A Dynamic Compensation Algorithm and System for Urea Level in Trucks Based on Big Data

Abstract

This invention discloses a big data-based dynamic compensation algorithm and system for truck urea level, specifically relating to the field of dynamic compensation technology for truck urea level. By integrating vehicle acceleration, pitch angle, and internal component layout information, a working condition-liquid level distortion feature description set is constructed. The original liquid level signal is separated into high-frequency fluctuations and low-frequency trends, and the intensity of non-consumable liquid level distortion is extracted. Then, the dominant distortion type is distinguished and the offset direction and compensation coefficient are obtained. Combined with the liquid level statistical benchmark obtained from long-term static or stable driving, a weighted fusion correction is performed to achieve accurate restoration of the liquid level measurement value to the actual volume under all working conditions. This effectively eliminates violent fluctuations in the liquid level display and significantly improves the accuracy of urea consumption calculation and the reliability of low liquid level alarm.

Description

Technical Field

[0001] This invention relates to the field of dynamic compensation technology for truck urea level, and more specifically, to a dynamic compensation algorithm and system for truck urea level based on big data. Background Technology

[0002] In truck operation, urea tanks are often designed with irregular shapes due to space constraints in the chassis. Internally, they house components such as baffles, oil return pipes, heating water pipes, and urea pumps. When the vehicle accelerates rapidly, brakes, or turns, the liquid inside the tank sloshes violently. These internal components severely disrupt the flow field, inducing turbulence, reflected waves, and standing waves, causing abrupt local distortion of the liquid level at the sensor's mounting location. At this point, the sensor only measures the transient height at that point, no longer maintaining a fixed correspondence with the overall remaining volume of the tank. Even with the vehicle's attitude angle and acceleration, it's difficult to calculate the true liquid volume geometrically. This is because the physical obstruction of the internal components can cause sudden jumps in the liquid level or localized accumulation and abnormal rise of liquid on the liquid-facing side of the components, directly generating false signals that deviate from the actual volume. This nonlinear distortion of the liquid level caused by the internal structure makes it impossible for traditional filtering and calibration methods to effectively eliminate the sloshing artifacts, resulting in violent fluctuations in the liquid level display during driving, severely inaccurate consumption calculations, and frequent false alarms for low liquid levels. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a big data-based dynamic compensation algorithm and system for truck urea levels to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A dynamic compensation algorithm for urea level in trucks based on big data includes the following steps: The original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle, and the pitch angle are obtained. Combined with the layout information of the internal components of the tank, the working condition-liquid surface distortion feature description set is obtained. Based on the working condition-liquid surface distortion feature description set, the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal are separated, the abnormal liquid surface jumps and accumulation features are extracted, and the non-consumable liquid surface distortion intensity under the current working condition is calculated. Based on the non-consumable liquid surface distortion intensity and the vehicle acceleration change rate, the dominant type of liquid surface distortion is determined, and combined with the urea tank temperature and the current liquid volume range, the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume are obtained. The original liquid level signal is dynamically corrected according to the compensation coefficient and fused with the liquid level statistical benchmark obtained by the vehicle under at least one state of long-term stationary and stable driving to output the compensated real-time liquid level value.

[0005] In a preferred embodiment, the acquisition of the original urea tank level signal, vehicle longitudinal and lateral acceleration, and pitch angle, combined with the internal component layout information of the tank, to obtain the working condition-liquid surface distortion feature description set specifically involves: simultaneously acquiring the original urea tank level signal, longitudinal acceleration, lateral acceleration, and pitch angle, and acquiring the internal component layout information of the tank; selecting segments of vehicles that have been stationary for extended periods and traveling at a constant speed in a straight line from historical driving data, taking the mean of the original level signal as the steady-state level reference value, and obtaining the steady-state level sequence through interpolation. The liquid level distortion under unsteady operating conditions is calculated; the total volume of the urea tank is discretized into multiple liquid volume intervals, a synthetic acceleration amplitude is defined, and the numerical range of the synthetic acceleration amplitude and the numerical range of the pitch angle are equally divided into several intervals to form a two-dimensional operating condition grid cell; for each liquid volume interval, the component influence factor is determined using the component layout information; sample points are extracted from the collected historical data, and the positive accumulation probability, negative exposure probability, average distortion amplitude, and abnormal jump frequency are statistically analyzed to obtain the operating condition-liquid surface distortion feature description set.

