Real-time monitoring-based carbon emission metering device and method for truck transportation

Abstract

This application relates to the field of data processing technology, and in particular to a real-time monitoring-based carbon emission metering device and method for truck transportation. The proposed scheme involves real-time acquisition of load change data using three-phase force sensors installed on the truck axles, combined with weight time series data to identify loading and unloading events and divide the transportation into segments; calculating segment-level carbon emissions based on load data, mileage, and vehicle characteristic parameters for each transportation segment; further determining multiple transportation data zones based on speed statistical characteristics within the transportation segment, and correcting the carbon emission baseline line along the vertical axis according to operating condition fluctuation characteristics to generate a refined carbon emission line reflecting energy consumption differences under actual operating conditions; finally, constructing a segment-ticket coverage matrix based on waybill loading and unloading times to achieve carbon emission data allocation and uploading at the waybill level.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a carbon emission metering device and method for truck transportation based on real-time monitoring. Background Technology

[0002] Existing methods for calculating carbon emissions from truck transportation generally rely on energy consumption estimation models, which often contain significant errors. Because they fail to reflect variations in load and operating conditions during actual transport, the carbon emission data calculated using existing technologies are typically only average or theoretical values, making it difficult to accurately represent the true energy consumption level. Traditional methods exhibit poor consistency and comparability of carbon emission results under conditions of multiple loading and unloading operations, mixed shipments, and complex road conditions. This fails to meet the requirements for refined accounting and allocation at the waybill level, and the carbon emission data is insufficient to support subsequent audits and carbon asset management, thus affecting the credibility and widespread application of carbon accounting results.

[0003] To address the above issues, this application presents a device and method for measuring carbon emissions from truck transportation based on real-time monitoring. Summary of the Invention

[0004] The technical problem this application aims to solve is to address the shortcomings of existing technologies by providing a real-time monitoring-based carbon emission metering device and method for truck transportation. This method involves real-time acquisition of load change data using three-phase force sensors installed on the truck axles, identifying loading and unloading events based on weight time series data, and dividing the transportation into segments. Segment-level carbon emissions are calculated based on the load data, mileage, and vehicle characteristic parameters of each segment. Furthermore, multiple transportation data zones are determined based on speed statistics within each segment, and the carbon emission baseline is corrected along the vertical axis according to operating condition fluctuations to generate a refined carbon emission curve reflecting energy consumption differences under actual operating conditions. Finally, a segment coverage matrix is ​​constructed by combining the loading and unloading times of the waybill, enabling the allocation and uploading of carbon emission data at the waybill level.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] A method for measuring carbon emissions from truck transportation based on real-time monitoring, the method comprising:

[0007] The load changes of the truck are monitored by stress sensors installed on the truck axles to obtain continuous time series weight data, and loading and unloading events are identified based on the weight data.

[0008] Based on the identified loading and unloading events, the transportation process is divided into multiple transportation segments, and the mileage of each transportation segment is calculated by combining the truck driving trajectory data.

[0009] The carbon emission data is calculated based on the weight data, mileage and vehicle characteristic parameters of each transportation segment, and then uploaded to the cloud server.

[0010] Identifying loading and unloading events based on the weight data includes:

[0011] The weight data is sampled and smoothed according to time sequence, and then converted into a weight change curve;

[0012] By calculating the rate of weight change between sampling points in the weight change curve, a weight change rate curve is obtained. Based on a preset rate of change threshold and duration threshold, the weight change rate curve is analyzed to identify segments where the weight change rate is greater than the rate of change threshold and the duration is greater than the duration threshold. These segments are then marked as candidate event areas.

[0013] Feature extraction is performed on the candidate event regions to obtain the direction of change, and the candidate event regions are classified into loading and unloading events based on the direction of change.

[0014] The formula for calculating the carbon emission data is as follows:

[0015] ,

[0016] in, This refers to the carbon emission data corresponding to the transportation segment. This refers to the carbon emissions per kilometer when the truck is unloaded. This refers to the carbon emissions per kilometer when the truck is fully loaded. This is the ratio of the truck's real-time load to its rated load. This refers to the mileage traveled by the freight truck within the transport section.

[0017] The method further includes:

[0018] Based on the carbon emission data and the corresponding transportation segments, a carbon emission baseline broken line is constructed, wherein the horizontal axis of the carbon emission baseline broken line is the corresponding transportation segment, the corresponding transportation segments are sorted according to the order of the travel process, and the vertical axis of the carbon emission baseline broken line is the carbon emission data corresponding to the transportation segment.

[0019] For each transportation segment, multiple transportation data areas are determined to characterize the operating conditions of the transportation segment, wherein the transportation data area is a time sub-interval within the transportation segment;

[0020] Based on the vehicle speed statistics of the transportation data area and the proportion of the transportation data area in the transportation segment, the coordinate movement of the transportation data area in the carbon emission baseline is calculated.

[0021] By summing the coordinate movement of multiple transportation data areas, a correction value is obtained. The carbon emission baseline is then corrected based on the correction value to obtain the carbon emission fine-tuning line corresponding to the transportation segment.

[0022] The stress sensor is a three-phase stress sensor, and the determination of multiple transportation data areas used to characterize the operating conditions of the transportation section includes:

[0023] The raw data collected by the three-phase force sensors in the transportation section is acquired, and the raw data is preprocessed to obtain the vehicle speed time series, wherein the preprocessing includes time synchronization, outlier removal and smoothing filtering;

[0024] The transportation segment is traversed by a sliding window based on the vehicle speed time series to generate candidate time sub-intervals. The window length of the sliding window is set according to the transportation segment duration, and the step size is set according to the vehicle speed time series.

[0025] For each candidate time sub-interval, calculate vehicle speed statistical features, wherein the vehicle speed statistical features include at least one of the following: average vehicle speed, vehicle speed variance, idling ratio, start-up frequency density, and rapid acceleration / deceleration event density.

[0026] According to the preset operating condition identification rules, the candidate time sub-intervals are classified in combination with vehicle speed statistical features. The transportation segments are then re-sorted based on the classification results to obtain multiple transportation data areas. Each transportation data area corresponds to a classification result, which includes congestion conditions, constant speed conditions, and slope conditions.

[0027] The calculation of the coordinate shift of the transportation data area within the carbon emission baseline includes:

[0028] For each transportation data area, the vehicle speed statistical characteristics of the candidate time sub-intervals in the transportation data area are summarized, and the comprehensive operating condition characteristic value of the transportation data area is calculated. The comprehensive operating condition characteristic value includes: the weighted average of the average vehicle speed of the candidate time sub-intervals, the root mean square value of the vehicle speed variance, the weighted average of the idling speed ratio, the average of the starting frequency density and the average of the rapid acceleration and deceleration event density.

[0029] The operating condition fluctuation index of the transportation data area is calculated based on the comprehensive operating condition characteristic value, wherein the operating condition fluctuation index is used to characterize the stability of the truck operating status within the transportation data area.

[0030] Based on the operating condition fluctuation index and the percentage duration, the carbon emission correction coefficient corresponding to the transportation data area is calculated, and the carbon emission correction coefficient is converted into the corresponding coordinate movement amount, which moves along the vertical axis of the carbon emission baseline line.

[0031] The method further includes:

[0032] Obtain waybill data corresponding to multiple waybills for a truck. Based on the loading and unloading times of each waybill data on the truck, construct a segment ticket coverage matrix, wherein the rows of the segment ticket coverage matrix are the transportation segments, the columns of the segment ticket coverage matrix are the waybills, and the matrix elements of the segment ticket coverage matrix are used to characterize the coverage ratio of the waybill data in the corresponding transportation segment.

[0033] For each waybill, a weighted average is performed based on the load percentage of the corresponding waybill within the transportation segment to obtain the allocation weight of the waybill in the corresponding transportation segment, thereby updating the segment ticket coverage matrix and obtaining the updated matrix.

[0034] The carbon emission data is updated according to the update matrix to obtain waybill-level carbon emission data, and the waybill-level carbon emission data is uploaded to the cloud server.

