Carbon emission determination method, device, equipment and storage medium
By acquiring historical and real-time operational data of new energy vehicles, identifying target characteristics, and constructing carbon emission models, the accuracy and real-time issues of carbon emission calculation for new energy vehicles have been resolved, enabling more precise carbon emission management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- RICHFIT INFORMATION TECH
- Filing Date
- 2024-12-05
- Publication Date
- 2026-06-05
AI Technical Summary
Existing methods for calculating carbon emissions from new energy vehicles rely on models of conventional energy vehicles, failing to consider the unique characteristics of new energy vehicles. This results in discrepancies between the calculation results and actual conditions, lacking accuracy and real-time performance.
By acquiring historical and real-time operational data of new energy vehicles, target characteristics associated with carbon emissions are identified, a targeted carbon emission determination model is constructed, and calculations are performed using real-time data.
It improves the accuracy and real-time performance of carbon emission calculations for new energy vehicles, ensuring that the model is closely related to the actual operating status of the vehicles, and facilitating the management and understanding of environmental impact.
Smart Images
Figure CN122153679A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of environmental protection technology, and in particular to a method, apparatus, equipment and storage medium for determining carbon emissions. Background Technology
[0002] With the severe challenges posed by global climate change, accurately determining carbon emissions is crucial for promoting environmental protection and achieving carbon peaking and carbon neutrality goals. While carbon emissions from conventional energy vehicles can usually be calculated directly using their fuel consumption, calculating the actual carbon emissions of new energy vehicles is more difficult because these vehicles themselves do not necessarily generate carbon emissions directly.
[0003] Related technologies and methods for calculating carbon emissions from new energy vehicles typically rely directly on existing carbon emission calculation models for conventional energy vehicles. However, these models suffer from issues such as poor real-time performance, low accuracy, and low correlation with the actual operating status of new energy vehicles, making it difficult for carbon emission data to accurately reflect the actual situation. Summary of the Invention
[0004] This disclosure provides a method, apparatus, device, and storage medium for determining carbon emissions, in order to address the problems of insufficient accuracy and poor reliability in related technologies for determining carbon emissions from new energy vehicles.
[0005] In a first aspect, embodiments of this disclosure provide a method for determining carbon emissions, the method comprising:
[0006] Acquire historical and real-time operating data of the vehicle to be analyzed. Historical operating data includes historical energy consumption records.
[0007] Identify target features from historical operational data that are associated with the carbon emission data of the vehicle to be analyzed;
[0008] Based on the historical operating data corresponding to the target characteristics, a carbon emission determination model is determined for the vehicle to be analyzed.
[0009] Real-time operational data is input into the carbon emission determination model, which outputs carbon emission data for the vehicle to be analyzed.
[0010] Secondly, embodiments of this disclosure provide a carbon emission determination apparatus, the data processing of which includes:
[0011] The acquisition module is used to acquire historical and real-time operating data of the vehicle to be analyzed. The historical operating data includes historical records of energy consumption.
[0012] The analysis module is used to identify target features that are associated with the carbon emission data of the vehicle to be analyzed from historical operating data;
[0013] The determination module is used to determine the carbon emission determination model for the vehicle to be analyzed based on the historical operating data corresponding to the target characteristics.
[0014] The calculation module inputs real-time operating data into the carbon emission determination model and outputs carbon emission data of the vehicle to be analyzed.
[0015] Thirdly, embodiments of this disclosure also provide a control device, which includes:
[0016] At least one processor;
[0017] and memory that is communicatively connected to at least one processor;
[0018] The memory stores instructions that can be executed by at least one processor to cause the control device to perform the carbon emission determination method as described in the first aspect of this disclosure.
[0019] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the carbon emission determination method as described in the first aspect of this disclosure.
[0020] Fifthly, embodiments of this disclosure also provide a computer program product comprising computer execution instructions, which, when executed by a processor, are used to implement the carbon emission determination method as described in the first aspect of this disclosure.
[0021] The carbon emission determination method, apparatus, device, and storage medium provided in this disclosure acquire historical and real-time operating data of the vehicle to be analyzed. Then, target features associated with the carbon emission data of the vehicle to be analyzed are determined from the historical operating data. Based on the historical operating data corresponding to the target features, a carbon emission determination model corresponding to the vehicle to be analyzed is determined. Finally, the real-time operating data is input into the carbon emission determination model, and the carbon emission data of the vehicle to be analyzed is output. Therefore, by combining historical and real-time operating data of new energy vehicles to jointly determine carbon emission data, the problems of low real-time performance, accuracy, and correlation with vehicle operating status in existing carbon emission estimation methods are solved. By extracting target features from historical data and establishing a targeted carbon emission model for the vehicle to be analyzed, the model is ensured to be closely related to the actual operating status of the vehicle to be analyzed, thereby improving the real-time performance and accuracy of the data. This facilitates relevant personnel to better understand and manage the environmental impact of new energy vehicles and to better manage the carbon emissions of new energy vehicles. Attached Figure Description
[0022] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.
[0023] Figure 1 A diagram illustrating an application scenario of the carbon emission determination method provided in this disclosure.
[0024] Figure 2 A flowchart illustrating a carbon emission determination method provided in one embodiment of this disclosure;
[0025] Figure 3a A flowchart of a carbon emission determination method provided in yet another embodiment of this disclosure;
[0026] Figure 3b for Figure 3a The flowchart of the carbon emission determination model provided in the embodiment shown is as follows;
[0027] Figure 4 A schematic diagram of the structure of a carbon emission determination device provided in yet another embodiment of this disclosure;
[0028] Figure 5 This is a schematic diagram of the structure of a control device provided in one embodiment of the present disclosure.
[0029] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0031] The technical solutions of this disclosure and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this disclosure will now be described with reference to the accompanying drawings.
