A method for processing motor vehicle exhaust telemetry drift data

Abstract

The present invention discloses a method for processing motor vehicle exhaust telemetry drift data, the method comprising: calculating the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide; calculating the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratios; dividing the fuel emission factor into a trend term, a periodic term and a fluctuation term using an STL time series decomposition algorithm, and correcting the data drift of the trend term and the periodic term to obtain corrected fuel emission factor data; screening the corrected fuel emission factor data for abnormal points using several different machine learning models, and marking each abnormal point screened by different machine learning models; performing threshold determination on each abnormal point based on a preset screening threshold, and determining target motor vehicles with excessive emissions. The present invention can effectively solve the data drift problem of exhaust telemetry equipment and screen vehicles with excessive exhaust emissions, and can be widely used in the field of data analysis technology.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a method for processing motor vehicle exhaust telemetry drift data. Background Art

[0002] Thanks to the rapid economic development, the number of motor vehicles in my country has increased rapidly in recent years, and motor vehicle exhaust has also become an important factor affecting the level of pollutant emissions in my country. In order to further reduce the level of motor vehicle pollutant emissions, it is necessary to accurately identify vehicles with excessive exhaust emissions in road traffic. Although the traditional idling method and operating condition method can accurately detect the emission level of vehicles, the detection cost is high and the detection cycle is long, and it is not suitable for the rapid measurement of exhaust emission levels of a large number of motor vehicles. In contrast, the motor vehicle exhaust remote sensing detection method based on optical principles can quickly detect the concentration of various exhaust pollutants of vehicles passing through the measuring equipment without affecting the operation of traffic flow, and is being increasingly widely used in the field of detecting motor vehicles with excessive exhaust emissions.

[0003] However, remote sensing equipment for vehicle exhaust emissions typically uses an open-ended detection method. The pollutant concentration data measured by the equipment is affected by environmental factors such as temperature, humidity, and wind speed. Furthermore, it can drift to varying degrees depending on the equipment's calibration and operating time. To ensure that the same equipment's measurements over different time periods consistently reflect the vehicle's true exhaust pollutant emissions, a method is required to correct the data and subsequently apply it to the screening of vehicles with excessive emissions. Currently, most research, both domestically and internationally, has overlooked the drift issue of telemetry data, lacking research on methods for correcting and screening drifted exhaust telemetry data. Summary of the Invention

[0004] In view of this, an embodiment of the present invention provides a method for processing motor vehicle exhaust telemetry drift data, which can effectively solve the data drift problem of exhaust telemetry equipment and screen vehicles with excessive exhaust emissions.

[0005] One aspect of an embodiment of the present invention provides a method for processing vehicle exhaust telemetry drift data, comprising:

[0006] Calculate the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide;

[0007] Calculating fuel emission factors for different types of motor vehicle exhaust pollutants based on the concentration ratios;

[0008] Using the STL time series decomposition algorithm to divide the fuel emission factor into a trend item, a period item, and a fluctuation item, and correcting the data drift of the trend item and the period item to obtain corrected fuel emission factor data;

[0009] Screening the corrected fuel emission factor data for outliers using several different machine learning models, and marking each outlier screened by the different machine learning models;

[0010] According to the preset screening threshold, a threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

[0011] Optionally, the step of calculating the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide includes:

[0012] Calculating a first concentration ratio of nitrogen monoxide to carbon dioxide in exhaust pollutants from a motor vehicle;

[0013] calculating a second concentration ratio of hydrocarbons to carbon dioxide in exhaust pollutants from motor vehicles;

[0014] Calculating a third concentration ratio between carbon monoxide and carbon dioxide in motor vehicle exhaust pollutants;

[0015] Wherein, the calculation formula of the first concentration ratio is:

[0016]

[0017] The calculation formula of the second concentration ratio is:

[0018]

[0019] The calculation formula of the third concentration ratio is:

[0020]

[0021] in, Represents the concentration ratio of NO to CO2; Represents the concentration ratio of hydrocarbons to CO2; Represents the concentration ratio of CO to CO2; C NO represents the measured concentration of NO; C HC Represents the measured concentration of hydrocarbons; C CO represents the measured concentration of CO; Represents the measured concentration of CO2.

