Methods and devices for assessing the quality of vehicle exhaust remote sensing monitoring data

By constructing a balanced vehicle set and using a random forest classification model to evaluate the quality of vehicle exhaust remote sensing monitoring data, the problem of poor remote sensing monitoring data quality was solved, achieving efficient data screening and quality assessment, and improving the accuracy and stability of the data.

CN123069099APending Publication Date: 2026-11-10TSINGHUA UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510609285.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-13
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing road remote sensing monitoring data is of poor quality and low accuracy, and is susceptible to falsification, leading to false positives and false negatives, which cannot effectively support the construction of high-precision remote sensing detection networks and the formulation of standards.

Method used

By acquiring remote sensing monitoring data of vehicle exhaust emissions and vehicle static information, a balanced vehicle set is constructed. A random forest classification model is used to identify the differences in emission characteristics between new and old vehicles. The data quality is evaluated by correlation analysis between the model output and vehicle static information.

Benefits of technology

It provides an objective and automated method to screen high-quality remote sensing monitoring data, eliminate unstable or low-quality data periods, and improve the effectiveness and stability of the data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123069099A_ABST
    Figure CN123069099A_ABST
Patent Text Reader

Abstract

This invention provides a method and apparatus for evaluating the quality of remote sensing monitoring data for vehicle exhaust emissions. The method includes: acquiring corresponding data from the same number of new and old vehicles to form a balanced vehicle set; dividing the remote sensing monitoring data into preset time windows; for each time window, obtaining statistical indicators of the quantiles of the nitrogen monoxide emission factor within the window, and using these statistical indicators as a sample library; training a random forest classification model using the data in the sample library to identify differences in emission characteristics between new and old vehicles; processing the remote sensing monitoring data to be evaluated using the trained model; performing a correlation analysis between the model output and vehicle static information to determine the relationship between new and old vehicle information, and evaluating the data quality of the remote sensing monitoring data within each time window based on the results of the correlation analysis. The standard deviation, kurtosis, and skewness of the processed data are reduced, the data distribution is closer to high-quality remote sensing data, and the effectiveness and stability of the data are significantly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of road vehicle remote sensing monitoring technology, and in particular to a method and apparatus for evaluating the quality of vehicle exhaust remote sensing monitoring data. Background Technology

[0002] With the continuous improvement of vehicle exhaust emission regulations and standards, the monitoring and control of vehicle exhaust pollution has become imperative. Compared with other monitoring methods, road remote sensing monitoring has significant advantages such as low cost, fast measurement speed, and no interference. However, due to the lack of effective operation and maintenance of monitoring equipment, remote sensing monitoring data has long suffered from poor data quality and low accuracy. Furthermore, remote sensing monitoring captures instantaneous data of vehicle emissions, which is inherently unstable. At the same time, there is also the problem of data falsification in road remote sensing monitoring, making false positives and false negatives prone to occur during vehicle exhaust emission monitoring.

[0003] It is evident that the large amount of remote sensing monitoring data currently collected has not been effectively analyzed, nor has it provided strong support for the construction and standardization of high-precision remote sensing detection networks. Therefore, further improving and standardizing the quality control of road remote sensing data is of paramount importance for enhancing the data effectiveness of existing remote sensing detection networks and promoting the control of vehicle exhaust emissions on roads. Summary of the Invention

[0004] In view of this, the present invention provides a method and apparatus for evaluating the quality of vehicle exhaust remote sensing monitoring data, in order to solve at least one of the problems mentioned above.

[0005] To achieve the above objectives, the present invention adopts the following solution:

[0006] According to a first aspect of the present invention, a method for evaluating the quality of vehicle exhaust remote sensing monitoring data is provided. The method includes: acquiring vehicle exhaust remote sensing monitoring data and corresponding vehicle static information and nitrogen monoxide emission factors; selecting qualified remote sensing monitoring data that meets preset standards from the vehicle exhaust remote sensing monitoring data, and constructing a sampled vehicle set from the qualified remote sensing monitoring data and the corresponding vehicle static information and nitrogen monoxide emission factors; extracting corresponding data from the same number of new and old vehicles from the sampled vehicle set to form a balanced vehicle set; dividing the remote sensing monitoring data according to preset time windows; and for each time window, obtaining the quality of the remote sensing monitoring data. The statistical indicators of the quantiles of the nitrogen monoxide emission factor within the specified time window are used as a sample library for the balanced vehicle set. A random forest classification model is trained using the data in the sample library to identify the differences in emission characteristics between new and old vehicles. The trained random forest classification model is used to process the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated, so as to generate the model output for each time window. The output of the random forest classification model is used to perform a correlation analysis between the new and old vehicle information and the vehicle static information. Based on the results of the correlation analysis, the data quality of the remote sensing monitoring data within each time window is evaluated.