[0032] Analyzing carbon emissions from conventional energy vehicles is a relatively mature technological solution, but there are significant shortcomings in calculating carbon emissions from new energy vehicles. On the one hand, new energy vehicles typically do not directly consume fuel, or consume very little fuel, and are easily assumed to have no carbon emissions (but in reality, the different forms of energy they consume, combined with the energy collection and conversion processes, all involve some carbon emissions). On the other hand, existing methods for calculating carbon emissions from new energy vehicles often rely on carbon emission models for conventional energy vehicles, directly using fuel consumption data for substitution calculations. This method fails to consider the unique characteristics of new energy vehicles (such as battery efficiency and charging methods), leading to discrepancies between the calculated results and actual conditions.
[0033] The diverse types of new energy vehicles, each with varying operating characteristics and energy consumption patterns, increase the difficulty of establishing a universal model. Furthermore, the diverse types of operational data associated with new energy vehicles present challenges in selecting and effectively processing this data to reflect their carbon emissions. This complexity also necessitates considering the influence of varying vehicle operating environments (such as climate and terrain) on carbon emissions, further complicating model design and validation. Consequently, related technologies lack methods for accurately determining the carbon emissions of new energy vehicles.
[0034] To address this issue, this disclosure provides a carbon emission determination method. By combining a single type of new energy vehicle and analyzing its historical and real-time operating data, target features associated with carbon emission data are determined. These target features are then used to generate carbon emission data corresponding to the vehicle to be analyzed. Combined with real-time operating data and a carbon emission determination model, the corresponding carbon emissions are calculated. This method significantly improves the accuracy of carbon emission calculation for new energy vehicles while ensuring the specificity of the analysis for the vehicle.
[0035] The application scenarios of the embodiments of this disclosure are explained below:
[0036] Figure 1 This diagram illustrates an application scenario of the carbon emission determination method provided in this embodiment of the disclosure. Figure 1 As shown, in the process of determining carbon emissions, server 100 receives real-time operating data uploaded by new energy vehicle 110 and combines it with historical operating data in historical database 101 to determine the corresponding carbon emissions.
[0037] It should be noted that, Figure 1 In the scenario shown, the server, new energy vehicle, and historical database are only used as examples of one or a specific number, but this disclosure is not limited to this. That is to say, the number of servers, new energy vehicles, and historical databases can be arbitrary.
[0038] The carbon emission determination method provided in this disclosure is described in detail below through specific embodiments.
[0039] Figure 2 A flowchart illustrating a carbon emission determination method provided in one embodiment of this disclosure. Figure 2 As shown, the carbon emission determination method provided in this embodiment includes the following steps:
[0040] Step S201: Obtain historical and real-time operating data of the vehicle to be analyzed.
[0041] Historical operating data includes historical energy consumption records.
[0042] Specifically, this embodiment describes the overall process of determining carbon emission data for new energy vehicles.
[0043] The execution entity of this embodiment is a server capable of collecting data generated during vehicle operation. It can be a server cluster of a data center used for overall analysis of various new energy vehicles, or it can be the manufacturer's server corresponding to the vehicle.
[0044] In carbon emission calculations, servers need to collect historical and real-time operating data from new energy vehicles. Historical operating data includes the vehicle's (all new energy vehicles) driving status data and energy consumption records over a period of time (such as the past year). Driving status data includes speed, acceleration, and real-time location, while energy consumption records include battery charging cycles, charging amount, and mileage.
[0045] Real-time operating data refers to the vehicle's current operating status information, such as current speed, acceleration, and real-time energy consumption.
[0046] If the amount of historical operating data for a specific vehicle to be analyzed is insufficient, it can be combined with the historical operating data of other new energy vehicles of the same type and model (such as other vehicles of the same model produced in the same batch) in the corresponding specific area for joint analysis.
[0047] Both historical and real-time operational data can be collected through the vehicle's built-in sensors and data recording devices and uploaded to a server database. Historical operational data can also be extracted from the cloud server database corresponding to the vehicle being analyzed, while real-time operational data is primarily collected in real time through onboard sensors.
[0048] Real-time data transmission can be carried out via wireless network to the server to ensure the timeliness and integrity of the data.
[0049] Therefore, by combining historical and real-time data, it is possible to ensure that the target features used to determine the model more comprehensively reflect the vehicle's operating mode and energy consumption characteristics, thus providing a foundation for the subsequent construction of carbon emission models.
[0050] Step S202: Determine target features associated with the carbon emission data of the vehicle to be analyzed from historical operating data.
[0051] Specifically, since the correlation between carbon emissions and vehicle operating characteristics varies among different types of vehicles to be analyzed, it is necessary to extract key features related to carbon emissions from a large amount of historical data. For example, the correlation between carbon emissions of hybrid vehicles and their driving range is higher than that between carbon emissions of pure electric vehicles and their driving range (because the electrical energy of pure electric vehicles is more likely to be consumed through external battery discharge devices when the vehicle is not in motion, such as charging external devices, while hybrid vehicles may not support this function, or the electrical energy that can be used for consumption is smaller, and the correlation with overall carbon emissions is lower).
[0052] These key characteristics (i.e. target characteristics) may include battery efficiency, charging frequency, energy consumption patterns, etc., which can reflect the energy consumption characteristics and carbon emission potential of a vehicle under different operating conditions.
[0053] The specific target features can be determined using data mining and machine learning techniques. Various driving status data and their correlation with energy consumption can be identified from historical data. By selecting appropriate feature selection algorithms (such as various filtering algorithms, packaging algorithms, and embedding algorithms, which are not limited in this embodiment), the most representative target features can be screened from historical operating data, reducing data dimensionality and enabling more accurate capture of key factors affecting carbon emissions, thereby improving the accuracy and efficiency of the model.
[0054] Step S203: Based on the historical operating data corresponding to the target characteristics, determine the carbon emission determination model corresponding to the vehicle to be analyzed.