[0006] In a preferred embodiment, the step of separating the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal according to the working condition-liquid surface distortion feature description set, extracting abnormal liquid level jumps and accumulation features, and calculating the non-consumable liquid level distortion intensity under the current working condition specifically involves: performing adaptive signal separation on the original liquid level signal to obtain intrinsic mode function components and residual trend terms arranged by frequency; reconstructing the high-frequency components into high-frequency fluctuation components of the liquid level, and merging and reconstructing the remaining components into low-frequency change trends of the liquid level; detecting abnormal jump events in the high-frequency fluctuation components of the liquid level, and detecting accumulation events in the low-frequency change trends of the liquid level; and calculating the non-consumable liquid level distortion intensity by combining the working condition-liquid surface distortion feature description set and component influence factors under the current working condition.

[0007] In a preferred embodiment, the non-consumable liquid surface distortion intensity is weighted and synthesized from the instantaneous energy under an abnormal jump event, the low-frequency offset under an accumulation event, the positive accumulation probability, the negative exposure probability, the average distortion amplitude, the frequency of the abnormal jump, and the component influence factor.

[0008] In a preferred embodiment, determining the dominant type of liquid surface distortion based on the non-consumable liquid surface distortion intensity and the vehicle acceleration change rate specifically involves: calculating the composite acceleration change rate; when the non-consumable liquid surface distortion intensity is greater than or equal to the distortion intensity determination threshold and the composite acceleration change rate is greater than or equal to the acceleration change rate determination threshold, it is determined to be an impact-jump type; when the non-consumable liquid surface distortion intensity is greater than or equal to the distortion intensity determination threshold and the composite acceleration change rate is less than the acceleration change rate determination threshold, it is determined to be a continuous-accumulation type; when the non-consumable liquid surface distortion intensity is less than the distortion intensity determination threshold, it is determined that the current liquid surface distortion degree is relatively mild, and no dominant type distinction is made.

[0009] In a preferred embodiment, the offset direction of the sensor measurement value relative to the actual volume is obtained as follows: when the dominant type is impact-jump type, the offset direction is determined according to the relative relationship between the instantaneous vector direction of the composite acceleration and the sensor installation position; when the dominant type is continuous-accumulation type, the offset direction is determined according to the sign of the pitch angle and the continuous direction of the lateral acceleration; when the liquid surface distortion is negligible, the offset direction is zero.

[0010] In a preferred embodiment, the original liquid level signal is dynamically corrected according to the compensation coefficient and fused with a liquid level statistical benchmark obtained under at least one of the vehicle's long-term stationary and stable driving conditions to output a compensated real-time liquid level value. Specifically, the following steps are taken: long-term stationary windows and stable driving windows are identified from the vehicle's historical driving data as liquid level statistical benchmark sampling windows; the arithmetic mean of the original liquid level signal is calculated for each of the liquid level statistical benchmark sampling windows as a benchmark liquid level value, and a liquid level statistical benchmark sequence is obtained by interpolation; the original liquid level signal is dynamically corrected according to the offset direction and the compensation coefficient to obtain a preliminary compensated liquid level; and a fusion weight is calculated to weight and fuse the preliminary compensated liquid level with the liquid level statistical benchmark sequence to output a compensated real-time liquid level value.

[0011] In a preferred embodiment, when the dominant type is impact-jump type, the compensation coefficient is determined based on the normalized results of the non-consumable liquid surface distortion intensity relative to its historical maximum value and the rate of change of the synthetic acceleration relative to its historical maximum value, combined with the component influence factor.

[0012] In a preferred embodiment, when the dominant type is a continuous-accumulation type, the compensation coefficient is determined based on the low-frequency offset, steady-state liquid level sequence value, and duration, combined with the component influence factor.

[0013] In a preferred embodiment, a truck urea level dynamic compensation system based on big data includes a distortion feature mapping module, a distortion intensity identification module, an offset and compensation decision module, and a dynamic compensation fusion module. The distortion feature mapping module is used to obtain the original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle and the pitch angle, and combine it with the internal component layout information of the tank to obtain the working condition-liquid surface distortion feature description set. The distortion intensity identification module is used to separate the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal according to the working condition-liquid surface distortion feature description set, extract the abnormal liquid surface jump and accumulation features, and calculate the non-consumable liquid surface distortion intensity under the current working condition. The offset and compensation decision module is used to determine the dominant type of liquid surface distortion based on the intensity of the non-consumable liquid surface distortion and the rate of change of vehicle acceleration, and to obtain the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume. The dynamic compensation fusion module is used to dynamically correct the original liquid level signal according to the compensation coefficient, and fuse it with the liquid level statistical reference obtained by the vehicle under at least one state of long-term stationary and stable driving, and output the compensated real-time liquid level value.