[0022] Optionally, the calculating of fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratios includes:

[0023] Calculating a fuel emission factor of NO based on the first concentration ratio;

[0024] calculating a fuel emission factor of hydrocarbons based on the second concentration ratio;

[0025] calculating a fuel emission factor for CO based on the third concentration ratio;

[0026] The calculation formula of the NO fuel emission factor is:

[0027]

[0028] The calculation formula of the HC fuel emission factor is:

[0029]

[0030] The calculation formula of the CO fuel emission factor is:

[0031]

[0032] Among them, EF NO represents the fuel emission factor of NO; EF HC represents the fuel emission factor of the hydrocarbon in question; EF CO represents the fuel emission factor of CO; constants 30, 44, and 28 are the molar masses of various pollutants respectively; M fuel represents the molar mass of the fuel type used; the constant 0.493 is the carbon mass conversion factor calculated based on propane.

[0033] Optionally, the STL time series decomposition algorithm is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term, and the data drift of the trend term and the period term is corrected to obtain corrected fuel emission factor data, including:

[0034] According to the additive splitting rule, the STL time series decomposition algorithm based on Fourier transform is used to divide the fuel emission factor data into three parts: trend term, period term, and fluctuation term. The trend term and period term are used to characterize the degree of data drift caused by different calibration conditions and different operating time of the exhaust gas telemetry equipment.

[0035] performing correction processing on the drift data using the trend item and the period item;

[0036] The expression of the correction process is:

[0037] EF A =EF O -EF T -EF S

[0038] Among them, EF A For the corrected data; EF O is the drift data before correction; EF Tis the trend item data; EF S It is the periodic item data.

[0039] Optionally, the performing outlier screening on the corrected fuel emission factor data using several different machine learning models, and marking each outlier screened by different machine learning models includes:

[0040] Using a first model constructed by an isolation forest algorithm and a second model constructed by a DBSCAN clustering algorithm, the corrected fuel emission factor data is screened for outliers to obtain outliers;

[0041] The outliers obtained by screening the first model and the second model are marked respectively.

[0042] Optionally, performing threshold determination on each of the abnormal points according to a preset screening threshold to determine target motor vehicles with excessive emissions includes:

[0043] Determine the screening threshold based on the fuel type and exhaust pollutant type of the vehicle to be screened, combined with the exhaust emission standard documents;

[0044] According to the screening threshold, threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

[0045] Another aspect of the present invention provides a vehicle exhaust telemetry drift data processing device.

[0046] include:

[0047] The first module is used to calculate the concentration ratio of different types of motor vehicle exhaust pollutants to carbon dioxide;

[0048] The second module is used to calculate the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratio;

[0049] The third module is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term using an STL time series decomposition algorithm, and correct the data drift of the trend term and the period term to obtain corrected fuel emission factor data;

[0050] A fourth module is configured to screen the corrected fuel emission factor data for outliers using several different machine learning models, and mark each outlier screened by different machine learning models;

[0051] The fifth module is used to perform threshold determination on each of the abnormal points according to a preset screening threshold, and determine the target motor vehicles with excessive emissions.

[0052] Another aspect of an embodiment of the present invention further provides an electronic device, including a processor and a memory;

[0053] The memory is used to store programs;

[0054] The processor executes the program to implement the method described above.

[0055] Another aspect of the embodiments of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method described above.

[0056] The present invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the above method.

[0057] An embodiment of the present invention calculates the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide; calculates the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratios; uses the STL time series decomposition algorithm to divide the fuel emission factor into trend terms, periodic terms and fluctuation terms, and corrects the data drift of the trend terms and the periodic terms to obtain corrected fuel emission factor data; uses several different machine learning models to screen the corrected fuel emission factor data for abnormal points, and marks the various abnormal points screened by different machine learning models; performs threshold judgment on each of the abnormal points based on a preset screening threshold, and determines the target motor vehicles with excessive emissions. The present invention can better solve the data drift problem of exhaust telemetry equipment and screen vehicles with excessive exhaust emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0059] Figure 1 An overall step flow chart provided for an embodiment of the present invention;

[0060] Figure 2 This is a data distribution diagram of fuel emission factors calculated by an embodiment of the present invention;

[0061] Figure 3This is an example diagram of the effect of the STL time series decomposition algorithm used in an embodiment of the present invention;

[0062] Figure 4 This is an example of the results of the motor vehicle exhaust telemetry drift data correction and screening method proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0064] In order to solve the problems existing in the prior art, an embodiment of the present invention provides a method for processing vehicle exhaust telemetry drift data, comprising:

[0065] Calculate the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide;

[0066] Calculating fuel emission factors for different types of motor vehicle exhaust pollutants based on the concentration ratios;

[0067] Using the STL time series decomposition algorithm to divide the fuel emission factor into a trend item, a period item, and a fluctuation item, and correcting the data drift of the trend item and the period item to obtain corrected fuel emission factor data;