[0055] Specifically, by using historical operational data corresponding to the target features as training data, the server constructs a corresponding carbon emission determination model for the vehicle to be analyzed, thereby predicting the vehicle's carbon emissions based on the target features. Furthermore, the calculation of carbon emissions comprehensively considers the unique characteristics of new energy vehicles, such as battery efficiency and charging methods, to ensure the accuracy of the carbon emission calculation results.
[0056] Specific model construction can employ various modeling techniques, such as regression analysis, neural networks, or decision trees, and be trained based on target features and historical data. Model selection and parameter adjustment require historical operational data corresponding to the target features, enabling optimization based on the specific characteristics of the vehicle being analyzed and the distribution of the data.
[0057] Therefore, by constructing a targeted carbon emission determination model, the server can more accurately predict the carbon emissions of vehicles under different operating conditions, thereby improving the reliability and accuracy of the calculation results.
[0058] Step S204: Input the real-time operating data into the carbon emission determination model and output the carbon emission data of the vehicle to be analyzed.
[0059] Specifically, after the carbon emission determination model is determined, the server can input the real-time vehicle operation data (i.e., real-time operation data) into the trained carbon emission determination model to accurately calculate the current carbon emissions of the vehicle to be analyzed.
[0060] Therefore, by inputting real-time operational data, the server can dynamically monitor the carbon emissions of the vehicles being analyzed, promptly reflecting the vehicle's operating status and environmental changes, ensuring the real-time nature and accuracy of carbon emission data. Furthermore, by recording the carbon emissions of each vehicle, the server can manage the overall carbon emissions of new energy vehicles, thereby contributing to promoting environmental protection and achieving carbon peaking and carbon neutrality goals.
[0061] The carbon emission determination method provided in this disclosure acquires historical and real-time operating data of the vehicle to be analyzed. Then, it identifies target features associated with the carbon emission data of the vehicle from the historical operating data. Based on the historical operating data corresponding to the target features, it determines a carbon emission determination model for the vehicle. Finally, it inputs the real-time operating data into the carbon emission determination model and outputs the carbon emission data of the vehicle. Therefore, by combining historical and real-time operating data of new energy vehicles to jointly determine carbon emission data, it solves the problems of low real-time performance, accuracy, and correlation with vehicle operating status in existing carbon emission estimation methods. By extracting target features from historical data and establishing a targeted carbon emission model for the vehicle to be analyzed, it ensures that the model is closely related to the actual operating status of the vehicle, thereby improving the real-time performance and accuracy of the data. This facilitates a better understanding and management of the environmental impact of new energy vehicles and enables better management of their carbon emissions.
[0062] Figure 3a A flowchart illustrating a carbon emission determination method provided in one embodiment of this disclosure. Figure 3a As shown, the carbon emission determination method provided in this embodiment includes the following steps:
[0063] Step S301: Obtain historical and real-time operating data of the vehicle to be analyzed.
[0064] The historical operating data includes historical records of energy consumption; the historical operating data includes historical records of the driving status of the vehicle to be analyzed, including the vehicle's position, speed, and distance traveled; the real-time operating data includes the real-time driving status of the vehicle to be analyzed, and the data types in the real-time driving status are the same as those in the historical driving status.
[0065] Specifically, this embodiment is used to further explain the detailed process of the carbon emission determination method.
[0066] In this embodiment, the server can use the high-precision positioning and timing functions of the BeiDou Navigation Satellite System to collect real-time operating data of new energy vehicles, including location, speed, acceleration, and energy consumption. It also collects historical operating data, including historical records of energy consumption and driving status, through a cloud server database. The driving status includes location, speed, and distance traveled.
[0067] The BeiDou Navigation Satellite System can efficiently collect real-time and historical vehicle data, ensuring the accuracy and timeliness of the data. This provides a reliable data foundation for the construction of carbon emission models and solves the problem of inaccurate data in traditional methods.
[0068] Step S302: Preprocess the historical operation data.
[0069] The preprocessing includes removing outlier data and data standardization. Outlier data includes data with format errors and data with numerical errors.
[0070] Specifically, preprocessing includes removing outlier data and data standardization. Outlier data includes data with format errors and data with numerical errors. Data cleaning removes invalid or erroneous data records, and data standardization is performed to eliminate the influence of different units and magnitudes.
[0071] This improves the quality and consistency of the data used for subsequent analysis, ensures the accuracy of the analysis, and solves the problems of data noise and inconsistency.
[0072] Step S303: Select the data of the specified type from the preprocessed historical operating data as candidate data related to the energy consumption of the vehicle to be analyzed.
[0073] Specifically, the preprocessed historical operational data includes multiple different types. By selecting data related to energy consumption as candidate data, the foundation for subsequent feature extraction is laid, which improves the relevance and effectiveness of the model and solves the problem of inappropriate data selection in traditional methods.
[0074] The specific data types corresponding to the alternative data can be pre-configured by the analysts to initially eliminate data types that are not related to energy consumption, such as remote door unlocking records and interior light colors, and reduce the amount of data required for subsequent analysis.
[0075] Step S304: Based on the correlation between candidate data and energy consumption, select data from the candidate data whose correlation with energy consumption is higher than the set correlation threshold, and use it as target data.
[0076] Specifically, by applying techniques such as logarithmic transformation, feature combination, normalization, and linear discriminant analysis, features are extracted and transformed from the candidate data to further identify data with high correlation to energy consumption as target data.
[0077] Therefore, through meticulous feature engineering, the predictive and generalization abilities of the model are improved, solving the problem of poor feature selection in traditional methods.
[0078] In some embodiments, the process of processing candidate data to obtain target data may include the following steps:
[0079] Step A1: Perform a logarithmic transformation on the candidate data, and then combine the different types of data after the logarithmic transformation in pairs to form derived feature data.