[0014] The technical effects and advantages of this invention are as follows: 1. This invention addresses the problem of nonlinear distortion of the liquid level caused by the irregular structure and internal components of a truck urea tank under dynamic operating conditions. By fusing vehicle acceleration, pitch angle, and internal component layout information, a set of operating condition-liquid level distortion feature descriptions is constructed. The original liquid level signal is separated into high-frequency fluctuations and low-frequency trends, and the intensity of non-consumable liquid level distortion is extracted. Then, the dominant distortion type is distinguished and the offset direction and compensation coefficient are obtained. Combined with the liquid level statistical benchmark obtained from long-term static or stable driving, a weighted fusion correction is performed to achieve accurate restoration of the liquid level measurement value to the actual volume under all operating conditions. This effectively eliminates violent fluctuations in the liquid level display and significantly improves the accuracy of urea consumption calculation and the reliability of low liquid level alarms. Attached Figure Description

[0015] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings; Figure 1 This is a flowchart of the algorithm in Embodiment 1 of the present invention; Figure 2 This is a module connection diagram of the system in Embodiment 2 of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1: Figure 1 This invention presents a dynamic compensation algorithm for urea level in trucks based on big data, comprising the following steps: The original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle, and the pitch angle are obtained. Combined with the layout information of the internal components of the tank, the mapping relationship between acceleration excitation and local liquid surface fluctuation characteristics under different liquid volume ranges is constructed to obtain the working condition-liquid surface distortion feature description set. Based on the working condition-liquid surface distortion feature description set, the high-frequency fluctuation component and low-frequency change trend of the original liquid level signal are separated, the abnormal liquid surface jump and accumulation features caused by obstruction and reflection of internal components are extracted, and the non-consumable liquid surface distortion intensity under the current working condition is calculated. Based on the non-consumable liquid surface distortion intensity and the vehicle acceleration change rate, the dominant type of liquid surface distortion is determined, and the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume are obtained. The original liquid level signal is dynamically corrected according to the compensation coefficient and fused with the liquid level statistical benchmark obtained by the vehicle under at least one state of long-term stationary and stable driving to output the compensated real-time liquid level value.

[0018] In this embodiment of the invention, the original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle, and the pitch angle are obtained. Combined with the layout information of the internal components of the tank, a mapping relationship between acceleration excitation and local liquid surface fluctuation characteristics under different liquid volume ranges is constructed to obtain the working condition-liquid surface distortion feature description set, specifically: The original liquid level signal is synchronously acquired via the vehicle's CAN bus and the T-BOX at a sampling frequency of no less than 1Hz. Longitudinal acceleration lateral acceleration and pitch angle The system obtains the layout information of the internal components of the enclosure, including at least the coordinates of the sensor installation positions and the top height of the partition. The bottom height, the routing of the heating water pipe and the urea pump suction oil pipe, and the space they occupy; The system selects segments from historical driving data showing vehicles stationary and traveling at a constant speed in a straight line for extended periods. The average value of the original liquid level signal within these segments is then used as the steady-state liquid level reference value. A time-continuous steady-state liquid level sequence was obtained through linear interpolation. Based on this, the liquid level distortion under unsteady conditions is calculated. .

[0019] The total volume of the urea tank Discretized into M liquid volume intervals ,in , Define the composite acceleration amplitude. The numerical range of the synthesized acceleration amplitude Divided into equal parts The interval defines the numerical range of the pitch angle. Divided into equal parts This forms a range, thus creating Each two-dimensional working condition grid cell corresponds to an acceleration range. With pitch angle range The combination of .

[0020] For each liquid volume range j, the corresponding component influence factor is determined using the component layout information. Based on the pre-calibrated liquid volume-liquid level function Calculate the midpoint of the liquid volume interval. Corresponding liquid level height ,when Critical height window falling near the top height of the partition At that time, take for The calibration value is between, otherwise take the value between. .