[0035] The method of obtaining the allocation weight of the waybill on the corresponding transportation segment includes:

[0036] For each waybill, the effective transportation range of the waybill is determined based on the matrix elements of the waybill in the segment coverage matrix, where the effective transportation range represents the transportation segment corresponding to the waybill.

[0037] Based on the weight data within the transportation segment, the load percentage of the waybill within the transportation segment is calculated, where the load percentage is the ratio of the load corresponding to the waybill to the real-time load of the transportation segment.

[0038] A real-time monitoring-based carbon emission metering device for truck transportation, the device comprising:

[0039] A stress sensing module, installed on the axle of a truck, is used to monitor the load changes of the truck in real time during loading, transportation and unloading, and output stress signals corresponding to the load; the stress sensing module includes at least one three-phase stress sensor and a signal conditioning circuit, the signal conditioning circuit being used to amplify, filter and convert the stress signal to analog-to-digital.

[0040] The data processing module is electrically connected to the stress sensing module and is used to collect and analyze stress signals in real time to obtain weight time series data. Based on the weight change curve, it identifies loading and unloading events, divides transportation segments, calculates the driving mileage and corresponding carbon emission data of each transportation segment, and calculates the carbon emission correction coefficient based on the operating condition fluctuation index and time proportion of each transportation segment to obtain the carbon emission result of the transportation segment after operating condition correction.

[0041] The cloud collaboration module is used to receive carbon emission data from the data processing module, perform data aggregation and analysis, construct carbon emission records for the entire journey of the truck and a carbon emission allocation report at the waybill level, and upload them to the cloud server.

[0042] The data processing module includes a microcontroller main control unit, a storage unit, and a processing unit for running carbon emission calculation algorithms, wherein:

[0043] The microcontroller main control unit is used to control the data acquisition frequency of the stress sensing module, synchronously trigger and time-calibrate the acquired three-phase stress signals, and transmit the digital signal converted by the signal conditioning circuit to the processing unit.

[0044] The storage unit is used to cache weight time series data, transportation segment division results, operating condition fluctuation index, carbon emission correction coefficient and carbon emission calculation results, and periodically packages and encrypts the data before uploading it to the cloud server through the cloud collaboration module.

[0045] The processing unit is used to execute data parsing and carbon emission calculation programs, including: generating weight change curves from stress signals; identifying loading and unloading events and dividing transportation segments based on the weight change rate; calculating the mileage of each transportation segment in combination with the acquired driving trajectory; and calculating the carbon emission data corresponding to the transportation segment based on the vehicle's characteristic parameters.

[0046] The processing unit is also used to perform operating condition identification and carbon emission correction procedures, including: obtaining the truck speed time series through three-phase force signals, identifying congested operating conditions, constant speed operating conditions and slope operating conditions in the transportation segment; calculating the operating condition fluctuation index and the proportion of duration for each transportation segment, determining the carbon emission correction coefficient, and correcting the carbon emission results of the transportation segment.

[0047] Compared with the prior art, the beneficial effects of this application are:

[0048] This application integrates key aspects such as axle stress monitoring, loading / unloading event identification, operating condition correction, and waybill allocation into a unified design, constructing a metering framework that can continuously and accurately reflect dynamic changes in carbon emissions in real transportation scenarios. The solution does not rely on fuel consumption interfaces or energy consumption sensors; it can calculate carbon emissions throughout the entire process using only axle stress signals and positioning data, avoiding the applicability issues of traditional methods in situations with limited interfaces and missing data. By introducing baseline broken lines and operating condition correction mechanisms into the segment-level carbon emission calculation, the carbon emission results can dynamically respond to complex operating conditions such as congestion, constant speed, and slopes, achieving continuous correction from ideal to reality. Attached Figure Description

[0049] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0050] Figure 1 An exemplary application scenario diagram provided for an embodiment of this application;

[0051] Figure 2A modular structure diagram of a real-time monitoring-based carbon emission metering device for truck transportation provided in this application embodiment;

[0052] Figure 3 A flowchart illustrating the real-time monitoring-based carbon emission measurement method for truck transportation provided in this application embodiment;

[0053] Figure 4 A flowchart illustrating another method for measuring carbon emissions from truck transportation based on real-time monitoring, provided as an embodiment of this application;

[0054] Figure 5 This is a schematic diagram of the carbon emission baseline broken line provided in the embodiments of this application;

[0055] Figure 6 This is a schematic diagram of the process for obtaining the transportation data area provided in an embodiment of this application;

[0056] Figure 7 This is a flowchart illustrating another method for measuring carbon emissions from truck transportation based on real-time monitoring, provided as an embodiment of this application. Detailed Implementation

[0057] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0058] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0059] Before presenting a specific method in this embodiment, it is necessary to define why it is necessary to calculate and refine transportation carbon emissions. For freight operators, carbon emissions are not simply an environmental indicator, but a statistic directly embedded in their operations, compliance, and settlement systems.

[0060] On the one hand, transportation companies need to establish verifiable carbon emission ledgers to meet regulatory reporting, customer disclosure, and internal emission reduction target assessment requirements. On the other hand, carbon cost-carbon performance clauses are increasingly appearing in freight rate structures, requiring the reasonable allocation of emissions from a single trip across specific waybills to achieve the principle of "whoever's goods, whose carbon." For the financial and supply chain sectors, transportation carbon intensity has become a key factor in credit granting, green procurement, and supplier evaluation. Only companies with auditable, traceable, and recalculated measurement chains can gain recognition in green finance and customer bidding.

[0061] In a typical road freight scenario, measuring the carbon emissions of a specific transport activity is not the same as estimating the average emissions per kilometer per vehicle. The stop-and-go nature of last-mile urban delivery, the half-load operation after transshipment on main lines, the long slopes and sharp bends of mountain roads, and the low-speed idling and temperature-controlled conditions of refrigerated vehicles all contribute to the fact that the marginal energy consumption per unit mileage is far from the ideal steady state.

[0062] In practice, companies often extrapolate linearly from transport distance, load capacity, and vehicle parameters to obtain a seemingly reasonable on-paper value. However, when there are frequent starts and stops and congestion during the journey, the actual carbon emissions are significantly higher than this ideal baseline.

[0063] Understandably, directly reading fuel consumption is the best method for calculating carbon emissions; however, in the commercial vehicle ecosystem, this approach is difficult to implement from both engineering and compliance perspectives.

[0064] Different OEMs' ECUs / gateways have varying degrees of openness to fuel consumption-related PID parameters such as fuel injection quantity and pulse width. This makes it difficult for fleets spanning different brands and emission stages to obtain stable, continuous, and verifiable fuel consumption timelines. For safety and warranty reasons, operators typically restrict third-party devices from permanent access to CAN / OBD; even if interface data can be read, its reliability and traceability are insufficient to meet settlement and regulatory requirements. Adding fuel flow meters to existing vehicles requires disconnecting fuel lines, bypassing, and periodic calibration, making the hardware and downtime costs unacceptable at scale. Therefore, this embodiment does not rely on direct fuel consumption measurement but instead focuses on joint modeling of measurable variables.

[0065] The core logic of this application stems from two easily overlooked but reliably extractable signals:

[0066] One is the micro-strain change caused by axle stress, which is highly correlated with the actual load of the vehicle and can provide a continuous weight time series without modifying the oil circuit and bus.

[0067] Secondly, the trajectory observations derived from the three-phase stress signals in conjunction with positioning data can calibrate the working conditions within a section without reading the power system parameters.

[0068] Based on these two types of signals, this embodiment first identifies loading and unloading events using weight transitions, dividing the entire process into natural transportation segments, and calculating the mileage and duration of each segment by combining the trajectory. Then, it establishes an ideal baseline for segment-level emissions using vehicle characteristic parameters and load ratios. Considering the non-steady-state nature of driving conditions, multiple representative transportation data areas are extracted within each transportation segment. These data areas are determined by speed statistical characteristics and the segment's proportion of time, used to characterize the offset effects of start-stop, congestion, constant speed cruising, and slope conditions on energy consumption. Finally, in a unified coordinate domain, the segment-level baseline is expressed as a cumulative broken line, and the nodes of the broken line are directionally corrected according to the operating intensity of each transportation data area, so that both segment-level and cumulative emissions closely approximate real operating loads and driving behaviors.