[0080] Specifically, logarithmic transformation of selected features can reduce data skewness or nonlinearity.
[0081] Based on this, features such as travel time and rate of change of acceleration are combined to derive new features from existing features, enrich the types of data, and more comprehensively reflect the correlation between various data and carbon emissions.
[0082] Logarithmic transformation makes the data distribution more normally distributed, which helps improve the linear separability of the model. Combined features enable the model to capture more complex relationships.
[0083] Step A2: Encode the derived feature data and alternative data based on the type of data.
[0084] Specifically, feature classification is labeled and encoded, converting various data features into numerical forms to facilitate model processing. This improves data processability, ensures that the model can utilize all features, and solves the problem of improper handling of classification features in traditional methods.
[0085] Step A3: Normalize the encoded data.
[0086] Specifically, by using normalization techniques, features are scaled to a uniform range or distribution to eliminate the influence of different dimensions, ensuring that all features are compared on the same scale, avoiding model bias caused by scale differences, and improving the stability and accuracy of the model. This solves the prediction error problem caused by different feature scales in traditional methods.
[0087] Step A4: Based on the linear discriminant analysis algorithm, determine the correlation between the encoded data and energy consumption.
[0088] Specifically, by applying Linear Discriminant Analysis (LDA), we can identify whether the changes in various types of data have a linear relationship with the changes in energy consumption (indicating a strong correlation). This removes redundant feature data, reduces the number of features, eliminates redundancy and noise, improves the model's generalization ability, and reduces computational complexity. It also solves the overfitting problem caused by feature redundancy in traditional methods.
[0089] Step A5: Remove encoded data whose relevance is lower than the set relevance threshold.
[0090] Specifically, by using the LDA algorithm, features with low correlation to energy consumption are eliminated, ensuring that the model uses only the most relevant data, which improves the efficiency and accuracy of the model and solves the problem of model complexity caused by low-correlation features in traditional methods.
[0091] Step A6: Determine the interrelationships among the data after the data removal process, and generate a decision tree model based on the interrelationships.
[0092] Specifically, based on the aforementioned elimination, interaction terms between the remaining features can be constructed to increase the variety of data. Through the decision tree model, complex interaction relationships between features can be identified, thereby enhancing the interpretability of the carbon emission determination model, better capturing complex feature relationships, and solving the problem that some interaction relationships between feature data are not identified in traditional methods.
[0093] Step A7: Based on the decision tree model, determine the number of data points of the specified categories that have the highest correlation with carbon emissions, and use them as target data.
[0094] Specifically, by applying the feature importance of the tree model to select the most influential (set number) features (such as three or five), the predictive power of the carbon emission determination model is improved, and the model is ensured to focus on the most important features.
[0095] By combining the aforementioned steps, multiple filtering and reconstruction of historical operational data are achieved, maximizing the mining of the relationship between various data and carbon emissions in historical operational data, thus solving the problem of inappropriate feature selection in traditional methods.
[0096] Step S305: Use the category corresponding to the target data as the target feature.
[0097] Specifically, the data corresponding to the target features can be determined through the aforementioned steps.
[0098] To facilitate subsequent model training, additional target data can be added if the amount of data is insufficient.
[0099] In some embodiments, the method for supplementing target data with insufficient data volume includes the following steps:
[0100] Step B1: If the amount of at least one type of target data is less than the average amount of all types of data in the target feature, then at least one type of target data is identified as data to be supplemented.
[0101] Specifically, by assessing the amount of data, we can identify the data that needs to be supplemented, ensure that the performance of features is consistent across different datasets or time points, improve the robustness of the model, ensure the integrity of the data, and solve the problem of data imbalance in traditional methods.
[0102] Step B2: Add the data to be supplemented to the automated feature generation library and output the corresponding simulation data.
[0103] Specifically, for this type of data that needs to be supplemented, the Featuretools automated feature engineering library can be used to generate features. The data to be supplemented and its related data can be input into the automated feature engineering library to generate corresponding simulation data, which can be used as feature data to supplement the corresponding data of the target feature. This improves the completeness of the data and the accuracy of the model, and solves the problem of insufficient data in traditional methods.
[0104] In some embodiments, once simulation data is obtained, the correlation between the simulation data and carbon emissions can be calculated. If the correlation between the simulation data and the correlation characteristics between the same type of supplementary data (such as the calculation results of linear discriminant analysis) differ, the difference information and the simulation data can be combined and input into an automated feature generation library to iteratively optimize the generation of feature data and ensure the usability of the generated simulation data.
[0105] Step B3 combines the simulation data with the corresponding types of supplementary data to be used as feature data for training the carbon emission determination model.
[0106] Specifically, by supplementing the simulation data, the integrity and diversity of the model training data are ensured, which improves the training effect and prediction ability of the model and solves the model bias problem caused by incomplete data in traditional methods.
[0107] Step S306: Determine the carbon emission factor corresponding to the vehicle to be analyzed based on the type of vehicle to be analyzed.
[0108] Specifically, by combining the vehicle's fuel type and energy consumption characteristics, a suitable carbon emission factor can be determined. For example, the carbon emission factor of hybrid vehicles is higher than that of pure electric vehicles. For instance, the carbon emission factor of gasoline consumed by new energy vehicles is 2.31 kg CO2 / L, and the carbon emission factor of electricity consumed can be 0.5 kg CO2 / kWh. By combining the fuel type of the vehicle to be analyzed and the proportion of various energy consumption in historical operating data, the corresponding carbon emission factor can be determined.
[0109] By specifically identifying carbon emission factors, the accuracy of carbon emission calculations has been improved, solving the problem of inaccurate factors in traditional methods.
[0110] Step S307: Based on the carbon emission factor and the energy consumption data in the historical operation data, determine the carbon emission data corresponding to the historical operation data.