[0021] The liquid volume-liquid level function The specific form is obtained by pre-calibrating the static volume of the urea tank, and it is expressed as a piecewise linear interpolation function based on the calibration nodes: Total volume of the urea tank Divide into Q calibration nodes, with each node having a volume of [missing information]. , Add a measured amount of urea solution to the empty box one at a time, and record the volume at each point after the liquid level stabilizes. Corresponding actual liquid level height This constitutes the calibration data pair For any volume to be determined The liquid level is calculated by linear interpolation: ,in, and These are volume nodes. and The calibrated liquid level height at the location, This represents the slope of the liquid level change within this volume range. This function fully characterizes the nonlinear relationship between volume and liquid height in the urea tank caused by its irregular structure and internal components, providing a basis for subsequently determining whether the liquid level falls within the critical height window of the baffle.

[0022] From the collected historical data, extract all data that simultaneously satisfy the conditions of liquid volume being in interval j and the combined acceleration amplitude being in interval j. And the pitch angle is within the range The sample points are statistically analyzed according to the following method for each element in the described working condition-liquid surface distortion feature set: Forward stacking probability : Statistics of the sample points Number of , with the total number of samples The ratio of, where This is the preset positive distortion threshold; Negative exposure probability : Statistics of the sample points Number of ,and The ratio; Mean distortion amplitude Calculate all sample points The arithmetic mean; abnormal jump frequency : Calculate the first-order difference of liquid level distortion at adjacent sampling times ,statistics Exceeding the transition threshold Number of events Divide the result by the time span corresponding to the sample point to obtain the frequency of abnormal jumps; The positive distortion threshold With the jump threshold The volume measurement accuracy of the urea tank is determined through actual vehicle calibration based on the required accuracy.

[0023] This leads to the construction of a three-dimensional description set. This is the working condition-liquid surface distortion feature description set, which fully records the statistical laws of liquid surface distortion corresponding to different combinations of acceleration and attitude under various liquid volume states.

[0024] In this embodiment of the invention, based on the operating condition-liquid surface distortion feature description set, the high-frequency fluctuation component and low-frequency change trend of the original liquid level signal are separated, the abnormal liquid surface jumps and accumulation features caused by obstruction and reflection of internal components are extracted, and the non-consumable liquid surface distortion intensity under the current operating condition is calculated, specifically: The original liquid level signal Signal separation is performed using ensemble empirical mode decomposition, yielding K eigenmode function components. With a residual trend term ,in The intrinsic mode function components are arranged in descending order of frequency, and the residual trend term... It reflects the low-frequency variation trend of the original liquid level signal.

[0025] The high-frequency eigenmode function components of the liquid level are reconstructed by the first p high-frequency eigenmode function components. The reconstruction order p is determined based on the ratio of the average frequency of each intrinsic mode function component to the highest frequency of liquid level change caused by urea consumption, and the smallest i-th value whose average frequency is greater than the highest frequency of urea consumption is taken as p; the first... Up to the Kth intrinsic mode function component and the residual trend term By merging and reconstructing, the low-frequency trend of liquid level changes can be obtained. .

[0026] The high-frequency fluctuation components of the liquid level In this process, the instantaneous energy within a sliding window is calculated segment by segment. Where W is the window width, and the product of the original liquid level signal sampling frequency and the lowest liquid surface sloshing period is rounded up; the instantaneous energy is detected. Exceeding the transition energy threshold During a given time period, and in conjunction with the zero-crossing statistics of the high-frequency fluctuation components of the liquid level during that time period, segments that continuously cross the zero point more than a preset threshold number are marked as abnormal liquid level jump events caused by obstruction and reflection from internal components.

[0027] The low-frequency change trend of the liquid level With the steady-state liquid level sequence The difference is used to obtain the low-frequency offset. When the low-frequency offset Continuously greater than the positive offset threshold The duration exceeds the duration threshold When this occurs, it is marked as a positive accumulation event on the liquid surface.

[0028] Based on the current sampling time, the liquid volume interval j, the synthetic acceleration amplitude interval k, and the pitch angle interval l, the positive stacking probability is retrieved from the operating condition-liquid surface distortion feature description set. The negative exposure probability The average distortion amplitude The frequency of abnormal jumps and the component influence factor Calculate the non-consumable liquid level distortion intensity under the current operating conditions as follows: : In the formula, This is the reference energy value calibrated under the condition of a calm liquid surface; The weighting coefficient for the jump is determined based on the proportion of liquid volume calculation deviation caused by abnormal jump events in historical data. The weighting coefficient for accumulation is determined based on the proportion of liquid volume calculation deviation caused by accumulation events in historical data. The non-consumable liquid level distortion intensity It comprehensively reflects the degree of signal distortion in the current liquid level measurement value caused by non-real consumption due to internal components and liquid surface sloshing.