[0069] Those skilled in the art will understand that the application scenarios of the method in this application are not limited to a single route or a single vehicle type. In multi-ticket, multi-station transportation with mixed trunk and branch lines, the natural segmentation of loading and unloading events can avoid metering overlap caused by mismatches in the timing windows of different orders; in congested conditions during morning and evening rush hours for urban distribution vehicles, the speed statistics characterize the idling ratio and starting density, making the energy consumption penalty of stop-and-go driving explicitly accounted for; in mountainous and windy road sections, the slope-resistance proxy formed by the positioning elevation and speed fluctuations constrains the net effect of uphill traction and downhill braking recovery; in cold chain transportation scenarios with low speed and constant temperature, the proportion of long periods of low speed and high idling within a segment will systematically increase the correction term.

[0070] refer to Figure 1 , Figure 1 This is an exemplary application scenario diagram provided for an embodiment of this application.

[0071] Figure 1 The diagram shows a truck body equipped with a metering device used to monitor load changes in real time during truck operation, identify loading and unloading events, and calculate carbon emission data for the corresponding transportation segment. It is understood that the multiple truck bodies shown represent the same truck in different transportation stages, and do not represent multiple independent vehicles.

[0072] Figure 1 The diagram further illustrates that the truck passes through four loading / unloading points during its transport process: Loading / Unloading Point 1, Loading / Unloading Point 2, Loading / Unloading Point 3, and Loading / Unloading Point 4, thus creating three transport segments: Transport Segment 1, Transport Segment 2, and Transport Segment 3. Within each transport segment, the metering device calculates the carbon emission data based on changes in the vehicle's load and its travel trajectory.

[0073] Figure 1Furthermore, it illustrates that during each transportation segment, the metering device transmits the corresponding carbon emission data for that segment to a cloud server. The dashed lines represent the data transmission process. It can be understood that in this application, the transmission process can occur at each loading / unloading point, or it can be performed uniformly after the transportation activity is completed. Figure 1 The illustrations are merely illustrative and are not intended to limit the scope of this application.

[0074] refer to Figure 2 , Figure 2 A modular structure diagram of a real-time monitoring-based carbon emission metering device for truck transportation provided in this application embodiment.

[0075] like Figure 2 As shown, the real-time monitoring-based truck transportation carbon emission metering device includes:

[0076] Device body 100;

[0077] The stress sensing module 101 is installed on the axle of the truck and is used to monitor the load changes of the truck in real time during loading, transportation and unloading, and output stress signals corresponding to the load. The stress sensing module includes at least one three-phase stress sensor and a signal conditioning circuit. The signal conditioning circuit is used to amplify, filter and convert the stress signal to analog-to-digital.

[0078] The data processing module 102 is electrically connected to the stress sensing module 101 and is used to collect and analyze stress signals in real time to obtain weight time series data. It also identifies loading and unloading events and divides transportation segments according to the weight change curve, calculates the driving mileage and corresponding carbon emission data of each transportation segment, calculates the carbon emission correction coefficient according to the working condition fluctuation index and time proportion of each transportation segment, and obtains the carbon emission result of the transportation segment after working condition correction.

[0079] The cloud collaboration module 103 is used to receive carbon emission data from the data processing module 102, perform data aggregation and analysis, construct carbon emission records for the entire journey of the truck and a waybill-level carbon emission allocation report, and upload them to the cloud server.

[0080] In one example, the data processing module 102 includes a microcontroller main control unit, a storage unit, and a processing unit for running carbon emission calculation algorithms, wherein:

[0081] The microcontroller main control unit is used to control the data acquisition frequency of the stress sensing module, synchronously trigger and time-calibrate the acquired three-phase stress signals, and transmit the digital signal converted by the signal conditioning circuit to the processing unit.

[0082] The storage unit is used to cache weight time series data, transportation segment division results, operating condition fluctuation index, carbon emission correction coefficient and carbon emission calculation results, and periodically packages and encrypts the data before uploading it to the cloud server through the cloud collaboration module.

[0083] The processing unit is used to execute data parsing and carbon emission calculation programs, including: generating weight change curves from stress signals; identifying loading and unloading events and dividing transportation segments based on the weight change rate; calculating the mileage of each transportation segment in combination with the acquired driving trajectory; and calculating the carbon emission data corresponding to the transportation segment based on the vehicle's characteristic parameters.

[0084] The processing unit is further used to execute the operating condition identification and carbon emission correction procedure, including: calculating the truck speed time series through three-phase force signals, identifying congested operating conditions, constant speed operating conditions and slope operating conditions in the transportation segment; calculating the operating condition fluctuation index and the proportion of duration for each transportation segment, determining the carbon emission correction coefficient, and correcting the carbon emission results of the transportation segment.

[0085] Next, combined Figure 3 The present application further describes the method for measuring carbon emissions from truck transportation based on real-time monitoring, which includes:

[0086] S1: Monitor the load changes of the truck by using stress sensors installed on the truck axle, obtain continuous time series weight data, and identify loading and unloading events based on the weight data;

[0087] In this embodiment, the stress sensor is specifically a three-phase stress sensor, installed on the front and rear axles of the vehicle, used to collect micro-strain signals of the truck in real time during driving and loading / unloading processes. The collected signals are amplified and filtered by a signal conditioning circuit, and then converted into digital data by an AD conversion module, forming a continuous weight time series. By performing a moving average and threshold discrimination on this time series, the load transition behavior can be stably identified in different states such as when the truck is stationary, loading, or unloading, thereby determining the timing of loading / unloading events and the corresponding weight change amplitude.

[0088] It is understandable that the axle load stress-based detection method in this application does not require modification of the vehicle's fuel system or engine system, does not rely on onboard OBD or bus interface information, and is applicable to truck platforms of different brands and models. On the one hand, it can avoid the limitations of traditional fuel consumption collection methods, such as interface privatization, discontinuous sampling, and untraceable data; on the other hand, through time-domain analysis of stress signals, it can achieve real-time judgment of loading status during transportation, avoid errors caused by human input, and enhance the objectivity and auditability of carbon emission measurement.

[0089] S2: Based on the identified loading and unloading events, the transportation process is divided into multiple transportation segments, and the mileage of each transportation segment is calculated by combining the truck driving trajectory data.

[0090] In this embodiment, the start and end times of loading and unloading events constitute natural segment boundaries, and each segment is an independent transportation unit of the vehicle under the same loading condition. The latitude and longitude trajectory of the truck within each segment is collected by the truck's built-in positioning module, and the corresponding mileage is calculated based on the time series between trajectory points. If the trajectory signal is temporarily interrupted, the path within the segment can be recovered by inertial calculation or interpolation.

[0091] Understandably, the purpose of dividing the transportation into segments is to divide the continuous transportation process into physically independent emission measurement units based on changes in load status, thereby maintaining consistency in measurement boundaries in subsequent carbon emission calculations and data reporting.

[0092] Unlike traditional methods that divide transportation segments by administrative regions, scheduling tasks, or mileage, the transportation segment division method in this embodiment can truly reflect the dynamic changes in cargo loading status, effectively distinguish between fully loaded, half-loaded, and empty stages, and make carbon emission calculations more precise and interpretable. It is especially suitable for complex business scenarios such as multiple tickets for one vehicle, multiple loading and unloading, or round-trip transportation.

[0093] S3: Calculate the corresponding carbon emission data based on the weight data, mileage and vehicle characteristic parameters of each transportation segment, and upload the carbon emission data to the cloud server;

[0094] In this embodiment, segment-level carbon emissions are calculated using a model combining empirical parameters and real-time load. Vehicle characteristic parameters include vehicle weight, rated load, and corresponding energy efficiency coefficient, which can be obtained through vehicle factory parameters or actual measurement calibration. This allows for approximate estimation of energy intensity without relying on real-time fuel consumption. Cloud-based uploading ensures that carbon emission data maintains consistent encoding and timestamps across different transportation segments, facilitating subsequent aggregation, analysis, and compliance verification.