[0111] Specifically, by combining historical energy consumption data and carbon emission factors, the historical carbon emissions of vehicles can be calculated, providing a basis for subsequent model training, validation, and optimization, thus solving the problem of lack of historical data support in traditional methods.
[0112] Step S308: Input the historical operating data corresponding to the target feature into at least two predetermined carbon emission estimation models, and output the first predicted carbon emission data corresponding to the carbon emission estimation models respectively.
[0113] Carbon emission estimation models include physical models, empirical models, and / or data-driven models.
[0114] Specifically, by using multiple models to make predictions together, the adaptability and accuracy of the models can be enhanced, solving the problem of inaccurate predictions by a single model.
[0115] For example, the physical model can be: carbon emissions = energy consumption × carbon emission factor; the empirical model can be: carbon emissions = coefficient 1 + coefficient 2 × driving speed + coefficient 3 × acceleration; and the data-driven model can be expressed as: carbon emissions = f(speed, acceleration, distance, energy consumption).
[0116] As can be seen from the examples above, the physical model can only simply calculate the carbon emissions corresponding to vehicle energy consumption. It cannot accurately analyze the changes in carbon emissions under different driving conditions, and its analysis granularity is the coarsest, but its calculation efficiency is the highest. The empirical model is relatively accurate in reflecting the carbon emissions of the vehicle under driving conditions, and its analysis granularity is moderate, but it cannot reflect the energy consumption of the vehicle under stationary conditions. The data-driven model considers more types of data and reflects the relationship between carbon emissions and different data more accurately.
[0117] In some embodiments, the actual carbon emission estimation model that can be selected may be adjusted based on the types of data in the defined target features, and is not limited to the types of data in the examples above.
[0118] Step S309: Based on the deviation between the first predicted energy consumption data and the corresponding carbon emission data, train each carbon emission estimation model.
[0119] Specifically, once the carbon emission estimation model is determined, it can be trained and optimized. Through model training, prediction errors are reduced and the reliability of the model is improved.
[0120] Step S310: Based on the trained carbon emission estimation model, determine the carbon emission determination model.
[0121] Specifically, as the analysis above shows, different types of models have their own focuses. Therefore, it is necessary to combine methods such as cluster analysis and model weight adjustment to determine the final carbon emission estimation model through the carbon emission estimation model that has been trained.
[0122] In some embodiments, such as Figure 3b The diagram shows the flowchart for determining the carbon emission determination model. The specific steps for determining the carbon emission model include:
[0123] Step S3101: Perform cluster analysis on the preprocessed historical running data, and use each cluster center obtained from the cluster analysis as the driving feature dataset corresponding to different driving modes.
[0124] Specifically, by clustering and analyzing different types of vehicle driving data (or by directly selecting historical operating data corresponding to the target feature), different driving modes, such as acceleration, constant speed, deceleration, and idling, can be identified. This allows for a more refined analysis granularity of the model, ensuring that the model can identify and process different driving modes and more accurately calculate the model's carbon emissions.
[0125] Step S3102: Determine the carbon emission data corresponding to the energy consumption data of each driving feature dataset, and determine the target data corresponding to the driving feature dataset with the highest average carbon emission.
[0126] Specifically, by analyzing energy consumption and carbon emissions, the energy consumption under different driving modes is assessed, and the mode with the highest energy consumption is identified to ensure that the model can identify high-energy-consuming modes.
[0127] Step S3103: Input the target data corresponding to the driving feature dataset with the highest average carbon emission into each trained carbon emission estimation model and output the corresponding second predicted energy consumption data.
[0128] Specifically, different models are used to calculate carbon emissions under the highest energy consumption mode, and the results are compared to select the carbon emission estimation model that can accurately predict carbon emissions under the high energy consumption mode. The weight of the model is increased to enhance the ability of the carbon emission determination model to accurately predict carbon emissions under the high energy consumption mode.
[0129] Step S3104: Determine the deviation between the second predicted energy consumption data corresponding to each trained carbon emission estimation model and the carbon emission data corresponding to the driving feature dataset with the highest average carbon emission.
[0130] Specifically, through error analysis, the strengths and weaknesses of various carbon emission estimation models in handling carbon emissions under high-energy-consumption models can be identified, so as to optimize the models.
[0131] Step S3105: Based on the deviation value, determine the weight coefficients of each trained carbon emission estimation model.
[0132] Specifically, error correction (weight adjustment) is performed based on the difference between model predictions and actual measurements to optimize the model's predictive ability and ensure that the contributions of different carbon emission estimation models are reasonably allocated, thereby improving the prediction accuracy of the comprehensive carbon emission determination model.
[0133] Step S3106: Combine the trained carbon emission estimation model with the weighting coefficients to obtain the carbon emission determination model corresponding to the vehicle to be analyzed.
[0134] Specifically, by combining the trained carbon emission estimation model with the weighting coefficients obtained in the aforementioned steps, a comprehensive carbon emission prediction model can be formed, ensuring the accuracy and reliability of the prediction.
[0135] Optionally, the above steps only use the driving mode with the highest carbon emissions to adjust the weight coefficients of different carbon emission estimation models. In practical applications, the corresponding weight coefficients can be determined for each different driving mode, and a driving mode recognition model can be established to identify driving modes based on real-time operating data. Then, the corresponding weight coefficients can be selected based on the driving mode identified by the driving mode recognition model to calculate carbon emissions, thereby improving the accuracy of the calculation (but this will significantly increase the amount of calculation, so those skilled in the art need to consider whether to choose this approach based on the actual situation).
[0136] Step S311: Preprocess the real-time running data.
[0137] The preprocessing includes removing outlier data and data standardization. Outlier data includes data with format errors and data with numerical errors.
[0138] Specifically, when calculating real-time carbon emissions, it is necessary to first remove outliers and standardize the real-time operating data to ensure the real-time performance and accuracy of the model predictions.