[0029] In this embodiment of the invention, the dominant type of liquid surface distortion is determined based on the intensity of the non-consumable liquid surface distortion and the rate of change of vehicle acceleration, and the offset direction and compensation coefficient of the sensor measurement value relative to the true volume are obtained, specifically: Obtain the first-order difference of the vehicle's longitudinal acceleration at the current sampling time. First-order difference with lateral acceleration Calculate the rate of change of the composite acceleration. ,in The sampling interval of the original liquid level signal, and the rate of change of the synthetic acceleration. It reflects the instantaneous and drastic changes in the vehicle's dynamic excitation.

[0030] The non-consumable liquid surface distortion intensity With the rate of change of the composite acceleration Perform joint determination: when and When the dominant type of current liquid level distortion is determined to be impact-jump type, the impact-jump type is caused by the impact of the liquid on the internal components of the tank during rapid acceleration, rapid deceleration, or sharp turning; when and When the dominant type of current liquid level distortion is determined to be persistent-accumulation type, this persistent-accumulation type is caused by the vehicle maintaining an inclined posture or driving on a stable curve for a long period of time, resulting in continuous congestion of liquid in the tank; when At that time, it was determined that the current degree of liquid surface distortion was relatively mild, and no dominant type was distinguished; among them The preset distortion intensity threshold is used to determine the distortion intensity. The preset threshold for the rate of change of acceleration is determined based on the interface between normal consumption and swaying interference in the actual vehicle calibration data.

[0031] Retrieve the component influence factor corresponding to the current liquid volume range j from the operating condition-liquid surface distortion feature description set. Determine the direction of the offset of the sensor measurement value relative to the actual volume. With compensation coefficient : When the dominant type is impact-jump type, the offset direction The offset direction is determined based on the relative relationship between the instantaneous vector direction of the composite acceleration and the sensor's installation position: if the composite acceleration vector direction points towards the side where the sensor is installed, the offset direction is positive, indicating that the sensor's measured value is too high; if the composite acceleration vector direction points away from the side where the sensor is installed, the offset direction is negative, indicating that the sensor's measured value is too low; the compensation coefficient... The calculation formula is as follows: In the formula, This represents the maximum value recorded in the historical data for the non-consumable liquid level distortion intensity. This refers to the maximum value recorded in the historical data regarding the rate of change of the composite acceleration. When the dominant type is continuous-stacking, the offset direction The pitch angle is determined based on its sign and the direction of the lateral acceleration: when the pitch angle... The value remains positive and its absolute value is greater than the preset pitch threshold. When the vehicle is in an uphill position, the liquid accumulates towards the rear of the tank. If the sensor is installed at the rear of the tank, the offset direction is positive; if it is installed at the front, the offset direction is negative. When the pitch angle... The value remains negative and its absolute value is greater than the preset pitch threshold. When the vehicle is descending a slope, the liquid accumulates towards the front of the tank. If the sensor is installed at the front of the tank, the offset direction is positive, indicating that the sensor measurement is too high; if the sensor is installed at the rear of the tank, the offset direction is negative, indicating that the sensor measurement is too low. The calculation formula for the compensation coefficient is as follows: In the formula, This represents the low-frequency offset at the current moment; This represents the steady-state liquid level sequence value at the current moment; This represents the current tilt posture and the duration of driving on the curve; To determine the reference duration, the statistical average of the durations of historical accumulated events is used; When the non-consumable liquid surface distortion intensity When the liquid surface distortion is deemed negligible, the offset direction... Take zero, the compensation coefficient Take 1.

[0032] In this embodiment of the invention, the original liquid level signal is dynamically corrected according to the compensation coefficient and fused with a liquid level statistical benchmark obtained under at least one state of long-term stationary and stable driving of the vehicle to output a compensated real-time liquid level value, specifically as follows: The following segments from historical vehicle driving data are identified as the baseline sampling window for liquid level statistics: the vehicle speed and engine speed remain at zero for a duration exceeding a first duration threshold. Marked as a long period of inactivity; vehicle speed continuously exceeds a preset driving speed threshold. The absolute value of longitudinal acceleration is consistently less than the acceleration fluctuation threshold. Furthermore, the absolute value of the lateral acceleration remains consistently less than the acceleration fluctuation threshold. The duration exceeds the second duration threshold. Marked as a stable driving window.