[0095] Before going into the specific technical details of each step, it is necessary to further explain the core metrological dilemma and its technical logic that this application embodiment aims to solve.

[0096] For a long time, the calculation of carbon emissions from road freight has remained at the statistical level, that is, the total amount is estimated by simply multiplying the distance traveled and the emission factor. Although this method has statistical significance at the regional or industry level, it is difficult to meet the needs of regulatory compliance, carbon footprint disclosure and carbon trading settlement in the refined management of individual vehicles, trips and even single shipments.

[0097] Furthermore, in actual transportation operations, the same truck may load and unload goods multiple times during a single transport mission, forming a series of transport segments with varying loads and operating conditions. The energy consumption characteristics of each transport segment are comprehensively affected by the current load, terrain, vehicle speed distribution, and traffic flow status, and these characteristics change frequently and irregularly.

[0098] Based on this, the core logic proposed in this embodiment is not merely to fit carbon emissions under the condition of missing traditional fuel consumption data, but to shift the focus to directly observable physical quantities:

[0099] The real-time changes in axle stress have a one-to-one mechanical correspondence with the load, and its variation curve naturally contains complete information about the vehicle's loading, unloading, operation, and stationary states. By combining the stress signal with the positioning trajectory, a load-mileage bivariate time series model can be established without intruding on the vehicle's power system. Based on this, by introducing vehicle characteristic parameters, the theoretical emission baseline for each transportation segment can be physically defined.

[0100] It is easy to understand that the theoretical emission baseline only describes the emission level under steady-state ideal conditions, and significant deviations will still occur in real-world driving due to non-steady-state operating conditions. To address this issue, this embodiment uses three-phase force sensors to calculate the truck's speed time series and extracts the speed fluctuation behavior within the transportation segment as representative operating data areas. Each operating data area corresponds to different driving characteristics, such as constant speed cruising, hill traction, or stop-and-go driving. These characteristics not only reflect the vehicle's power output state but also imply the trend of energy consumption increase or decrease. Through statistical modeling of the operating data areas, this embodiment can superimpose a correction amount related to the intensity of operating condition fluctuations onto the carbon emission baseline, enabling the segment-level emission estimate to dynamically conform to the actual driving environment.

[0101] Those skilled in the art will understand that this application utilizes the vehicle's own perceptible structural stress signals to reconstruct the dynamic distribution pattern of transportation energy consumption and achieves a continuous transition from ideal emissions to actual emissions through operating condition mapping. For fleet operations, the method of this application can provide a unified carbon emission standard across vehicle models and routes without increasing hardware intrusion costs; for carbon auditing and carbon settlement, it can provide a continuous and traceable measurement chain as evidence, significantly improving the credibility of carbon emission data and the reproducibility of regulatory oversight.

[0102] In one example, identifying loading and unloading events based on the weight data includes:

[0103] The weight data is sampled and smoothed according to time sequence, and then converted into a weight change curve;

[0104] By calculating the rate of weight change between sampling points in the weight change curve, a weight change rate curve is obtained. Based on a preset rate of change threshold and duration threshold, the weight change rate curve is analyzed to identify segments where the weight change rate is greater than the rate of change threshold and the duration is greater than the duration threshold. These segments are then marked as candidate event areas.

[0105] In one example, the rate of change threshold is a proportional threshold related to the vehicle's rated load, and its value range can be a percentage corresponding to the rated load. The duration threshold is the minimum time length for continuously satisfying the weight change rate condition, and for example, its value range can be from 10 seconds to 300 seconds.

[0106] Feature extraction is performed on the candidate event regions to obtain the direction of change, and the candidate event regions are classified into loading and unloading events based on the direction of change.

[0107] It is understandable that the identification of loading and unloading events in this application is a prerequisite for achieving full-process carbon emission measurement. The identification principle in this application is based on the physical correspondence between changes in vehicle load and stress response.

[0108] When a truck is stationary or moving at a constant speed, its axle stress signal is in a relatively stable state; however, during loading or unloading, the total mass of the truck body changes abruptly, and the axle stress also undergoes a significant step or slope change.

[0109] To accurately capture this change, this embodiment first samples and smooths the original weight data according to a time series, eliminating instantaneous noise caused by uneven road surfaces, braking, or suspension bounce, ensuring the curve truly reflects the overall trend of load change. Subsequently, by calculating the weight change rate between adjacent sampling points, a continuous curve of the change rate over time can be obtained. When the truck is in loading / unloading mode, this change rate continuously exceeds the background noise level for a certain period, forming significant peak or trough intervals. By setting a combination of change rate threshold and duration threshold, short-term false signals caused by road disturbances or vehicle attitude changes can be effectively eliminated, retaining only segments with stable load transition characteristics as candidate event areas. Further discrimination is made based on the direction of change: when the weight change direction is upward and the magnitude exceeds the loading threshold, it is determined as a loading event; when the change direction is downward and the magnitude exceeds the unloading threshold, it is determined as an unloading event. Through this derivation process, automatic identification of loading / unloading events can be achieved solely based on stress sensing signals without external manual intervention or vehicle interface data, thus providing accurate and traceable time boundaries for subsequent transportation segment division and carbon emission calculation.

[0110] It is important to note that since each loading / unloading event typically corresponds to a loading or unloading node in the waybill lifecycle, loading / unloading events can be matched and bound to waybill events one-to-one, one-to-many, or many-to-one, i.e.:

[0111] The effective range of a waybill (from successful initial loading to final unloading) is determined by constraints such as event timestamps, geofence hits, and the tolerance matching between weight jumps and declared weights on the waybill. When the same event involves multiple shipments loading simultaneously or the same shipment being unloaded in batches, allocation is based on a priority order of time proximity, geofence consistency, and weight matching. If conflicts exist (such as weight differences exceeding tolerance or abnormal time overlap), the relevant events are marked as unique constraints to be verified, ensuring that the ranges do not overlap. Thus, the loading and unloading event sequence is mapped to the start and end boundaries of the waybill, which can be used to construct segment-shipment coverage relationships and the weight percentage within a segment, ensuring the traceability and verifiability of segment-level measurement results at the waybill level.

[0112] In one example, the carbon emission data is calculated using the following formula:

[0113] ,

[0114] in, This refers to the carbon emission data corresponding to the transportation segment. This refers to the carbon emissions per kilometer when the truck is unloaded. This refers to the carbon emissions per kilometer when the truck is fully loaded. This is the ratio of the truck's real-time load to its rated load. This refers to the mileage traveled by the freight truck within the transport section.

[0115] Next, we will further elaborate on the technical aspects of the carbon emission data correction method in this application.

[0116] refer to Figure 4 , Figure 4 This is a flowchart illustrating another method for measuring carbon emissions from truck transportation based on real-time monitoring, provided as an embodiment of this application.

[0117] S1: Monitor the load changes of the truck by using stress sensors installed on the truck axle, obtain continuous time series weight data, and identify loading and unloading events based on the weight data;

[0118] S2: Based on the identified loading and unloading events, the transportation process is divided into multiple transportation segments, and the mileage of each transportation segment is calculated by combining the truck driving trajectory data.

[0119] S3: Calculate the corresponding carbon emission data based on the weight data, mileage and vehicle characteristic parameters of each transportation segment;

[0120] S4: Correct the carbon emission data to obtain the carbon emission refinement line corresponding to each transportation segment, and upload the carbon emission refinement line to the cloud server.

[0121] In one example, the specific steps of S4 are as follows:

[0122] S4.1: Based on the carbon emission data and the corresponding transportation segment, construct a carbon emission baseline broken line, wherein the horizontal axis of the carbon emission baseline broken line is the corresponding transportation segment, the corresponding transportation segment is sorted according to the order of the travel process, and the vertical axis of the carbon emission baseline broken line is the carbon emission data corresponding to the transportation segment.

[0123] Specifically, constructing a carbon emission baseline polygon aims to establish emission relationships at the transportation segment level within a unified coordinate domain, enabling subsequent condition-based corrections to be implemented in the form of measurable node offsets. This polygon uses the order of transportation segment travel as the horizontal axis sequence and the baseline emissions within a segment as the vertical axis sequence. Through sequential accumulation and node-based representation, it forms a temporally consistent segment-level emission trajectory. This facilitates the inclusion of intra-segment and inter-segment variation information within the same data structure, avoiding issues such as discontinuities at segment interfaces or mismatches in accumulated emissions during subsequent corrections.