[0007] As an embodiment of the present invention, the method described above for obtaining vehicle exhaust remote sensing monitoring data and corresponding vehicle static information and nitric oxide emission factors includes: matching the obtained vehicle exhaust remote sensing monitoring data with the vehicle static information using the vehicle identification number; and obtaining the corresponding nitric oxide emission factor based on the molar mass of nitric oxide, the ratio of nitric oxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-to-carbon ratio of the fuel in the vehicle exhaust remote sensing monitoring data.

[0008] As an embodiment of the present invention, the method described above for extracting the same number of corresponding data of new and old vehicles from the vehicle set to be sampled to form a balanced vehicle set includes: determining, based on the vehicle static information contained in the vehicle set to be sampled, whether the vehicle corresponding to each data point in the vehicle set to be sampled belongs to a new vehicle or an old vehicle category label, wherein the category label is determined based on a comparison between the registration year or service life in the vehicle static information and a preset threshold; randomly selecting N vehicle samples from the data with new vehicle labels and old vehicle labels respectively; and merging the data corresponding to the selected 2N vehicle samples to form a balanced vehicle set.

[0009] As an embodiment of the present invention, the statistical indicators of the quantiles of the nitric oxide emission factor in the above method include: the mean, standard deviation, kurtosis, and skewness of the quantiles of the nitric oxide emission factor.

[0010] As an embodiment of the present invention, the above method performs a correlation analysis between the model output and vehicle static information to determine the new and old vehicle information. Based on the results of the correlation analysis, the data quality of remote sensing monitoring data within each time window is evaluated by: comparing the prediction results of the new and old vehicle categories output by the random forest classification model for the time window with the actual new and old vehicle category labels in the vehicle static information one by one; when the prediction results output by the random forest classification model are completely consistent with the actual new and old vehicle category labels in all vehicle static information within the time window, the remote sensing monitoring data within the time window is determined to be high-quality data; when the prediction results output by the random forest classification model are not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window, the remote sensing monitoring data within the time window is determined to be low-quality data, and the low-quality data is deleted.

[0011] According to a second aspect of the present invention, a vehicle exhaust remote sensing monitoring data quality assessment device is provided. The device includes: a data acquisition unit for acquiring vehicle exhaust remote sensing monitoring data and corresponding vehicle static information and nitrogen oxide emission factors; a vehicle set construction unit for selecting qualified remote sensing monitoring data that meets preset standards from the vehicle exhaust remote sensing monitoring data, and constructing a vehicle set to be sampled from the qualified remote sensing monitoring data and the corresponding vehicle static information and nitrogen oxide emission factors; a balanced vehicle set construction unit for extracting corresponding data from the same number of new and old vehicles from the vehicle set to be sampled to form a balanced vehicle set; and a time window. The system includes a segmentation unit for dividing the remote sensing monitoring data into preset time windows; a sample library construction unit for obtaining statistical indicators of the quantiles of the nitrogen monoxide emission factor within each time window and using these statistical indicators as the sample library for the balanced vehicle set; a model training unit for training a random forest classification model using the data in the sample library to identify differences in emission characteristics between new and old vehicles; and a data processing unit for processing the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated using the trained random forest classification model to generate model outputs for each time window.

[0012] The data quality assessment unit is used to perform correlation analysis between the output of the random forest classification model and the vehicle static information to determine the new and old information of the vehicle, and to evaluate the data quality of the remote sensing monitoring data within each time window based on the results of the correlation analysis.

[0013] As an embodiment of the present invention, the above-mentioned data acquisition unit includes: a data matching module, used to match the acquired vehicle exhaust remote sensing monitoring data with vehicle static information through vehicle identification number; and an emission factor calculation module, used to obtain the corresponding nitrogen monoxide emission factor based on the molar mass of nitrogen monoxide, the ratio of nitrogen monoxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-carbon ratio of fuel in the vehicle exhaust remote sensing monitoring data.

[0014] As an embodiment of the present invention, the above-mentioned balanced vehicle set construction unit includes: a category label determination module, used to determine, based on the vehicle static information contained in the vehicle set to be sampled, whether the vehicle corresponding to each piece of data in the vehicle set to be sampled belongs to a new vehicle or an old vehicle, wherein the category label is determined based on a comparison between the registration year or service life in the vehicle static information and a preset threshold; a data selection module, used to randomly select N vehicle samples from the data with new vehicle labels and old vehicle labels respectively; and a data merging module, used to merge the data corresponding to the selected 2N vehicle samples to form a balanced vehicle set.