[0139] Step S312: Input the preprocessed real-time operating data into the carbon emission determination model and output the carbon emission data of the vehicle to be analyzed.
[0140] Specifically, the model calculates and outputs the vehicle's carbon emissions in real time, providing dynamic monitoring and management of vehicle carbon emissions, thus solving the problem of lack of real-time performance in traditional methods.
[0141] The carbon emission determination method provided in this disclosure utilizes the BeiDou Navigation Satellite System to collect high-precision historical and real-time vehicle operating data, providing a reliable data foundation for the model. Data preprocessing and feature extraction remove outliers, and standardization eliminates the influence of different units of measurement, ensuring data consistency. Feature engineering enhances the model's predictive ability, while the combination of linear discriminant analysis and decision tree models optimizes feature selection and model structure. Automated feature generation and simulation data supplementation enhance data integrity and model robustness. Cluster analysis identifies energy consumption characteristics under different driving modes, ensuring the model can accurately predict carbon emissions under high-energy-consumption modes. Multi-model combination and weight adjustment improve the model's comprehensive predictive ability, and error analysis and correction further optimize model accuracy. Finally, the system dynamically outputs vehicle carbon emission data through real-time data input and model calculation, providing a scientific basis for policy formulation and user behavior guidance.
[0142] Based on the carbon emission data obtained in the above embodiments, since the obtained carbon emission data can be accurately correlated with factors such as the driving status, vehicle type, and energy type of the vehicle to be analyzed, the carbon emission data can be effectively applied.
[0143] For example, carbon emission data of the vehicle to be analyzed can be recorded, and then a carbon emission chart of the vehicle to be analyzed can be generated and output based on the set chart type.
[0144] Specifically, the server can record carbon emission data and display it in chart form to analyze trends in carbon emissions over time and comparisons of emissions under different driving modes. It can also generate detailed analysis reports to provide a basis for policy making, vehicle design improvements (to optimize carbon emissions under different modes), and user behavior guidance (such as how to operate the vehicle to save energy).
[0145] In some embodiments, the collected carbon emissions and vehicle operation data can also be used to inventory and identify carbon emission-related resources, and to develop a unified data format and standard.
[0146] Specifically, by comprehensively reviewing various types of data, identifying datasets with potential value, and establishing unified formats and standards, the consistency and usability of the data are ensured, providing data support for the subsequent optimization of energy consumption in new energy vehicles. This allows for the determination of carbon emissions processes, improving the efficiency and accuracy of data management, laying the foundation for subsequent data assetization, and solving the management difficulties caused by inconsistent data formats and unclear data characteristics in traditional methods.
[0147] In some embodiments, since the carbon emission determination process involves the operating data of the vehicle to be analyzed and other vehicles of the same type, it is also necessary to clarify the rights and interests of the data asset owners, users and managers to ensure the legal and compliant use of the data asset. This can ensure the legality and security of the data asset, reduce the risks in the data assetization process, and solve the problem of non-compliant data use in traditional methods.
[0148] In some embodiments, because the calculation of carbon emissions is relatively accurate, the carbon emission estimates can be transformed into quantifiable data asset indicators, such as carbon emission rights and carbon credits per vehicle. This transforms carbon emission data into specific economic indicators, giving it practical value in the carbon trading market. It also realizes the economic value of carbon emission data, promotes the quantitative management of carbon emission reduction, and solves the problem that traditional methods make carbon emission data difficult to apply directly.
[0149] In some embodiments, for various data in the process of determining various carbon emissions, data assets can also be registered on a data asset platform or system to record detailed information and transaction history of data assets, thereby improving the transparency and transaction efficiency of data assets, providing new business models and market opportunities for the new energy vehicle industry, and solving the problem of opaque data asset management in traditional methods.
[0150] In summary, the carbon emission determination method provided by this invention achieves comprehensive analysis of new energy vehicle operation data and accurate estimation of carbon emissions through multi-dimensional data acquisition, intelligent data processing, and precise carbon emission calculation. By utilizing the high-precision positioning and time synchronization functions of the BeiDou Navigation Satellite System, it can accurately track vehicle status. Furthermore, it utilizes big data analytics to store, process, and analyze massive amounts of data, and combines this with machine learning algorithms to further optimize the carbon emission determination model, improving the accuracy and reliability of carbon emission estimation. Simultaneously, this solution possesses an adaptive adjustment mechanism, enabling continuous performance optimization based on real-time data and feedback.
[0151] Furthermore, through data asset transformation, carbon emission data is converted into quantifiable data assets, supporting carbon trading and environmental regulation. Data visualization facilitates user management and querying of carbon emission data and assets. Accurate determination of carbon emission data ensures future functional expansion and upgrade capabilities, adapting to future technological development needs. The integrated nature of the overall solution allows for seamless integration with existing vehicle monitoring systems and carbon trading platforms, providing a unified solution. Therefore, this invention not only improves the efficiency and accuracy of carbon emission management for new energy vehicles but also provides new business models and market opportunities for the development of a low-carbon economy.
[0152] Figure 4 This is a schematic diagram of a carbon emission determination device provided in one embodiment of the present disclosure. Figure 4 As shown, the carbon emission determination device 400 includes: an acquisition module 410, an analysis module 420, a determination module 430, and a calculation module 440. Wherein:
[0153] The acquisition module 410 is used to acquire historical and real-time operating data of the vehicle to be analyzed. The historical operating data includes historical records of energy consumption.
[0154] Analysis module 420 is used to determine target features associated with the carbon emission data of the vehicle to be analyzed from historical operating data;
[0155] The determination module 430 is used to determine the carbon emission determination model corresponding to the vehicle to be analyzed based on the historical operating data corresponding to the target characteristics.
[0156] The calculation module 440 inputs real-time operating data into the carbon emission determination model and outputs carbon emission data of the vehicle to be analyzed.