[0124] In this embodiment, the entire process is first divided into orderly transportation segments based on the timestamps of loading and unloading events. For each transportation segment, the baseline emissions within the segment are obtained based on the weight time series and vehicle characteristic parameters. This value is then written sequentially into a polyline structure in the form of nodes, with the horizontal axis of the node being the transportation segment number and the vertical axis being the baseline emissions of that segment.

[0125] refer to Figure 5 , Figure 5 This is a schematic diagram of the carbon emission baseline broken line provided in an embodiment of this application.

[0126] Figure 5 The diagram shows the correspondence between carbon emission data and transportation segments. The horizontal axis represents multiple transportation segments arranged in the order of travel, including transportation segment one, transportation segment two, and transportation segment three. The vertical axis represents the carbon emission data of the corresponding transportation segment, including carbon emission data one, carbon emission data two, and carbon emission data three.

[0127] Understandable, Figure 5 Each node in the graph corresponds to an independent transport segment. The vertical coordinate value of the node represents the carbon emissions of that transport segment. The dashed lines in the graph only indicate the changes in carbon emission data.

[0128] S4.2: For each transportation segment, determine multiple transportation data areas to characterize the operating conditions of the transportation segment, wherein the transportation data area is a time sub-interval within the transportation segment;

[0129] Specifically, identifying multiple transport data regions to characterize the operating conditions of a transport segment aims to extract representative time sub-intervals for energy consumption within the segment. This ensures that subsequent corrections are derived from actual speed behavior rather than segment-wide averages, reducing the risk of short-term extreme conditions being masked by smoothing. The transport data regions are all generated from speed time series within the segment, possessing clear start and end indices and duration quotas. They cover the main speed patterns within the segment and provide categorized representations of factors such as low-speed idling, acceleration / deceleration fluctuations, long-term constant speed, and gradient effects.

[0130] In one example, determining multiple transport data regions used to characterize the operating conditions of the transport segment includes:

[0131] S4.2.1: Obtain the raw data collected by the three-phase force sensor in the transportation section, and preprocess the raw data to obtain the vehicle speed time series, wherein the preprocessing includes time synchronization, outlier removal and smoothing filtering;

[0132] Specifically, under steady-state rolling conditions, there exists a dynamically modifiable relationship between axle stress and wheel-end angular velocity: changes in axle load cause an observable stress phase difference between ground deformation and micro-displacement at the wheel end. By coherently processing the three-phase signals, a speed proxy consistent with the vehicle's longitudinal motion can be obtained without connecting to the power bus. Furthermore, the three-phase layout provides spatial redundancy, which helps to offset single-point noise caused by road disturbances and local adhesion changes, resulting in better stability and computability of speed derivation.

[0133] In this embodiment, the three-phase signals are acquired at a fixed sampling frequency. The raw data is first synchronized in time using a dual-source correction method combining external time synchronization and a local high-stability clock to eliminate transmission and reception delays and jitter. Then, outlier removal is performed, using robust discrimination based on neighborhood residuals to exclude hardware jitter, transient interruptions, or extreme impact points. Next, smoothing filtering is applied, preferably using a bandpass-sliding window joint strategy to suppress low-frequency drift and high-frequency vibration while preserving effective components. After preprocessing, three-phase synthesis is performed to obtain the principal direction stress waveform. Combined with wheel circumference geometric parameters and phase drift relationships, a speed proxy sequence monotonically corresponding to the actual vehicle speed is generated. When positioning data is available, the speed proxy is weakly constrained and corrected using trajectory mileage to maintain consistency between the two within the segment.

[0134] In one optional implementation, the three-phase force signals are first synchronously acquired and then subjected to DC drift removal and bandpass filtering to retain the effective components related to tire rolling and structural vibration.

[0135] Then, phase consistency analysis and periodic feature extraction are performed on the three-phase signals. By identifying periodic variation features or dominant frequency components, time features related to wheel rotation are obtained.

[0136] Based on the vehicle tire circumference, axle structure parameters, or factory calibration parameters, the periodic characteristics are mapped to corresponding speed proxy values;

[0137] When vehicle positioning data is available, the speed agent is calibrated for consistency using the positioning mileage to keep the cumulative mileage error within a preset range;

[0138] When the positioning data is unavailable or abnormal, the most recent valid calibration result is used to continue outputting the velocity time series.

[0139] S4.2.2: Perform a sliding window traversal on the transportation segment according to the vehicle speed time series to generate candidate time sub-intervals, wherein the window length of the sliding window is set according to the transportation segment duration, and the step size is set according to the vehicle speed time series;

[0140] Specifically, the purpose of sliding window traversal of the transportation segment is to form a set of local segments with uniform granularity and comparability on the time axis, so that different speed states can be locally expressed and subjected to the same evaluation criteria in subsequent steps. The selection of window length and step size is related to the segment duration, speed fluctuation level, and target coverage. If the window is too short, the statistics will be unstable; if it is too long, it will mask short-term conditions and affect the sensitivity of subsequent corrections.

[0141] In this embodiment, the window length is proportionally set based on the duration of the transport segment, supplemented by upper and lower limits to accommodate ultra-short and ultra-long segments. The step size is determined based on the volatility of the speed time series; segments with high volatility use smaller step sizes to refine the identification of transport conditions, while segments with low volatility use larger step sizes to reduce redundancy. During the traversal, an alignment strategy is executed for each window boundary to ensure that the start and end of the window avoid speed abrupt changes as much as possible, reducing statistical confusion caused by crossing different transport conditions. When a window crosses a loading / unloading event or segment interface, it is automatically truncated and assigned to the corresponding transport segment, ensuring that the candidate sub-interval is consistent with the segment boundary.

[0142] S4.2.3: For each candidate time sub-interval, calculate the vehicle speed statistical characteristics, wherein the vehicle speed statistical characteristics include at least one of the following: average vehicle speed, vehicle speed variance, idling ratio, start-up frequency density, and rapid acceleration / deceleration event density.

[0143] Specifically, average vehicle speed is used to reflect the overall operating level, vehicle speed variance characterizes the stability, idling ratio reflects the duration of low-speed stopping, and the density of start-up frequency and the density of rapid acceleration and deceleration events reflect the frequency and intensity of driving events. These features together cover the main performance of congestion, cruising, hill traction and combined operating conditions.

[0144] It is understandable that vehicle speed statistical features can be extracted using conventional algorithms in the field. For example, the average vehicle speed can be obtained by using the weighted average of the speed proxy within the window, and the density of starting times can be obtained by identifying the crossing events of speed from below the low speed threshold to above the stable threshold. This application will not elaborate on these points here.

[0145] S4.2.4: According to the preset working condition identification rules, the candidate time sub-intervals are classified in combination with vehicle speed statistical features. The transportation segments are re-sorted according to the classification results to obtain multiple transportation data areas, where each transportation data area corresponds to a classification result. The classification results include congestion working conditions, constant speed working conditions, and slope working conditions.

[0146] Specifically, candidate time sub-intervals are classified according to preset operating condition identification rules, and the transportation segments are reordered accordingly. The aim is to cluster and merge candidates with the same or similar driving characteristics to form a transportation data area that can represent the main operating conditions within the segment.

[0147] In this embodiment, the operating condition identification rule consists of threshold judgment and hierarchical voting: congested operating conditions are mainly judged by high idling ratio, low average vehicle speed and high start density; constant speed operating conditions are mainly judged by low vehicle speed variance and medium-to-high average vehicle speed; and slope operating conditions are mainly judged by slope level label and continuous traction events. When the same candidate meets multiple category conditions at the same time, hierarchical voting is carried out according to priority, and the priority is set according to the strength and persistence of the impact on energy consumption. After classification, candidates of the same category that are adjacent or close in time are merged to generate continuous transportation data areas. If the length of the merged data area exceeds the upper limit, it is segmented according to the uniformity of statistics to maintain the consistency of features within each data area.