[0015] As an embodiment of the present invention, the statistical indicators of the quantiles of the nitric oxide emission factor include: the mean, standard deviation, kurtosis, and skewness of the quantiles of the nitric oxide emission factor.

[0016] As an embodiment of the present invention, the above-mentioned data quality assessment unit includes: a result comparison module, used to compare the prediction results of new and old vehicle categories output by the random forest classification model for the time window with the actual new and old vehicle category labels in the vehicle static information one by one; a high-quality judgment module, used to judge the remote sensing monitoring data in the time window as high-quality data when the prediction results output by the random forest classification model are completely consistent with the actual new and old vehicle category labels in all vehicle static information within the time window; and a low-quality judgment module, used to judge the remote sensing monitoring data in the time window as low-quality data and delete the low-quality data when the prediction results output by the random forest classification model are not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window.

[0017] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0018] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0019] As can be seen from the above technical solution, the vehicle exhaust remote sensing monitoring data quality assessment method and apparatus provided by this invention utilizes the inherent emission characteristic differences (nitric oxide emission factor distribution) between new and old vehicles, which can be quantified by statistical methods, as a reference standard. A machine learning model capable of accurately identifying these emission characteristic differences between new and old vehicles is trained, and this trained model is applied to the remote sensing monitoring data to be evaluated, divided into time windows. This application cleverly uses the model's accuracy in identifying known new and old vehicles within a specific time window as an indicator to indirectly assess the quality of remote sensing monitoring data within that time window. If the model can accurately distinguish between known new and old vehicles within that time window, it indicates that the remote sensing monitoring data for that period can reliably reflect the true emission characteristics, and the data quality is high; conversely, if the model makes an identification error, it suggests that the monitoring data for that period may contain noise, bias, or other unstable factors, causing it to fail to accurately reflect the expected emission differences, and therefore the data quality is low. This method judges data quality by the consistency of internal characteristics of the data, and provides an objective and automated means to screen out high-quality remote sensing monitoring data, eliminate unstable or low-quality data periods, and reduce the standard deviation, kurtosis and skewness of the processed data. The data distribution is closer to that of high-quality remote sensing data, and the effectiveness and stability of the data are significantly improved. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0021] Figure 1 This is a flowchart illustrating a method for assessing the quality of remote sensing monitoring data for vehicle exhaust emissions provided in an embodiment of this application.

[0022] Figure 2 This is a schematic diagram of the process for obtaining remote sensing monitoring data of vehicle exhaust emissions and corresponding vehicle static information and nitrogen oxide emission factors provided in the embodiments of this application.

[0023] Figure 3 This is a schematic diagram of the process of constructing a set of balancing vehicles provided in an embodiment of this application;

[0024] Figure 4 This is a flowchart illustrating the process of evaluating the data quality of remote sensing monitoring data within each time window, as provided in the embodiments of this application.

[0025] Figure 5This is a schematic diagram of the structure of a vehicle exhaust remote sensing monitoring data quality assessment device provided in an embodiment of this application;

[0026] Figure 6 This is a schematic diagram of the structure of the data acquisition unit provided in an embodiment of this application;

[0027] Figure 7 This is a schematic diagram of the structure of the balanced vehicle assembly building unit provided in the embodiments of this application;

[0028] Figure 8 This is a schematic diagram of the structure of the data quality assessment unit provided in the embodiments of this application;

[0029] Figure 9 This is a schematic block diagram of the system configuration of the electronic device provided in the embodiments of this application. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0031] like Figure 1 The diagram shown is a flowchart illustrating a method for assessing the quality of vehicle exhaust remote sensing monitoring data provided in an embodiment of this application. The method includes the following steps:

[0032] Step S101: Obtain vehicle exhaust remote sensing monitoring data and corresponding vehicle static information and nitrogen monoxide emission factor.

[0033] In this embodiment, vehicle exhaust remote sensing monitoring data is primarily collected by a Remote Sensing Device (RSD) system. This system typically places the monitoring equipment on the side of the road or above a road crossing, enabling real-time monitoring of normally moving vehicles in a non-contact manner. When a vehicle passes through the detection area, a vehicle trigger (radar, induction coil, etc.) senses its entry, and a high-speed camera automatically captures images of the license plate and vehicle exterior, achieving unique vehicle identification. Simultaneously, as the vehicle passes the monitoring point, exhaust gases from the exhaust pipe instantly pass through the system's optical measurement channel. The device penetrates the exhaust gases using infrared and ultraviolet beams, and the receiver detects changes in gas absorbance, thereby calculating the concentrations of pollutants such as NO and CO2. In this embodiment, vehicle static information includes, but is not limited to, registration year, emission standards, and engine displacement.