[0157] Optionally, the acquisition module 410 specifically includes: historical operating data including historical records of the driving status of the vehicle to be analyzed, the driving status including the position, speed and driving distance of the vehicle to be analyzed; and real-time operating data including the real-time driving status of the vehicle to be analyzed, wherein the data types in the real-time driving status are the same as the data types in the historical driving status.
[0158] Optionally, the analysis module 420 is specifically used to: preprocess historical operating data, including removing outlier data and data standardization, whereby outlier data includes data with format errors and data with numerical errors; identify data of a set category from the preprocessed historical operating data as candidate data related to the energy consumption of the vehicle to be analyzed; based on the correlation between the candidate data and energy consumption, identify data from the candidate data whose correlation with energy consumption is higher than a set correlation threshold as target data; and use the category corresponding to the target data as target features.
[0159] Optionally, the analysis module 420 is specifically used to: perform logarithmic transformation on the candidate data, and combine the different types of data after logarithmic transformation in pairs as derived feature data; encode the derived feature data and candidate data based on the types of data; normalize the encoded data; determine the correlation between the encoded data and energy consumption based on the linear discriminant analysis algorithm; remove the encoded data whose correlation is lower than a set correlation threshold; determine the mutual correlation between the removed data and generate a decision tree model based on the mutual correlation; and determine the set number of data types with the highest correlation to carbon emissions based on the decision tree model as target data.
[0160] Optionally, the determining module 430 is specifically used to: determine the carbon emission factor corresponding to the vehicle to be analyzed based on the type of the vehicle to be analyzed; determine the carbon emission data corresponding to the historical operating data based on the carbon emission factor and the energy consumption data in the historical operating data; input the historical operating data corresponding to the target feature into at least two pre-determined carbon emission estimation models, and output the first predicted carbon emission data corresponding to the carbon emission estimation models respectively, wherein the carbon emission estimation models include physical models, empirical models and / or data-driven models; train each carbon emission estimation model based on the deviation between the first predicted energy consumption data and the corresponding carbon emission data; and determine the carbon emission determination model based on the trained carbon emission estimation models.
[0161] Optionally, the determination module 430 is specifically used to: perform cluster analysis on the preprocessed historical operating data, and use each cluster center obtained from the cluster analysis as a driving feature dataset corresponding to different driving modes; determine the carbon emission data corresponding to the energy consumption data of each driving feature dataset, and determine the target data corresponding to the driving feature dataset with the highest average carbon emission; input the target data corresponding to the driving feature dataset with the highest average carbon emission into each trained carbon emission estimation model, and output the corresponding second predicted energy consumption data; determine the deviation value between the second predicted energy consumption data corresponding to each trained carbon emission estimation model and the carbon emission data corresponding to the driving feature dataset with the highest average carbon emission; determine the weight coefficient of each trained carbon emission estimation model based on the deviation value; and combine the trained carbon emission estimation model with the weight coefficient to obtain the carbon emission determination model corresponding to the vehicle to be analyzed.
[0162] Optionally, the calculation module 440 is specifically used to preprocess the real-time operating data, including removing outlier data and data standardization, whereby outlier data includes data with format errors and data with numerical errors; input the preprocessed real-time operating data into the carbon emission determination model, and output the carbon emission data of the vehicle to be analyzed.
[0163] Optionally, the analysis module 420 is further configured to, after taking the type corresponding to the target data as the target feature, if the amount of data of at least one type of target data is less than the average amount of data of all types of data in the target feature, determine at least one type of target data as data to be supplemented; add the data to be supplemented to the automated feature generation library and output the corresponding simulation data; combine the simulation data with the data to be supplemented of the corresponding type as feature data for training the carbon emission determination model.
[0164] Optionally, the calculation module 440 is also used to input real-time running data into the carbon emission determination model, output the carbon emission data of the vehicle to be analyzed, record the carbon emission data of the vehicle to be analyzed, and generate and output the carbon emission chart of the vehicle to be analyzed based on the set chart type.
[0165] In this embodiment, the carbon emission determination device, through the combination of various modules, solves the problems of insufficient accuracy and poor reliability of carbon emission determination methods for new energy vehicles in related technologies. It ensures that the model is closely related to the actual operating status of the vehicle to be analyzed, thereby improving the real-time performance and accuracy of the data. This makes it easier for relevant personnel to better understand and manage the environmental impact of new energy vehicles and to better manage the carbon emissions of new energy vehicles.
[0166] Figure 5 This is a schematic diagram of the structure of a control device provided in one embodiment of the present disclosure, as shown below. Figure 5 As shown, the control device 500 includes a memory 510 and a processor 520.
[0167] The memory 510 stores a computer program that can be executed by at least one processor 520. This computer program is executed by at least one processor 520 to enable the control device to implement the material removal method or the carbon emission determination method provided in any of the above embodiments.
[0168] The memory 510 and the processor 520 can be connected via a bus 530.
[0169] The relevant explanations can be understood by referring to the corresponding descriptions and effects in the method embodiments, and will not be repeated here.
[0170] The relevant explanations can be understood by referring to the corresponding descriptions and effects in the method embodiments, and will not be repeated here.
[0171] One embodiment of this disclosure provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the material removal method as provided in any of the above method embodiments or the carbon emission determination method as provided in any of the above embodiments.
[0172] The computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.
[0173] One embodiment of this disclosure provides a computer program product comprising computer-executable instructions that, when executed by a processor, are used to implement a material removal method as described in the above method embodiments or a carbon emission determination method as provided in any of the foregoing embodiments.
[0174] In the several embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.
[0175] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.
[0176] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.