[0033] The liquid level statistical benchmark sampling window includes a long-term static window and a stable driving window; For each of the liquid level statistical reference sampling windows, calculate the original liquid level signal within the window. The arithmetic mean of these values ​​is used as the baseline liquid level value at the end of the window. By sequentially connecting adjacent reference liquid level values ​​and performing linear interpolation, a time-continuous statistical reference sequence of liquid levels is obtained. The first duration threshold With the second duration threshold The determination is based on the distinguishable duration of the liquid level drop caused by urea consumption.

[0034] Based on the direction of the offset of the sensor measurement value relative to the true volume With the compensation coefficient The original liquid level signal is dynamically corrected using the following formula to obtain the preliminary compensated liquid level. : In the formula, when the offset direction is determined to be that the sensor measurement value is too high, Pick When the offset direction is determined to indicate that the sensor measurement value is too low, Pick When the non-consumable liquid level distortion intensity hour, Pick .

[0035] Calculate fusion weights The fusion weight is adjusted according to the current non-consumable liquid surface distortion intensity. It increases with the increase of [amount], and the calculation formula is as follows: In the formula The preset distortion intensity threshold is used to determine the initial compensation liquid level. With the liquid level statistical reference sequence Weighted fusion is used to output the compensated real-time liquid level value. : ; When the intensity of the non-consumable liquid surface distortion is weak, the fusion weight Approaching The real-time liquid level value is mainly determined by the liquid level statistical reference sequence; when the liquid level distortion is severe, the fusion weight... Approaching The real-time liquid level value is mainly determined by the preliminary compensation liquid level, thereby achieving stable compensation of the original liquid level signal under different dynamic operating conditions.

[0036] This invention constructs a working condition-liquid surface distortion feature description set, coupling vehicle acceleration, pitch angle, and internal component layout information to characterize the statistical laws governing the interaction between liquid sloshing and internal components under different liquid volume ranges. It quantitatively represents previously difficult-to-model nonlinear distortions such as localized liquid surface jumps and accumulations. By adaptively separating high-frequency fluctuation components and low-frequency change trends from the original liquid level signal, it can effectively identify abnormal jumps and continuous accumulation characteristics caused by component obstruction and reflection from violently sloshing measurements, avoiding misjudging transient liquid surface distortions as actual consumption. The calculated non-consumption liquid surface distortion intensity is combined with the vehicle acceleration change rate to further analyze the dominant distortion type. The system differentiates between different urea tanks and introduces temperature and volume ranges to obtain offset direction and compensation coefficients, achieving precise matching of distortion amplitude and direction under different dynamic operating conditions. This overcomes the shortcomings of traditional methods that rely solely on fixed parameter filtering and cannot adapt to complex swaying patterns. The system uses compensation coefficients to dynamically correct the original signal and weightedly integrates it with the liquid level statistical benchmark obtained under long-term static or stable driving conditions. When swaying is severe, the system mainly relies on dynamic compensation results to track liquid level changes in a timely manner. When swaying is weak, the system automatically reverts to the statistical benchmark to ensure long-term measurement accuracy. This effectively eliminates severe fluctuations in liquid level display across the entire operating range, significantly improving the accuracy of urea consumption calculation and the reliability of low liquid level alarms.

[0037] This invention addresses the problem of nonlinear distortion of the liquid level caused by the irregular structure and internal components of a truck urea tank under dynamic operating conditions. By fusing vehicle acceleration, pitch angle, and internal component layout information, a set of operating condition-liquid level distortion feature descriptions is constructed. The original liquid level signal is separated into high-frequency fluctuations and low-frequency trends to extract the intensity of non-consumable liquid level distortion. Then, the dominant distortion type is distinguished and the offset direction and compensation coefficient are obtained. Combined with the liquid level statistical benchmark obtained from long-term static or stable driving, a weighted fusion correction is performed to achieve accurate restoration of the liquid level measurement value to the actual volume under all operating conditions. This effectively eliminates violent fluctuations in the liquid level display and significantly improves the accuracy of urea consumption calculation and the reliability of low liquid level alarms.