[0034] In order to correlate the acquired vehicle exhaust remote sensing monitoring data with vehicle static information and nitrogen oxide emission factors, preferably, such as Figure 2 As shown, this step may further include the following sub-steps:

[0035] Step S1011: Match the acquired vehicle exhaust remote sensing monitoring data with the vehicle static information using the vehicle identification number.

[0036] After collecting remote sensing data on vehicle exhaust emissions, the first step is to extract the Vehicle Identification Number (VIN) for each vehicle. The system then links with the vehicle management database to retrieve and obtain the corresponding vehicle's static information, including registration year, emission standard, vehicle model, and fuel type. This matching ensures that each piece of remote sensing data accurately corresponds to the vehicle's static attributes, providing fundamental data support for subsequent data analysis and classification.

[0037] Step S1012: Based on the molar mass of nitric oxide, the ratio of nitric oxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-carbon ratio of the fuel in the vehicle exhaust remote sensing monitoring data, obtain the corresponding nitric oxide emission factor.

[0038] Specifically, in this embodiment, the nitric oxide emission factor can be obtained using the following formula (1):

[0039]

[0040] In the above formula, EF is the NO emission factor; Pollutant(g) / Fuel(kg) is the number of grams of NO emitted when 1 kg of fuel is burned; M pollutant 1 is the molar mass of NO; ​​Pollutant(%) / CO2(%) is the ratio of NO concentration to CO2 obtained by remote sensing measurement; MC is the molar mass of carbon, 12 g / mol; MH is the molar mass of hydrogen, 1 g / mol; r is the atomic hydrogen-to-carbon ratio of the fuel, such as 1.87 for gasoline.

[0041] Step S102: Select qualified remote sensing monitoring data that meet the preset standards from the vehicle exhaust remote sensing monitoring data, and construct a sample vehicle set by combining the qualified remote sensing monitoring data, the corresponding vehicle static information, and the nitrogen monoxide emission factor.

[0042] The raw remote sensing monitoring data is filtered by applying preset validity criteria to remove data records that are highly volatile, unstable, or potentially falsified, retaining only high-quality datasets. To ensure a high-quality dataset, in this embodiment, data meeting the preset criteria can be measured by remote sensing equipment that has undergone rigorous operation and maintenance; only data from these devices is retained in this step, while data from other devices is deleted. Furthermore, to further improve data quality, this embodiment also performs data cleaning on the selected qualified remote sensing monitoring data that meets the preset criteria to remove invalid data outside the effective range. The data retained after cleaning constitutes the sample vehicle set, which also includes remote sensing monitoring data, corresponding vehicle static information, and nitrogen oxide emission factors.

[0043] Step S103: Extract the same number of corresponding data of new and old vehicles from the vehicle set to be sampled to form a balanced vehicle set.

[0044] Because the number of new and old vehicles in the vehicle set to be sampled obtained in step S102 above is unbalanced, the subsequent random forest classification model may be biased towards predicting new vehicles. Therefore, this step creates a balanced vehicle dataset by randomly re-sampling new and old vehicle samples from the high-quality dataset (vehicle set to be sampled) to ensure that the number of samples drawn from new and old vehicles is the same.

[0045] Preferred, such as Figure 3 As shown, this step may further include:

[0046] Step S1031: Based on the vehicle static information contained in the vehicle set to be sampled, determine the category label of whether the vehicle corresponding to each piece of data in the vehicle set to be sampled is a new vehicle or an old vehicle. The category label is determined based on the comparison between the registration year or usage years in the vehicle static information and a preset threshold.

[0047] For example, based on the registration year, vehicles 15 years or older are classified as "used cars," while those 6 years or younger are classified as "new cars." It should be noted that in this embodiment, to improve the feature differentiation between new and old vehicles in the training dataset, vehicles between 6 and 15 years old were removed when constructing the balanced vehicle set. However, in subsequent model applications, vehicles of all ages can be used as input data.

[0048] Step S1032: Randomly select N vehicle samples from the data with new vehicle tags and old vehicle tags respectively.

[0049] The labeled data are categorized into "New Vehicles" and "Old Vehicles" groups. Then, N new vehicles are randomly selected from the "New Vehicles" group, and N old vehicles are randomly selected from the "Old Vehicles" group. The value of N is determined by the smaller sample size of the two groups and the actual needs, to ensure data balance and representativeness.

[0050] Step S1033: Merge the data corresponding to the selected 2N vehicle samples to form a balanced vehicle set.