Claims
1. A method for determining carbon emissions, characterized in that, include: Acquire historical and real-time operating data of the vehicle to be analyzed, wherein the historical operating data includes historical records of energy consumption; Determine target features associated with the carbon emission data of the vehicle to be analyzed from the historical operating data; Based on the historical operating data corresponding to the target features, a carbon emission determination model is determined for the vehicle to be analyzed. The real-time operating data is input into the carbon emission determination model, and the carbon emission data of the vehicle to be analyzed is output.
2. The method according to claim 1, characterized in that, The historical operation data includes a historical record of the driving status of the vehicle to be analyzed, which includes the vehicle's position, speed, and distance traveled. The real-time operation data includes the real-time driving status of the vehicle to be analyzed, and the data types in the real-time driving status are the same as those in the historical driving status.
3. The method according to claim 2, characterized in that, The step of determining the target features associated with the carbon emission data of the vehicle to be analyzed from the historical operating data includes: The historical operational data is preprocessed, including the removal of abnormal data and data standardization. The abnormal data includes data with format errors and data with numerical errors. Data of a specified type from the preprocessed historical operating data are identified as candidate data associated with the energy consumption of the vehicle to be analyzed. Based on the correlation between the candidate data and energy consumption, data with a correlation higher than a set correlation threshold is selected from the candidate data and used as target data. The category corresponding to the target data is used as the target feature.
4. The method according to claim 3, characterized in that, The step of determining data from the candidate data with a correlation higher than a set correlation threshold with energy consumption as target data based on the correlation between the candidate data and energy consumption includes: The candidate data is logarithmically transformed, and the different types of logarithmically transformed data are combined in pairs to form derived feature data. Based on the type of data, the derived feature data and alternative data are encoded; The encoded data is then normalized. Based on the linear discriminant analysis algorithm, the correlation between the encoded data and energy consumption is determined; Data with a correlation lower than a set correlation threshold will be removed. Determine the interrelationships among the data after the data removal process, and generate a decision tree model based on the interrelationships; Based on the decision tree model, the data of the set number of categories with the highest correlation to the carbon emissions are determined as the target data.
5. The method according to claim 4, characterized in that, The step of determining the carbon emission determination model for the vehicle to be analyzed based on the historical operating data corresponding to the target features includes: Based on the type of vehicle to be analyzed, determine the corresponding carbon emission factor of the vehicle to be analyzed; Based on the carbon emission factor and the energy consumption data in the historical operating data, determine the carbon emission data corresponding to the historical operating data; The historical operating data corresponding to the target feature is input into at least two predetermined carbon emission estimation models, and the first predicted carbon emission data corresponding to the carbon emission estimation models are output respectively. The carbon emission estimation models include physical models, empirical models and / or data-driven models. Based on the deviation between the first predicted energy consumption data and the corresponding carbon emission data, each carbon emission estimation model is trained. Based on the trained carbon emission estimation model, a carbon emission determination model is determined.
6. The method according to claim 5, characterized in that, The carbon emission determination model, based on the trained carbon emission estimation model, includes: Cluster analysis is performed on the preprocessed historical running data, and each cluster center obtained from the cluster analysis is used as a driving feature dataset corresponding to different driving modes. Determine the carbon emission data corresponding to the energy consumption data for each driving feature dataset, and determine the target data corresponding to the driving feature dataset with the highest average carbon emission. The target data corresponding to the driving feature dataset with the highest average carbon emissions is input into each trained carbon emission estimation model, and the corresponding second predicted energy consumption data is output. Determine the deviation between the second predicted energy consumption data corresponding to each trained carbon emission estimation model and the carbon emission data corresponding to the driving feature dataset with the highest average carbon emission. Based on the aforementioned deviation value, the weight coefficients of each trained carbon emission estimation model are determined; The trained carbon emission estimation model is combined with the weighting coefficients to obtain the carbon emission determination model corresponding to the vehicle to be analyzed.
7. The method according to claim 3, characterized in that, The step of inputting the real-time operating data into the carbon emission determination model and outputting the carbon emission data of the vehicle to be analyzed includes: The real-time running data is preprocessed, including the removal of abnormal data and data standardization. The abnormal data includes data with format errors and data with numerical errors. The preprocessed real-time operating data is input into the carbon emission determination model, and the carbon emission data of the vehicle to be analyzed is output.
8. The method according to any one of claims 3 to 7, characterized in that, After using the category corresponding to the target data as the target feature, the method further includes: If the amount of data for at least one target data is less than the average amount of data for all types of data in the target feature, then the at least one target data is identified as data to be supplemented. Add the data to be supplemented to the automated feature generation library and output the corresponding simulation data; The simulation data is combined with the corresponding types of supplementary data to be used as feature data for training the carbon emission determination model.
9. The method according to any one of claims 1 to 7, characterized in that, After inputting the real-time operating data into the carbon emission determination model and outputting the carbon emission data of the vehicle to be analyzed, the method further includes: Record the carbon emission data of the vehicle to be analyzed; Based on the set chart type, a chart of the carbon emissions of the vehicle to be analyzed is generated and output.
10. A carbon emission determination device, characterized in that, The carbon emission determination device includes: The acquisition module is used to acquire historical and real-time operating data of the vehicle to be analyzed, wherein the historical operating data includes historical records of energy consumption. An analysis module is used to determine target features associated with the carbon emission data of the vehicle to be analyzed from the historical operating data; The determination module is used to determine the carbon emission determination model corresponding to the vehicle to be analyzed based on the historical operating data corresponding to the target features; The calculation module inputs the real-time operating data into the carbon emission determination model and outputs the carbon emission data of the vehicle to be analyzed.
11. A control device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, cause the control device to perform the carbon emission determination method as described in any one of claims 1 to 9.
12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the carbon emission determination method as described in any one of claims 1 to 9.
13. A computer program product, characterized in that, The computer program product includes computer execution instructions that, when executed by a processor, are used to implement the carbon emission determination method as described in any one of claims 1 to 9.