[0038] Example 2: This example introduces a truck urea level dynamic compensation system based on big data, such as... Figure 2As shown, it includes a distortion feature mapping module, a distortion intensity identification module, an offset and compensation decision module, and a dynamic compensation fusion module; The distortion feature mapping module is used to acquire the original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle and the pitch angle. Combined with the layout information of the internal components of the tank, it constructs the mapping relationship between acceleration excitation and local liquid surface fluctuation characteristics under different liquid volume ranges, and obtains the working condition-liquid surface distortion feature description set. The distortion intensity identification module is used to separate the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal according to the working condition-liquid surface distortion feature description set, extract the abnormal liquid surface jump and accumulation features caused by obstruction and reflection of internal components, and calculate the non-consumable liquid surface distortion intensity under the current working condition. The offset and compensation decision module is used to determine the dominant type of liquid surface distortion based on the intensity of the non-consumable liquid surface distortion and the rate of change of vehicle acceleration, and to obtain the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume. The dynamic compensation fusion module is used to dynamically correct the original liquid level signal according to the compensation coefficient, and fuse it with the liquid level statistical reference obtained by the vehicle under at least one state of long-term stationary and stable driving, and output the compensated real-time liquid level value.

[0039] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0040] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0041] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0042] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and method described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0043] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways.

[0044] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A dynamic compensation algorithm for truck urea level based on big data, characterized in that: Includes the following steps: The original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle, and the pitch angle are obtained. Combined with the layout information of the internal components of the tank, the working condition-liquid surface distortion feature description set is obtained. Based on the working condition-liquid surface distortion feature description set, the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal are separated, the abnormal liquid surface jumps and accumulation features are extracted, and the non-consumable liquid surface distortion intensity under the current working condition is calculated. Based on the non-consumable liquid surface distortion intensity and the vehicle acceleration change rate, the dominant type of liquid surface distortion is determined, and combined with the urea tank temperature and the current liquid volume range, the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume are obtained. The original liquid level signal is dynamically corrected according to the compensation coefficient and fused with the liquid level statistical benchmark obtained by the vehicle under at least one state of long-term stationary and stable driving to output the compensated real-time liquid level value.

2. The truck urea level dynamic compensation algorithm based on big data according to claim 1, characterized in that: The process of acquiring the original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle, and the pitch angle, combined with the internal component layout information of the tank, to obtain the working condition-liquid surface distortion feature description set, specifically involves: simultaneously acquiring the original liquid level signal, longitudinal acceleration, lateral acceleration, and pitch angle of the urea tank, and acquiring the internal component layout information of the tank; selecting segments of the vehicle that have been stationary for a long time and are traveling in a straight line at a constant speed from historical driving data, taking the mean of the original liquid level signal as the steady-state liquid level reference value, obtaining the steady-state liquid level sequence through interpolation, and calculating the liquid level distortion under unsteady working conditions; discretizing the total volume of the urea tank into multiple liquid volume intervals, defining the synthetic acceleration amplitude, and dividing the numerical range of the synthetic acceleration amplitude and the numerical range of the pitch angle into several intervals to form a two-dimensional working condition grid cell; for each liquid volume interval, determining the component influence factor using the component layout information; extracting sample points from the acquired historical data, and statistically analyzing the positive accumulation probability, negative exposure probability, average distortion amplitude, and abnormal jump frequency to obtain the working condition-liquid surface distortion feature description set.

3. The truck urea level dynamic compensation algorithm based on big data according to claim 1, characterized in that: The step involves separating the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal based on the working condition-liquid surface distortion feature description set, extracting abnormal liquid surface jumps and accumulation features, and calculating the non-consumable liquid surface distortion intensity under the current working condition. Specifically, this involves: performing adaptive signal separation on the original liquid level signal to obtain the intrinsic mode function components and residual trend terms arranged by frequency; reconstructing the high-frequency part into high-frequency fluctuation components of the liquid level, and merging and reconstructing the remaining parts into low-frequency change trends of the liquid level. Abnormal jump events are detected in the high-frequency fluctuation components of the liquid level, and accumulation events are detected in the low-frequency change trend of the liquid level; The non-consumable liquid surface distortion intensity is calculated by combining the current working condition-liquid surface distortion feature description set and component influence factors.

4. The truck urea level dynamic compensation algorithm based on big data according to claim 3, characterized in that: The non-consumable liquid surface distortion intensity is weighted and synthesized from the instantaneous energy under the abnormal jump event, the low-frequency offset under the accumulation event, the positive accumulation probability, the negative exposure probability, the average distortion amplitude, the frequency of the abnormal jump, and the component influence factor.