[0051] The N randomly selected new vehicle samples and N used vehicle samples are combined to obtain a total of 2N vehicle data points with known categories. This balanced vehicle set ensures that the number of new and used vehicle samples is equal, effectively avoiding category bias in the subsequent machine learning modeling process, and facilitating accurate discovery and comparison of differences in emission characteristics between new and used vehicles.

[0052] Step S104: Divide the remote sensing monitoring data into preset time windows.

[0053] The entire remote sensing monitoring data was divided into multiple subsets according to preset time windows, such as hourly, three-hourly, and daily, based on time sequence. This step takes into account the influence of time-dependent factors such as road traffic flow and weather, and also facilitates independent quality assessment of data for each time period, providing a basis for subsequent phased and segmented output judgments. Sensitivity experiments showed that the accuracy of the subsequent random forest classification model was highest when the data was divided into 3-hour time intervals.

[0054] Step S105: For each time window, obtain the statistical index of the quantile of the nitrogen monoxide emission factor within the time window, and use the statistical index as the sample library of the balanced vehicle set.

[0055] For each window, statistical features of the quantiles of the nitric oxide emission factor within the window are extracted, including but not limited to mean, standard deviation, kurtosis, and skewness. These features can fully characterize the overall distribution of exhaust emission behavior of new and old vehicles. The statistical features of each time window are combined with the corresponding new and old vehicle category information to form a high-quality sample library of balanced vehicle sets to support subsequent machine learning modeling.

[0056] Step S106: Train a random forest classification model using the data in the sample library to identify the differences in emission characteristics between new and old vehicles.

[0057] In this embodiment, 70% of the samples in the sample library can be used as the training set, and the remaining 30% can be used for model testing. The model evaluation results show that the accuracy of balancing high-quality remote sensing data from the vehicle set is 86.95%. The model's Kappa coefficient is 0.737, indicating a high consistency between the predicted results and the actual values. The AUC value is 0.939, further demonstrating that the model can effectively distinguish between new and old vehicles and capture macroscopic emission patterns in high-quality remote sensing data.

[0058] Step S107: Use the trained random forest classification model to process the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated, so as to generate the model output for each time window.

[0059] In this step, when assessing the quality of remote sensing monitoring data with unknown quality, statistical indicators are extracted on a time-window basis and input into a pre-trained random forest model. The model outputs the classification results of new and old vehicles based on the data performance within each window. The model's output can provide quantitative basis for subsequent correlation analysis.

[0060] Step S108: Perform a correlation analysis between the output of the random forest classification model and the vehicle static information to determine the new and old information of the vehicle, and evaluate the data quality of the remote sensing monitoring data within each time window based on the results of the correlation analysis.

[0061] Finally, the discriminant output of the random forest classification model is compared one by one with the actual new and old real labels of vehicles within the same time window. If the model can correspond well to the real labels, that is, correctly distinguish between new and old vehicles, it indicates that the data within this window can truly reflect the essential emission characteristics of new and old vehicles, and the data quality is high. If the comparison results are chaotic, that is, if there are misclassifications, it indicates that the data in this window may be interfered with or abnormal, and the quality is low. Based on this, an automatic, objective, and quantitative assessment of the quality of remote sensing monitoring data for each time period can be achieved.

[0062] Preferred, such as Figure 4 As shown, this step may further include the following sub-steps:

[0063] Step S1081: Compare the prediction results of the new and old vehicle categories output by the random forest classification model for the time window with the actual new and old vehicle category labels in the vehicle static information one by one.

[0064] Step S1082: When the prediction result output by the random forest classification model is completely consistent with the actual new and old vehicle category labels in all vehicle static information within the time window, the remote sensing monitoring data within the time window is determined to be high-quality data.

[0065] Step S1083: When the prediction result output by the random forest classification model is not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window, the remote sensing monitoring data within the time window is determined to be low-quality data and the low-quality data is deleted.

[0066] By using these three sub-steps to verify the consistency of the data using the established classification model, we can effectively identify and eliminate time window data where the model predictions deviate significantly from the actual situation, thereby achieving further screening and purification of remote sensing monitoring data quality.