[0148] refer to Figure 6 , Figure 6 This is a schematic diagram of the process for obtaining the transportation data area provided in an embodiment of this application.

[0149] Figure 6 Taking transportation segment one as an example, seven candidate time sub-intervals, from one to seven, are shown, and the corresponding vehicle speed statistical features one to seven are calculated in sequence.

[0150] Figure 6 The diagram further illustrates how candidate time sub-intervals are identified according to the operating condition identification rules, resulting in the operating condition corresponding to each candidate time sub-interval.

[0151] Figure 6 The process of reordering and merging candidate time sub-intervals after completing the work condition identification is further illustrated to generate a representative transportation data area.

[0152] Specifically, Figure 6In the process, candidate time sub-intervals one through seven, after being classified according to the operating condition identification rules, were identified into different categories such as congested operating conditions, constant speed operating conditions, and climbing operating conditions. To ensure the continuity and statistical consistency of similar types of driving behaviors in terms of time sequence, this embodiment reorders similar candidate time sub-intervals according to their temporal order and statistical feature similarity, and merges sub-intervals with adjacent times and similar features into a unified transportation data area. Figure 6 Taking the section shown as an example, candidate time sub-interval 1, candidate time sub-interval 3, candidate time sub-interval 5, and candidate time sub-interval 6 are all identified as congested conditions, and therefore are merged into transportation data area 1 for congested conditions during the re-sorting stage; similarly, candidate time sub-interval 2 and sub-interval 7 are both constant speed conditions, and are formed into transportation data area 2 for constant speed conditions after re-sorting; candidate time sub-interval 4 is an uphill condition with obvious gradient changes, and is formed independently into transportation data area 3.

[0153] S4.3: Based on the vehicle speed statistics of the transportation data area and the proportion of the transportation data area in the transportation segment, calculate the coordinate movement of the transportation data area in the carbon emission baseline broken line;

[0154] Specifically, based on the vehicle speed statistics of the transportation data area and the proportion of time the data area occupies within the transportation segment, the coordinate shift of the transportation data area in the carbon emission baseline broken line is calculated. This maps the impact of operating conditions within the segment on energy consumption into an observable offset on the vertical axis of the broken line, allowing differences in operating conditions to be reflected in segment-level emissions through node movement. The coordinate shift is calculated using the data area as the unit, and the direction and magnitude of the offset are determined based on statistics such as speed fluctuations, idling ratio, start-up density, and ramp labels. Then, it is weighted according to the proportion of time the data area occupies within the entire segment, ensuring that the impact of operating conditions that dominate for a long time on segment-level emissions is more fully realized, and that short-term anomalies are not excessively amplified.

[0155] In one example, calculating the coordinate shift of the transport data area within the carbon emission baseline polygon includes:

[0156] For each transportation data area, the vehicle speed statistical characteristics of the candidate time sub-intervals in the transportation data area are summarized, and the comprehensive operating condition characteristic value of the transportation data area is calculated. The comprehensive operating condition characteristic value includes: the weighted average of the average vehicle speed of the candidate time sub-intervals, the root mean square value of the vehicle speed variance, the weighted average of the idling speed ratio, the average of the starting frequency density and the average of the rapid acceleration and deceleration event density.

[0157] The operating condition fluctuation index of the transportation data area is calculated based on the comprehensive operating condition characteristic value, wherein the operating condition fluctuation index is used to characterize the stability of the truck operating status within the transportation data area.

[0158] Based on the operating condition fluctuation index and the percentage duration, the carbon emission correction coefficient corresponding to the transportation data area is calculated, and the carbon emission correction coefficient is converted into the corresponding coordinate movement amount, which moves along the vertical axis of the carbon emission baseline line.

[0159] Understandably, the determination of the coordinate displacement comes from mapping the increase or decrease in energy consumption due to the velocity behavior within a segment to a bounded offset of the polyline's vertical axis:

[0160] Specifically, firstly, within each transportation data region, the candidate time sub-intervals are weighted and aggregated. The weights are determined by the sub-interval duration, data integrity, and feature stability, ensuring that average vehicle speed is primarily weighted by duration, idling ratio by continuous stop weight, and speed variance by confidence weight after anomaly removal, resulting in a set of comprehensive operating condition characteristic values. Then, an operating condition fluctuation index is constructed based on this set of characteristic values. The operating condition fluctuation index follows a monotonic relationship: the index increases with increasing fluctuation and load, and decreases with stable cruising. Higher speed dispersion, greater density of start-up and rapid acceleration / deceleration events, and higher idling ratio all result in larger indices. When slope or long downhill labels are present, additional factors such as uphill traction or recovery attenuation are applied to the index, respectively, without altering the overall monotonicity. Finally, the operating condition fluctuation index is compared with the data for that transportation data region. The proportion of time spent in each segment is merged to form an intermediate amount of the carbon emission correction coefficient. This intermediate amount is then converted into a coordinate shift in the vertical direction through a set of pre-defined mapping intervals: congestion and frequent starts and stops correspond to positive offset, constant speed cruising corresponds to weak positive or zero offset, and long downhill sections with braking recovery correspond to weak negative offset. To avoid excessive amplification of local extreme segments, the mapping process employs standardization and truncation rules to compress inputs that exceed empirical boundaries. Physical rationality and continuity constraints are also applied to the shift amount to ensure that the shift amount of multiple data areas within the same transportation segment does not jump or accumulate in the opposite direction after time splicing. The final output coordinate shift amount, along with the time index and feature summary of the source data area, is recorded for subsequent directed correction of the carbon emission baseline polygonal nodes and for recalculation and auditing in the cloud.

[0161] S4.4: Summarize the coordinate movement of multiple transportation data areas to obtain correction values, and correct the carbon emission baseline line according to the correction values ​​to obtain the carbon emission fine-tuning line corresponding to the transportation segment;

[0162] Specifically, the coordinate shifts of multiple transportation data areas are aggregated to obtain correction values, which are then used to correct the carbon emission baseline polygon. This integrates the influence of different time-period operating conditions scattered within a segment into a single segment-level correction value, which is directly applied to the polygon node to form the refined carbon emission polygon. The refined carbon emission polygon maintains structural consistency with the baseline polygon, applying only a bounded offset based on operating conditions to the vertical axis value, thereby achieving unified calibration of segment-level emissions. At the same time, it preserves the comparison relationship between the baseline and refined trajectories, facilitating recalculation and verification.

[0163] In this embodiment, the coordinate movement of all transportation data areas within the same transportation segment is first weighted and summed according to their duration proportion and category weight to obtain the total correction amount for that segment. For data segments with compound working conditions, a category conflict handling rule is introduced. When the movement trends of two categories are opposite and the time overlap is high, the dominant trend is determined according to the priority of speed stability and slope level, and the other trend is attenuated according to the overlap ratio to avoid the correction amounts canceling each other out or excessively superimposing. After obtaining the total correction amount, the vertical axis value of the corresponding node is moved according to the correction amount, and the node difference field is updated. The consistency of the cumulative relationship between adjacent nodes is checked. If anomalies such as no cumulative increase or cross-segment reversal occur, the boundary smoothing strategy is triggered to make slight adjustments in the node neighborhood to ensure that the overall polyline remains monotonic and interpretable.

[0164] Next, a specific and complete calculation example will be used to illustrate the whole process. The example is only to illustrate the feasibility of the calculation. The mileage section and specific working conditions can be adjusted according to the actual task and do not represent the actual values.

[0165] For example, the mileage of a certain transportation segment is 120 km, and the baseline calculation shows that the carbon emission of the transportation segment is 132 (this value is obtained from vehicle characteristic parameters, load ratio and mileage, and belongs to the baseline output result, in kg), and the total transportation time is 120 minutes.

[0166] For example, this application provides three transport data areas:

[0167] Data Zone 1 (Congestion): Duration 30 minutes, accounting for 25%; the overall score is in the high fluctuation range, with a corresponding adjustment ratio of 12%.

[0168] Data Section 2 (Constant Speed): Duration 70 minutes, accounting for approximately 58.3%; Overall score falls into the low fluctuation range, corresponding to a correction ratio of 2%.