[0068] Screening the corrected fuel emission factor data for outliers using several different machine learning models, and marking each outlier screened by the different machine learning models;

[0069] According to the preset screening threshold, a threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

[0070] Optionally, the step of calculating the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide includes:

[0071] Calculating a first concentration ratio of nitrogen monoxide to carbon dioxide in exhaust pollutants from a motor vehicle;

[0072] calculating a second concentration ratio of hydrocarbons to carbon dioxide in exhaust pollutants from motor vehicles;

[0073] Calculating a third concentration ratio between carbon monoxide and carbon dioxide in exhaust pollutants from motor vehicles;

[0074] Wherein, the calculation formula of the first concentration ratio is:

[0075]

[0076] The calculation formula of the second concentration ratio is:

[0077]

[0078] The calculation formula of the third concentration ratio is:

[0079]

[0080] in, Represents the concentration ratio of NO to CO2; Represents the concentration ratio of hydrocarbons to CO2; Represents the concentration ratio of CO to CO2; C NO represents the measured concentration of NO; C HC represents the measured concentration of HC; C CO represents the measured concentration of CO; Represents the measured concentration of CO2.

[0081] Optionally, the calculating of fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratios includes:

[0082] Calculating a fuel emission factor of NO based on the first concentration ratio;

[0083] calculating a fuel emission factor of hydrocarbons based on the second concentration ratio;

[0084] calculating a fuel emission factor for CO based on the third concentration ratio;

[0085] The calculation formula of the NO fuel emission factor is:

[0086]

[0087] The calculation formula of the HC fuel emission factor is:

[0088]

[0089] The calculation formula of the CO fuel emission factor is:

[0090]

[0091] Among them, EF NO represents the fuel emission factor of NO; EF HC represents the fuel emission factor of the hydrocarbon in question; EF CO represents the fuel emission factor of CO; constants 30, 44, and 28 are the molar masses of various pollutants respectively; M fuelrepresents the molar mass of the fuel type used; the constant 0.493 is the carbon mass conversion factor calculated based on propane.

[0092] Optionally, the STL time series decomposition algorithm is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term, and the data drift of the trend term and the period term is corrected to obtain corrected fuel emission factor data, including:

[0093] According to the additive splitting rule, the STL time series decomposition algorithm based on Fourier transform is used to divide the fuel emission factor data into three parts: trend term, period term, and fluctuation term. The trend term and period term are used to characterize the degree of data drift caused by different calibration conditions and different operating time of the exhaust gas telemetry equipment.

[0094] performing correction processing on the drift data using the trend item and the period item;

[0095] The expression of the correction process is:

[0096] EF A =EF O -EF T -EF S

[0097] Among them, EF A For the corrected data; EF O is the drift data before correction; EF T is the trend item data; EF S It is the periodic item data.

[0098] Optionally, the performing outlier screening on the corrected fuel emission factor data using several different machine learning models, and marking each outlier screened by different machine learning models includes:

[0099] Using a first model constructed by an isolation forest algorithm and a second model constructed by a DBSCAN clustering algorithm, the corrected fuel emission factor data is screened for outliers to obtain outliers;

[0100] The outliers obtained by screening the first model and the second model are marked respectively.

[0101] Optionally, performing threshold determination on each of the abnormal points according to a preset screening threshold to determine target motor vehicles with excessive emissions includes:

[0102] Determine the screening threshold based on the fuel type and exhaust pollutant type of the vehicle to be screened, combined with the exhaust emission standard documents;

[0103] According to the screening threshold, threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

[0104] Another aspect of the present invention provides a vehicle exhaust telemetry drift data processing device.

[0105] include:

[0106] The first module is used to calculate the concentration ratio of different types of motor vehicle exhaust pollutants to carbon dioxide;

[0107] The second module is used to calculate the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratio;

[0108] The third module is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term using an STL time series decomposition algorithm, and correct the data drift of the trend term and the period term to obtain corrected fuel emission factor data;

[0109] A fourth module is configured to screen the corrected fuel emission factor data for outliers using several different machine learning models, and mark each outlier screened by different machine learning models;

[0110] The fifth module is used to perform threshold determination on each of the abnormal points according to a preset screening threshold, and determine the target motor vehicles with excessive emissions.

[0111] Another aspect of an embodiment of the present invention further provides an electronic device, including a processor and a memory;

[0112] The memory is used to store programs;

[0113] The processor executes the program to implement the method described above.

[0114] Another aspect of the embodiments of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method described above.

[0115] The present invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the above method.