5. The truck urea level dynamic compensation algorithm based on big data according to claim 1, characterized in that: The process of determining the dominant type of liquid surface distortion based on the non-consumable liquid surface distortion intensity and the vehicle acceleration change rate specifically involves: calculating the composite acceleration change rate; when the non-consumable liquid surface distortion intensity is greater than or equal to the distortion intensity determination threshold and the composite acceleration change rate is greater than or equal to the acceleration change rate determination threshold, it is determined to be an impact-jump type; when the non-consumable liquid surface distortion intensity is greater than or equal to the distortion intensity determination threshold and the composite acceleration change rate is less than the acceleration change rate determination threshold, it is determined to be a continuous-accumulation type; when the non-consumable liquid surface distortion intensity is less than the distortion intensity determination threshold, it is determined that the current liquid surface distortion degree is relatively mild, and no dominant type distinction is made.

6. The truck urea level dynamic compensation algorithm based on big data according to claim 5, characterized in that: The offset direction of the sensor measurement value relative to the actual volume is obtained as follows: when the dominant type is impact-jump type, the offset direction is determined according to the relative relationship between the instantaneous vector direction of the composite acceleration and the sensor installation position; when the dominant type is continuous-accumulation type, the offset direction is determined according to the sign of the pitch angle and the continuous direction of the lateral acceleration; when the liquid surface distortion is negligible, the offset direction is zero.

7. The truck urea level dynamic compensation algorithm based on big data according to claim 1, characterized in that: The original liquid level signal is dynamically corrected according to the compensation coefficient and fused with the liquid level statistical benchmark obtained by the vehicle under at least one state of long-term stationary and stable driving to output the compensated real-time liquid level value. Specifically, the long-term stationary window and stable driving window are identified from the vehicle's historical driving data as liquid level statistical benchmark sampling windows; the arithmetic mean of the original liquid level signal is calculated for each of the liquid level statistical benchmark sampling windows as the benchmark liquid level value, and the liquid level statistical benchmark sequence is obtained by interpolation; the original liquid level signal is dynamically corrected according to the offset direction and the compensation coefficient to obtain the preliminary compensated liquid level; and the fusion weight is calculated, and the preliminary compensated liquid level is weighted and fused with the liquid level statistical benchmark sequence to output the compensated real-time liquid level value.

8. The truck urea level dynamic compensation algorithm based on big data according to claim 5, characterized in that: When the dominant type is impact-jump type, the compensation coefficient is determined based on the normalized results of the non-consumable liquid surface distortion intensity relative to its historical maximum value and the rate of change of the synthetic acceleration relative to its historical maximum value, combined with the component influence factor.

9. The truck urea level dynamic compensation algorithm based on big data according to claim 5, characterized in that: When the dominant type is continuous-accumulation type, the compensation coefficient is determined based on the low-frequency offset, steady-state liquid level sequence value and duration, combined with the component influence factor.

10. A truck urea level dynamic compensation system based on big data, used to implement the truck urea level dynamic compensation algorithm based on big data as described in any one of claims 1-9, characterized in that: It includes a distortion feature mapping module, a distortion intensity identification module, an offset and compensation decision module, and a dynamic compensation fusion module; The distortion feature mapping module is used to obtain the original liquid level signal of the urea tank, the longitudinal and lateral acceleration of the vehicle and the pitch angle, and combine it with the internal component layout information of the tank to obtain the working condition-liquid surface distortion feature description set. The distortion intensity identification module is used to separate the high-frequency fluctuation components and low-frequency change trends of the original liquid level signal according to the working condition-liquid surface distortion feature description set, extract the abnormal liquid surface jump and accumulation features, and calculate the non-consumable liquid surface distortion intensity under the current working condition. The offset and compensation decision module is used to determine the dominant type of liquid surface distortion based on the intensity of the non-consumable liquid surface distortion and the rate of change of vehicle acceleration, and to obtain the offset direction and compensation coefficient of the sensor measurement value relative to the actual volume. The dynamic compensation fusion module is used to dynamically correct the original liquid level signal according to the compensation coefficient, and fuse it with the liquid level statistical reference obtained by the vehicle under at least one state of long-term stationary and stable driving, and output the compensated real-time liquid level value.

Images