[0169] Data Zone 3 (Raft Traction): Duration 20 minutes, accounting for approximately 16.7%; the overall score is in the medium fluctuation range, with a corresponding correction ratio of 8%.

[0170] It is understood that the correction ratio here is specifically obtained through a mapping between the comprehensive score level and the correction ratio range: the comprehensive score of the transportation data area is mapped to the corresponding fluctuation level, and the correction ratio is determined from the correction ratio range corresponding to that level according to a preset value selection rule; the value selection rule is to take the median value or default benchmark value of the range, and the mapping table and value selection rule can be configured by the cloud and distributed and updated according to vehicle type / emission stage / fleet operation scenario to ensure that the results are recalculated and auditable. This application is only providing an exemplary mapping method, and the specific mapping boundaries and default values ​​can be adjusted according to the actual calibration data without affecting the core idea of ​​this application.

[0171] The correction increments for each data region are as follows:

[0172] Congestion correction increment: Based on 132 kg, scaled by 12% and 25% of the time duration, resulting in approximately 3.96 kg;

[0173] Uniformly corrected increment: Based on 132 kg, scaled by 2% and according to the duration ratio of 58.3%, we get approximately 1.54 kg;

[0174] Ramp correction increment: Based on 132 kg, it is scaled by 8% and by 16.7% of the duration, resulting in approximately 1.76 kg.

[0175] The total correction for this transport segment is approximately 7.26 kg, therefore the corrected carbon emissions for this segment are approximately 139.26 kg. The corresponding line segment node value is moved up from 132 kg to 139.26 kg, and the category, score, correction ratio, duration percentage, and increment of the three data areas are uploaded to the cloud as audit fields.

[0176] Next, we will further elaborate on the technical content of the method in this application regarding carbon emission data at the waybill level.

[0177] refer to Figure 7 , Figure 7 This is a flowchart illustrating another method for measuring carbon emissions from truck transportation based on real-time monitoring, provided as an embodiment of this application.

[0178] S1: Monitor the load changes of the truck by using stress sensors installed on the truck axle, obtain continuous time series weight data, and identify loading and unloading events based on the weight data;

[0179] S2: Based on the identified loading and unloading events, the transportation process is divided into multiple transportation segments, and the mileage of each transportation segment is calculated by combining the truck driving trajectory data.

[0180] S3: Calculate the corresponding carbon emission data based on the weight data, mileage and vehicle characteristic parameters of each transportation segment;

[0181] S4: Correct the carbon emission data to obtain the carbon emission trimmed line corresponding to each transportation segment;

[0182] S5: Obtain truck waybill data and carbon emission data corresponding to the carbon emission refinement curve, process them to obtain waybill-level carbon emission data, and upload the waybill-level carbon emission data to the cloud server.

[0183] In one example, the specific steps of S5 are as follows:

[0184] S5.1: Obtain the waybill data corresponding to multiple waybills for the truck, and construct a segment ticket coverage matrix based on the loading and unloading time of each waybill data on the truck. The columns of the segment ticket coverage matrix are the transportation segments, the columns of the segment ticket coverage matrix are the waybills, and the matrix elements of the segment ticket coverage matrix are used to characterize the coverage ratio of the waybill data in the corresponding transportation segment.

[0185] Specifically, the segment coverage matrix is ​​used to establish a calculable mapping relationship between segment-level measurement dominated by loading and unloading events and the lifecycle of waybills, avoiding time window mismatches and duplicate measurement caused by different business calibers. The rows of the matrix are arranged in the order of transportation segments obtained by dividing loading and unloading events, and the columns are the waybills involved in the same transportation task. The matrix elements express the effective coverage ratio of a waybill within the corresponding transportation segment, so that subsequent allocation can be stably implemented through matrix operations.

[0186] In this embodiment, the loading and unloading timestamps, loading and unloading locations, declared weight, and waybill identifier fields in the waybill data are first read. Alignment is then achieved using a combination of geofence triggering and event timestamp constraints: when a vehicle enters the loading fence and the weight time series shows an upward jump with the amplitude falling within the tolerance band of the declared weight, the fence is determined to match the waybill loading action. The unloading process uses logic identification in the opposite direction. If multiple shipments are loaded or unloaded in batches within the same time window, a preliminary allocation is made based on time proximity and weight matching, followed by a secondary decision based on location consistency. Shipments that fail to meet the tolerance are marked as pending verification and temporarily excluded from coverage calculations, ensuring that the uniqueness constraint of the matrix is ​​not violated.

[0187] S5.2: For each waybill, perform a weighted fusion based on the load ratio of the corresponding waybill in the transportation segment to obtain the allocation weight of the waybill in the corresponding transportation segment, so as to update the segment ticket coverage matrix and obtain the updated matrix;

[0188] Specifically, segment-level carbon emissions are a vehicle-level energy consumption metric. To ensure fairness and comparability of this metric across waybills, it needs to be allocated based on the contribution of each waybill within the segment to the vehicle's effective load. This embodiment only introduces the load ratio as a single, but recalculated, quantity as the basis for allocation, avoiding inconsistencies caused by introducing multiple weight factors. The load ratio is determined by the weight time series within the segment and the waybill coverage ratio: Within each transport segment, firstly, based on the coverage ratio obtained in S5.1, the weight transition sequence of the segment is decomposed to each waybill, forming a waybill-level segment load curve; when there is partial unloading or batch loading, the load curve is updated at the event granularity to maintain a new step value after the event occurs, ensuring that the sum of the loads of each waybill at the same time equals the real-time load within the segment; when there are occasional gaps or local fluctuations in the load curve, neighborhood interpolation and sliding window median correction are used to ensure the stability of the time integral of the load ratio. Subsequently, the load curve is normalized within the effective time of the transportation segment to obtain the average load percentage of each waybill in the segment. This average load percentage is used as the allocation weight of the waybill in the segment and written back to the segment ticket coverage matrix to form an update matrix. When two tickets completely overlap in the segment and have the same declared weight, they are allocated proportionally and each retains its own chain of evidence.

[0189] In one example, obtaining the allocation weight of the waybill on the corresponding transportation segment includes:

[0190] For each waybill, the effective transportation range of the waybill is determined based on the matrix elements of the waybill in the segment coverage matrix, where the effective transportation range represents the transportation segment corresponding to the waybill.

[0191] Based on the weight data within the transportation segment, the load percentage of the waybill within the transportation segment is calculated, where the load percentage is the ratio of the load corresponding to the waybill to the real-time load of the transportation segment.

[0192] S5.3: Update the carbon emission data according to the update matrix to obtain the waybill-level carbon emission data, and upload the waybill-level carbon emission data to the cloud server;

[0193] Specifically, the carbon emission at the waybill level is formed by using an update matrix as a bridge, mapping the carbon emission at the depot level to each waybill according to the allocation weight, and obtaining a carbon emission metric that corresponds one-to-one with the business document, which facilitates clearing, reconciliation and disclosure.

[0194] In this embodiment, the carbon emissions of each transportation segment are multiplied by the allocation weight of the corresponding column in the update matrix segment by segment and accumulated at the waybill dimension to form the total carbon emissions of the waybill throughout the entire transportation process. When the waybill spans multiple segments and partial unloading occurs midway, the total is naturally spliced ​​from the allocation results of each segment without additional processing. To adapt to the phased settlement of long-cycle tasks, the waybill-level carbon emissions simultaneously generate phased aggregate values. The phase granularity is driven by the settlement rules in the task order and aligned with the segment boundaries, which facilitates reconciliation on the business side.