[0116] The specific implementation process of the present invention is described in detail below with reference to the accompanying drawings:

[0117] The embodiment of the present invention proposes a correction and screening method for vehicle exhaust telemetry drift data. The input of the method is the original data of vehicle exhaust telemetry, and the output is the information of vehicles with excessive exhaust emissions, such as Figure 1 As shown, the method of the present invention comprises the following steps:

[0118] Step 1: Calculate the concentration ratio of various motor vehicle exhaust pollutants to carbon dioxide;

[0119] Step 2: Calculate the fuel emission factors of various types of motor vehicle exhaust pollutants;

[0120] Step 3: Use the STL time series decomposition algorithm to divide the data into three parts: trend term, period term, and fluctuation term, and correct the data drift problem caused by the trend term and period term;

[0121] Step 4: Use several machine learning algorithms to screen data on vehicles that may exceed emission standards, and mark the data points that are determined to be outliers by each machine learning algorithm for further judgment;

[0122] Step 5: Filter the data points marked in step 4 according to the set threshold to determine the final data of vehicles with excessive emissions.

[0123] It should be noted that the method for correcting and screening the drift data of motor vehicle exhaust telemetry proposed in the embodiment of the present invention is aimed at valid data obtained during the normal operation of the exhaust telemetry equipment. Therefore, the measurement data generated during the calibration and maintenance of the exhaust telemetry equipment should be removed, and the measurement data that does not meet the data validity requirements of the industry standards should also be removed. In addition, since the exhaust pollution generated by road traffic flow mainly comes from diesel and gasoline vehicles, and the exhaust pollution generated by other types of new energy vehicles is relatively low, the embodiment of the present invention only processes and analyzes the data of diesel and gasoline vehicles. Furthermore, since the operating principles of diesel engines and gasoline engines are different, the concentration levels of various exhaust pollutants generated are also different, so the correction and screening processes of diesel vehicle data and gasoline vehicle data should be carried out separately.

[0124] In step 1, the calculation formulas for the concentration ratios of various motor vehicle exhaust pollutants to CO2 are shown in Equations 1, 2, and 3, respectively, and the calculated concentration ratios are mainly used to calculate the fuel emission factors of various motor vehicle exhaust pollutants in step 2:

[0125]

[0126] In the formula are the concentration ratios of NO, HC, CO and CO2, C NO 、C HC 、C CO 、 These are the measured concentrations of NO, HC, and CO, respectively. It's worth noting that the original measurement data for NO and HC (hydrocarbons) is in parts per million (ppm), while the measurement units for CO and CO2 are in percentage (%). Therefore, the units must be unified using Equation 4 before calculation:

[0127] C ppm =1000*C ％ #(4)

[0128] Where C ppm is the gas concentration expressed in ppm, C ％ is the gas concentration expressed as a percentage.

[0129] In step 2, the fuel emission factor of each type of motor vehicle exhaust pollutant is an indicator proposed by existing relevant research based on the carbon balance principle to measure the emission level of motor vehicle exhaust pollutants. The calculation process is shown in Equation 5, Equation 6, and Equation 7 respectively:

[0130]

[0131]

[0132] Where EF NO EF HC EF CO are the fuel emission factors of NO, HC, and CO, respectively. The constants 30, 44, and 28 are the molar masses of the pollutants (the molar mass of HC is calculated using C3H8, which is often used as a standard gas for calibrating telemetry equipment). fuel is the molar mass of the fuel type used, which can be 0.01385 kg*mol for diesel -1 , the constant 0.493 is the carbon mass conversion coefficient calculated based on propane.

[0133] Steps 1 and 2 convert the concentration data of various pollutants measured by the exhaust telemetry equipment into fuel emission factor data of various pollutants. The fuel emission factor can better reflect the actual emission level of the vehicle. Its unit is grams per kilogram of fuel (g*kg -1 ).

[0134] In step 3, the fuel emission factor data calculated in step 2 is treated as a time series data that changes continuously with the equipment's operating time. The STL time series decomposition algorithm based on Fourier transform can then be used to decompose the fuel emission factor data into three components: trend term, period term, and fluctuation term according to the additive splitting rule. The trend term and period term characterize the degree of data drift caused by the different calibration conditions and operating durations of the exhaust gas telemetry equipment. Therefore, step 3 uses the decomposed trend term and period term to correct the fuel emission factor data before decomposition. The process is shown in Equation 8:

[0135] EF A =EF O -EF T -EF S #(8)

[0136] Where EF A is the corrected data, EF O is the drift data before correction, EF T is the trend item data, EF S It is the periodic item data.

[0137] After the correction in step 3, it can be considered that the drift problem of the exhaust telemetry data has been well solved, so the same algorithm and standards can be used for screening.