[0067] As described above, the vehicle exhaust remote sensing monitoring data quality assessment method provided by this invention utilizes the inherent, statistically quantifiable differences in emission characteristics (nitrogen monoxide emission factor distribution) between new and old vehicles as a reference standard. A machine learning model capable of accurately identifying these emission characteristic differences between new and old vehicles is trained, and this trained model is applied to the remote sensing monitoring data to be evaluated, divided into time windows. This application cleverly uses the model's accuracy in identifying known new and old vehicles within a specific time window as an indicator to indirectly assess the quality of the remote sensing monitoring data within that time window. If the model can accurately distinguish between known new and old vehicles within that time window, it indicates that the remote sensing monitoring data for that period reliably reflects the true emission characteristics, and the data quality is high. Conversely, if the model makes an identification error, it suggests that the monitoring data for that period may contain noise, bias, or other unstable factors, causing it to fail to accurately reflect the expected emission differences, and therefore the data quality is low. This method judges data quality by the consistency of internal characteristics of the data, and provides an objective and automated means to screen out high-quality remote sensing monitoring data, eliminate unstable or low-quality data periods, and reduce the standard deviation, kurtosis and skewness of the processed data. The data distribution is closer to that of high-quality remote sensing data, and the effectiveness and stability of the data are significantly improved.

[0068] like Figure 5 The diagram shown is a structural schematic of a vehicle exhaust remote sensing monitoring data quality assessment device provided in an embodiment of this application. The device includes: a data acquisition unit 510, a vehicle set construction unit 520, a balanced vehicle set construction unit 530, a time window division unit 540, a sample library construction unit 550, a model training unit 560, a data processing unit 570, and a data quality assessment unit 580, which are connected sequentially.

[0069] The data acquisition unit 510 is used to acquire remote sensing monitoring data of vehicle exhaust emissions and corresponding vehicle static information and nitrogen monoxide emission factors.

[0070] The vehicle set construction unit 520 is used to select qualified remote sensing monitoring data that meet preset standards from the vehicle exhaust remote sensing monitoring data, and construct the qualified remote sensing monitoring data, the corresponding vehicle static information, and the nitrogen monoxide emission factor into a vehicle set to be sampled.

[0071] The vehicle set construction unit 530 is used to extract the same number of corresponding data of new and old vehicles from the vehicle set to be sampled to form a vehicle set.

[0072] The time window division unit 540 is used to divide the remote sensing monitoring data according to a preset time window;

[0073] The sample library construction unit 550 is used to obtain the statistical index of the quantile of the nitrogen monoxide emission factor within each time window, and use the statistical index as the sample library of the balanced vehicle set.

[0074] The model training unit 560 is used to train a random forest classification model using data from the sample library to identify differences in emission characteristics between new and old vehicles.

[0075] The data processing unit 570 is used to process the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated using the trained random forest classification model, so as to generate the model output of each time window.

[0076] The data quality assessment unit 580 is used to perform correlation analysis between the output of the random forest classification model and the vehicle static information to determine the new and old information of the vehicle, and to evaluate the data quality of the remote sensing monitoring data within each time window based on the results of the correlation analysis.

[0077] Preferred, such as Figure 6 As shown, the data acquisition unit 510 includes:

[0078] The data matching module 511 is used to match the acquired vehicle exhaust remote sensing monitoring data with the vehicle static information through the vehicle identification number.

[0079] The emission factor calculation module 512 is used to obtain the corresponding nitrogen monoxide emission factor based on the molar mass of nitrogen monoxide, the ratio of nitrogen monoxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-carbon ratio of fuel in the vehicle exhaust remote sensing monitoring data.

[0080] Preferred, such as Figure 7 As shown, the aforementioned balanced vehicle assembly unit 530 includes:

[0081] The category label determination module 531 is used to determine the category label of whether the vehicle corresponding to each piece of data in the vehicle set to be sampled belongs to a new vehicle or an old vehicle based on the vehicle static information contained in the vehicle set to be sampled. The category label is determined based on the comparison between the registration year or the years of use in the vehicle static information and a preset threshold.

[0082] The data selection module 532 is used to randomly select N vehicle samples from the data with new vehicle labels and old vehicle labels respectively.

[0083] The data merging module 533 merges the data corresponding to the selected 2N vehicle samples to form a balanced vehicle set.

[0084] Preferably, the statistical indicators for the quantiles of the nitric oxide emission factor include: the mean, standard deviation, kurtosis, and skewness of the quantiles of the nitric oxide emission factor.

[0085] Preferred, such as Figure 8 As shown, the data quality assessment unit 580 includes:

[0086] The result comparison module 581 is used to compare the prediction results of the new and old vehicle categories output by the random forest classification model for the time window with the actual new and old vehicle category labels in the vehicle static information one by one.

[0087] The high-quality judgment module 582 is used to judge the remote sensing monitoring data in the time window as high-quality data when the prediction result output by the random forest classification model is completely consistent with the real new and old vehicle category labels in all vehicle static information within the time window.

[0088] The low-quality judgment module 583 determines the remote sensing monitoring data in the time window as low-quality data and deletes it when the prediction result output by the random forest classification model is not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window.