[0195] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for measuring carbon emissions from truck transportation based on real-time monitoring, characterized in that, The method includes: The load changes of the truck are monitored by stress sensors installed on the truck axles to obtain continuous time series weight data, and loading and unloading events are identified based on the weight data. Based on the identified loading and unloading events, the transportation process is divided into multiple transportation segments, and the mileage of each transportation segment is calculated by combining the truck driving trajectory data. The corresponding carbon emission data is calculated based on the weight data, mileage and vehicle characteristic parameters of each transportation segment, and the carbon emission data is uploaded to the cloud server. The formula for calculating the carbon emission data is as follows: ， in, This refers to the carbon emission data corresponding to the transportation segment. This refers to the carbon emissions per kilometer when the truck is unloaded. This refers to the carbon emissions per kilometer when the truck is fully loaded. This is the ratio of the truck's real-time load to its rated load. This refers to the mileage traveled by the truck within the transport section. The method further includes: Based on the carbon emission data and the corresponding transportation segments, a carbon emission baseline broken line is constructed, wherein the horizontal axis of the carbon emission baseline broken line is the corresponding transportation segment, the corresponding transportation segments are sorted according to the order of the travel process, and the vertical axis of the carbon emission baseline broken line is the carbon emission data corresponding to the transportation segment. For each transportation segment, multiple transportation data areas are determined to characterize the operating conditions of the transportation segment, wherein the transportation data area is a time sub-interval within the transportation segment; Based on the vehicle speed statistics of the transportation data area and the proportion of the transportation data area in the transportation segment, the coordinate movement of the transportation data area in the carbon emission baseline is calculated. The coordinate movement of multiple transportation data areas is summarized to obtain correction values. The carbon emission baseline line is then corrected based on the correction values ​​to obtain the carbon emission fine-tuning line corresponding to the transportation segment. The method further includes: Obtain waybill data corresponding to multiple waybills for a truck. Based on the loading and unloading times of each waybill data on the truck, construct a segment ticket coverage matrix, wherein the rows of the segment ticket coverage matrix are the transportation segments, the columns of the segment ticket coverage matrix are the waybills, and the matrix elements of the segment ticket coverage matrix are used to characterize the coverage ratio of the waybill data in the corresponding transportation segment. For each waybill, a weighted average is performed based on the load percentage of the corresponding waybill within the transportation segment to obtain the allocation weight of the waybill in the corresponding transportation segment, thereby updating the segment ticket coverage matrix and obtaining the updated matrix. The carbon emission data is updated according to the update matrix to obtain waybill-level carbon emission data, and the waybill-level carbon emission data is uploaded to the cloud server.

2. The method for measuring carbon emissions from truck transportation based on real-time monitoring according to claim 1, characterized in that, Identifying loading and unloading events based on the weight data includes: The weight data is sampled and smoothed according to time sequence, and then converted into a weight change curve; By calculating the rate of weight change between sampling points in the weight change curve, a weight change rate curve is obtained. Based on a preset rate of change threshold and duration threshold, the weight change rate curve is analyzed to identify segments where the weight change rate is greater than the rate of change threshold and the duration is greater than the duration threshold. These segments are then marked as candidate event areas. Feature extraction is performed on the candidate event regions to obtain the direction of change, and the candidate event regions are classified into loading and unloading events based on the direction of change.

3. The method for measuring carbon emissions from truck transportation based on real-time monitoring according to claim 1, characterized in that, The stress sensor is a three-phase stress sensor, and the determination of multiple transportation data areas used to characterize the operating conditions of the transportation section includes: The raw data collected by the three-phase force sensors in the transportation section is acquired, and the raw data is preprocessed to obtain the vehicle speed time series, wherein the preprocessing includes time synchronization, outlier removal and smoothing filtering; The transportation segment is traversed by a sliding window based on the vehicle speed time series to generate candidate time sub-intervals. The window length of the sliding window is set according to the transportation segment duration, and the step size is set according to the vehicle speed time series. For each candidate time sub-interval, calculate vehicle speed statistical features, wherein the vehicle speed statistical features include at least one of the following: average vehicle speed, vehicle speed variance, idling ratio, start-up frequency density, and rapid acceleration / deceleration event density. According to the preset operating condition identification rules, the candidate time sub-intervals are classified in combination with vehicle speed statistical features. The transportation segments are then re-sorted based on the classification results to obtain multiple transportation data areas. Each transportation data area corresponds to a classification result, which includes congestion conditions, constant speed conditions, and slope conditions.

4. The method for measuring carbon emissions from truck transportation based on real-time monitoring according to claim 1, characterized in that, The calculation of the coordinate shift of the transportation data area within the carbon emission baseline includes: For each transportation data area, the vehicle speed statistical characteristics of the candidate time sub-intervals in the transportation data area are summarized, and the comprehensive operating condition characteristic value of the transportation data area is calculated. The comprehensive operating condition characteristic value includes: the weighted average of the average vehicle speed of the candidate time sub-intervals, the root mean square value of the vehicle speed variance, the weighted average of the idling speed ratio, the average of the starting frequency density and the average of the rapid acceleration and deceleration event density. The operating condition fluctuation index of the transportation data area is calculated based on the comprehensive operating condition characteristic value, wherein the operating condition fluctuation index is used to characterize the stability of the truck operating status within the transportation data area. Based on the operating condition fluctuation index and the percentage duration, the carbon emission correction coefficient corresponding to the transportation data area is calculated, and the carbon emission correction coefficient is converted into the corresponding coordinate movement amount, which moves along the vertical axis of the carbon emission baseline line.

5. The method for measuring carbon emissions from truck transportation based on real-time monitoring according to claim 1, characterized in that, The method of obtaining the allocation weight of the waybill on the corresponding transportation segment includes: For each waybill, the effective transportation range of the waybill is determined based on the matrix elements of the waybill in the segment coverage matrix, where the effective transportation range represents the transportation segment corresponding to the waybill. Based on the weight data within the transportation segment, the load percentage of the waybill within the transportation segment is calculated, where the load percentage is the ratio of the load corresponding to the waybill to the real-time load of the transportation segment.

6. A real-time monitoring-based truck transportation carbon emission metering device, used to implement the real-time monitoring-based truck transportation carbon emission metering method as described in any one of claims 1-5, characterized in that, The device includes: A stress sensing module, installed on the axle of a truck, is used to monitor the load changes of the truck in real time during loading, transportation and unloading, and output stress signals corresponding to the load; the stress sensing module includes at least one three-phase stress sensor and a signal conditioning circuit, the signal conditioning circuit being used to amplify, filter and convert the stress signal to analog-to-digital. The data processing module is electrically connected to the stress sensing module and is used to collect and analyze stress signals in real time to obtain weight time series data. Based on the weight change curve, it identifies loading and unloading events, divides transportation segments, calculates the driving mileage and corresponding carbon emission data of each transportation segment, and calculates the carbon emission correction coefficient based on the operating condition fluctuation index and time proportion of each transportation segment to obtain the carbon emission result of the transportation segment after operating condition correction. The cloud collaboration module is used to receive carbon emission data from the data processing module, perform data aggregation and analysis, construct carbon emission records for the entire journey of the truck and a carbon emission allocation report at the waybill level, and upload them to the cloud server.

7. The truck transportation carbon emission metering device based on real-time monitoring according to claim 6, characterized in that, The data processing module includes a microcontroller main control unit, a storage unit, and a processing unit for running carbon emission calculation algorithms, wherein: The microcontroller main control unit is used to control the data acquisition frequency of the stress sensing module, synchronously trigger and time-calibrate the acquired three-phase stress signals, and transmit the digital signal converted by the signal conditioning circuit to the processing unit. The storage unit is used to cache weight time series data, transportation segment division results, operating condition fluctuation index, carbon emission correction coefficient and carbon emission calculation results, and periodically packages and encrypts the data before uploading it to the cloud server through the cloud collaboration module. The processing unit is used to execute data parsing and carbon emission calculation programs, including: generating weight change curves from stress signals; identifying loading and unloading events and dividing transportation segments based on the weight change rate; calculating the mileage of each transportation segment in combination with the acquired driving trajectory; and calculating the carbon emission data corresponding to the transportation segment based on the vehicle's characteristic parameters. The processing unit is also used to perform operating condition identification and carbon emission correction procedures, including: obtaining the truck speed time series through three-phase force signals, identifying congested operating conditions, constant speed operating conditions and slope operating conditions in the transportation segment; calculating the operating condition fluctuation index and the proportion of duration for each transportation segment, determining the carbon emission correction coefficient, and correcting the carbon emission results of the transportation segment.

Images