive least squares with a forgetting factor is as follows: make The observed output is the regression input. The parameters to be identified are The simplified rolling resistance model is then converted into a standard linear identification form using the least squares method. Calculate the iterative gain: ; Update vehicle weight: ; Update the covariance matrix: ; in, for The iterative gain of the least squares method at each time step is used to control the magnitude of parameter correction. This is a forgetting factor used to reduce the weight of old data; for Time-regression input vector; for The time-varying covariance matrix.

5. The method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle according to claim 4, characterized in that... The process of calculating the real-time drag coefficient using the recursive least squares method is as follows: First, the estimated vehicle mass is substituted as a known constant into the longitudinal dynamics equation, and its expression is: ； in, This represents the current discrete sampling time. for The equivalent driving force of the entire vehicle drive unit at all times; for Real-time longitudinal acceleration of the vehicle; Standard air density; The standard frontal area for passenger vehicles; for Real-time drag coefficient at any given moment; for Real-time vehicle speed; Then, let To observe the output quantity, The regression input is the parameter to be identified. The least squares method with a forgetting factor is used for calculation. After stable convergence, the real-time drag coefficient under the current operating conditions is obtained. .

6. The method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle according to claim 5, characterized in that... It also includes: combining air resistance benchmark values With real-time drag coefficient Compare the real-time dynamic aerodynamic drag with the baseline value of air resistance at the same vehicle speed. The difference is used to determine whether the current real-time dynamic aerodynamic drag deviates from the preset range. If it deviates from the preset range, it is determined that the current mass estimate has a large error, and the mass estimation stage is re-entered. The expression for the current real-time dynamic aerodynamic drag is: ； in, For real-time dynamic aerodynamic drag; Standard air density; The standard frontal area for passenger vehicles; This refers to the vehicle's real-time speed.

7. A device for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle, used to implement the method for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle as described in any one of claims 1-6, characterized in that, include: The semi-physical initialization unit is used to collect the current vehicle speed in real time, obtain the standard frontal area and initial drag coefficient of the bus, and calculate the air resistance benchmark value. The low-speed mass locking unit is used to estimate the vehicle mass based on a simplified rolling resistance model and recursive least squares method when the current vehicle speed is detected to be lower than a set first threshold, ignoring the air resistance term. The high-speed drag calculation unit is used to calculate the real-time drag coefficient under the current operating condition when the current vehicle speed is detected to be higher than the second threshold. The estimated vehicle mass is substituted into the longitudinal dynamic equation as a known constant. The initial drag coefficient and the air resistance reference value are used as the initial reference for iteration. The recursive least squares method is used to calculate the real-time drag coefficient under the current operating condition. The closed-loop verification unit is used to generate the theoretical resistance envelope under the current operating conditions based on the estimated vehicle mass, and to perform online rationality verification by combining the measured resistance, the air resistance benchmark value and the real-time drag coefficient, and to trigger the observer to reset when the deviation exceeds the limit.

8. A device for jointly estimating the overall mass and aerodynamic drag of a passenger vehicle, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that can be executed by the processor to implement a method for jointly estimating the overall mass and aerodynamic drag of a bus as described in any one of claims 1-6.

Images