[0089] As described above, the vehicle exhaust remote sensing monitoring data quality assessment device provided by this invention utilizes the inherent emission characteristic differences (nitric oxide emission factor distribution) between new and old vehicles, which can be quantified statistically, as a reference standard. A machine learning model capable of accurately identifying these emission characteristic differences between new and old vehicles is trained, and this trained model is applied to the remote sensing monitoring data to be evaluated, divided into time windows. This application cleverly uses the model's accuracy in identifying known new and old vehicles within a specific time window as an indicator to indirectly assess the quality of the remote sensing monitoring data within that time window. If the model can accurately distinguish between known new and old vehicles within that time window, it indicates that the remote sensing monitoring data for that period reliably reflects the true emission characteristics, and the data quality is high. Conversely, if the model makes an identification error, it suggests that the monitoring data for that period may contain noise, bias, or other unstable factors, causing it to fail to accurately reflect the expected emission differences, and therefore the data quality is low. This method judges data quality by the consistency of internal characteristics of the data, and provides an objective and automated means to screen out high-quality remote sensing monitoring data, eliminate unstable or low-quality data periods, and reduce the standard deviation, kurtosis and skewness of the processed data. The data distribution is closer to that of high-quality remote sensing data, and the effectiveness and stability of the data are significantly improved.

[0090] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method.

[0091] This invention also provides a computer-readable storage medium storing a computer program for performing the above-described methods.

[0092] like Figure 9 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 9 All components shown; in addition, the electronic device 600 may also include Figure 9 For components not shown, please refer to existing technologies.

[0093] like Figure 9 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0094] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.

[0095] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0096] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.

[0097] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0098] The communication module 110 is a transmitter / receiver that transmits and receives signals via the antenna 111. The communication module 110 (transmitter / receiver) is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0099] Based on different communication technologies, multiple communication modules 110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module 110 (transmitter / receiver) is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling typical telecommunications functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 130 is coupled to a central processing unit 100, enabling on-device recording via the microphone 132 and on-device playback of stored audio via the speaker 131.

[0100] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0102] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0103] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0104] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for evaluating the quality of vehicle exhaust remote sensing monitoring data, characterized in that, The method includes: Acquire remote sensing monitoring data of vehicle exhaust emissions and corresponding vehicle static information and nitrogen oxide emission factors; Qualified remote sensing monitoring data that meet preset standards are selected from the vehicle exhaust remote sensing monitoring data, and the qualified remote sensing monitoring data, along with the corresponding vehicle static information and nitrogen oxide emission factor, are used to construct a set of vehicles to be sampled. The same number of corresponding data from new and old vehicles are extracted from the vehicle set to be sampled to form a balanced vehicle set; The remote sensing monitoring data is divided into preset time windows; For each time window, obtain the statistical index of the quantile of the nitrogen monoxide emission factor within the time window, and use the statistical index as the sample library of the balanced vehicle set; A random forest classification model was trained using data from the aforementioned sample library to identify differences in emission characteristics between new and old vehicles; The trained random forest classification model is used to process the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated, so as to generate the model output for each time window; The output of the random forest classification model is used to perform a correlation analysis between the new and old vehicle information and the vehicle static information. Based on the results of the correlation analysis, the data quality of the remote sensing monitoring data within each time window is evaluated.

2. The method for evaluating the quality of vehicle exhaust remote sensing monitoring data as described in claim 1, characterized in that, The acquisition of vehicle exhaust remote sensing monitoring data and corresponding vehicle static information and nitrogen oxide emission factors includes: The acquired vehicle exhaust remote sensing monitoring data is matched with the vehicle static information using the vehicle identification number. The corresponding nitrogen monoxide emission factor is obtained based on the molar mass of nitrogen monoxide, the ratio of nitrogen monoxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-to-carbon ratio of the fuel in the vehicle exhaust remote sensing monitoring data.

3. The method for evaluating the quality of vehicle exhaust remote sensing monitoring data as described in claim 1, characterized in that, The step of extracting the same number of corresponding data from new and old vehicles from the vehicle set to be sampled to form a balanced vehicle set includes: Based on the vehicle static information contained in the vehicle set to be sampled, determine the category label of whether the vehicle corresponding to each piece of data in the vehicle set to be sampled is a new vehicle or an old vehicle. The category label is determined based on the comparison between the registration year or the years of use in the vehicle static information and a preset threshold. N vehicles are randomly selected from the data with new vehicle tags and old vehicle tags respectively; The data corresponding to the selected 2N vehicle samples are merged to form a balanced vehicle set.

4. The method for evaluating the quality of vehicle exhaust remote sensing monitoring data as described in claim 1, characterized in that, The statistical indicators of the quantiles of the nitric oxide emission factor include: the mean, standard deviation, kurtosis, and skewness of the quantiles of the nitric oxide emission factor.

5. The method for evaluating the quality of vehicle exhaust remote sensing monitoring data as described in claim 1, characterized in that, The aforementioned method of performing correlation analysis between the output of the random forest classification model and vehicle static information to determine the new and old vehicle information, and evaluating the data quality of remote sensing monitoring data within each time window based on the results of the correlation analysis, includes: The prediction results of the new and old vehicle categories output by the random forest classification model for the time window are compared one by one with the actual new and old vehicle category labels in the vehicle static information. When the prediction result output by the random forest classification model is completely consistent with the actual new and old vehicle category labels in all vehicle static information within the time window, the remote sensing monitoring data within the time window is judged as high-quality data. When the prediction result output by the random forest classification model is not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window, the remote sensing monitoring data within the time window is judged as low-quality data and the low-quality data is deleted.

6. A device for evaluating the quality of vehicle exhaust remote sensing monitoring data, characterized in that, The device includes: The data acquisition unit is used to acquire remote sensing monitoring data of vehicle exhaust emissions and corresponding vehicle static information and nitrogen oxide emission factors. The vehicle set construction unit is used to select qualified remote sensing monitoring data that meet preset standards from the vehicle exhaust remote sensing monitoring data, and construct the qualified remote sensing monitoring data, the corresponding vehicle static information and nitrogen oxide emission factor into a vehicle set to be sampled. The vehicle set construction unit is used to extract the same number of corresponding data of new and old vehicles from the vehicle set to be sampled to form a vehicle set. A time window division unit is used to divide the remote sensing monitoring data according to a preset time window; The sample library construction unit is used to obtain the statistical index of the quantile of the nitrogen monoxide emission factor within each time window, and use the statistical index as the sample library of the balanced vehicle set. The model training unit is used to train a random forest classification model using data from the sample library to identify differences in emission characteristics between new and old vehicles. The data processing unit to be evaluated is used to process the statistical indicators obtained from each time window of the remote sensing monitoring data to be evaluated using the trained random forest classification model, so as to generate the model output of each time window. The data quality assessment unit is used to perform correlation analysis between the output of the random forest classification model and the vehicle static information to determine the new and old information of the vehicle, and to evaluate the data quality of the remote sensing monitoring data within each time window based on the results of the correlation analysis.

7. The vehicle exhaust remote sensing monitoring data quality assessment device as described in claim 6, characterized in that, The data acquisition unit includes: The data matching module is used to match the acquired vehicle exhaust remote sensing monitoring data with the vehicle static information through the vehicle identification number. The emission factor calculation module is used to obtain the corresponding nitrogen monoxide emission factor based on the molar mass of nitrogen monoxide, the ratio of nitrogen monoxide concentration to carbon dioxide, the molar mass of carbon, the molar mass of hydrogen, and the atomic hydrogen-carbon ratio of the fuel in the vehicle exhaust remote sensing monitoring data.

8. The vehicle exhaust remote sensing monitoring data quality assessment device as described in claim 6, characterized in that, The balanced vehicle assembly building unit includes: The category label determination module is used to determine the category label of each data point in the vehicle set to be sampled, based on the vehicle static information contained in the vehicle set to be sampled, whether the vehicle is a new vehicle or an old vehicle. The category label is determined based on a comparison between the registration year or service life in the vehicle static information and a preset threshold. The data selection module is used to randomly select N vehicle samples from the data with new vehicle labels and old vehicle labels respectively. The data merging module merges the data corresponding to the selected 2N vehicle samples to form a balanced vehicle set.

9. The vehicle exhaust remote sensing monitoring data quality assessment device as described in claim 6, characterized in that, The statistical indicators of the quantiles of the nitric oxide emission factor include: the mean, standard deviation, kurtosis, and skewness of the quantiles of the nitric oxide emission factor.

10. The vehicle exhaust remote sensing monitoring data quality assessment device as described in claim 6, characterized in that, The data quality assessment unit includes: The result comparison module is used to compare the prediction results of the new and old vehicle categories output by the random forest classification model for the time window with the actual new and old vehicle category labels in the vehicle static information one by one. The high-quality judgment module is used to judge the remote sensing monitoring data in the time window as high-quality data when the prediction result output by the random forest classification model is completely consistent with the real new and old vehicle category labels in all vehicle static information within the time window. The low-quality judgment module determines the remote sensing monitoring data in the time window as low-quality data and deletes it when the prediction result output by the random forest classification model is not completely consistent with the actual new and old vehicle categories in all vehicle static information within the time window.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.