[0138] In step 4, this embodiment of the present invention uses several machine learning algorithms to screen data for vehicles that may exceed emission standards. It should be noted that the reason for using multiple machine learning algorithms is that a single machine learning algorithm has its limitations, making it prone to misjudgment when screening large amounts of data. Step 4 uses different machine learning algorithms to screen the corrected fuel emission factor data and marks data points that are identified as outliers by each machine learning algorithm for further analysis.

[0139] The threshold determination in step 5 takes into account that the effectiveness of machine learning algorithms is influenced not only by pre-set hyperparameters but also by measurement differences between different devices and the characteristics of the measurement data. Without a large amount of data samples to determine appropriate hyperparameters for a specific point device, the algorithm's output may be suboptimal. Therefore, further determining vehicles exceeding emission standards based on thresholds based on the output of several machine learning algorithms can reduce the possibility of false positives.

[0140] In step 5, when screening vehicles with excessive exhaust pollutant emissions, a targeted screening threshold should be determined. Taking diesel vehicles as an example, the current National Environmental Protection Standard of the People's Republic of China, "Measurement Methods and Technical Requirements for Exhaust Pollutants from In-Use Diesel Vehicles (Remote Sensing Detection Method)" (HJ 845-2017), sets a NOx emission limit of 1500 ppm for diesel vehicles. Therefore, when screening diesel vehicles with excessive NOx emissions, a multiple of this limit can be considered as the threshold. It should be noted that while steps 3 and 4 correct the fuel emission factor data and identify outliers, step 5 should filter the concentration data of the marked data points (the ID field or license plate field of the telemetry data can be traced to enable the mapping from fuel emission factor data to concentration data). Furthermore, due to errors between telemetry devices and variations in environmental factors such as wind speed and temperature at each device installation location, the threshold should be optimized based on the type of exhaust pollutant being screened and tailored to the devices at each location. A blanket threshold determination method will yield suboptimal results.

[0141] The following combination Figure 2 、 Figure 3 as well as Figure 4 , explain the correction and screening results.

[0142] First, let's explain the basics of the data. The raw data comes from two telemetry points (hereafter referred to as Point 1 and Point 2, respectively). Each data entry consists of 42 fields, including basic vehicle information, speed information as it passes through the telemetry point, and environmental factors during the equipment measurement. Point 1 contains 6,697 raw data entries, covering a period of 19 days, from July 11 to July 29, 2022; Point 2 contains 28,482 raw data entries, covering a period of 6 days, from July 13 to July 18, 2022. Figures 2 to 4 The correction and screening results proposed in the embodiment of the present invention will be demonstrated by taking the NO data of diesel vehicles at point 1 as an example.

[0143] Figure 2 The figure shows the NO fuel emission factor data distribution obtained after calculation of the measurement data at point 1 through steps 1 and 2. It can be seen that after converting the concentration index into the fuel emission factor index, the absolute value of the data is reduced, but it can better reflect the actual emission level of the vehicle.

[0144] Figure 3 The figure shows the result of decomposing the fuel emission factor data of point 1 using the STL time series decomposition algorithm. It can be seen that the algorithm can divide the fuel emission factor data into three parts. The trend term and the periodic term represent the degree of drift of the equipment measurement results. Therefore, after using the two to correct the biased data, it can be considered that the drift problem has been well solved.

[0145] Figure 4 Shown are the machine learning algorithm judgment results and threshold judgment results. Figure 4 (a) shows an example of the results of machine learning algorithm judgment. This example uses two machine learning algorithms, the Isolation Forest algorithm and the DBSCAN clustering algorithm. The principles and hyperparameter settings of the two algorithms are briefly described below:

[0146] The Isolation Forest algorithm is an unsupervised anomaly detection algorithm that identifies outliers as sparsely distributed points that are far from a denser sample population. During training, the one-dimensional Isolation Forest algorithm continuously randomly selects a split point between the maximum and minimum values in the sequence, splitting the data into two parts. This process is then recursively repeated to build a binary tree for both parts, until all data points are leaf nodes or the depth of the binary tree reaches a pre-set upper depth limit. Once the binary tree is established, the depth of outliers in the sample population is significantly lower than that of non-outliers. Therefore, the algorithm can identify outliers based on a pre-set average depth. In this example, 2% of the data points are considered outliers during the screening process.

[0147] The DBSCAN clustering algorithm is an unsupervised algorithm based on density clustering. The main hyperparameters of the algorithm are the neighborhood radius e and the minimum number of data points required to become a core object, MinPts. The algorithm randomly selects data points during the training process, defines the data points within the neighborhood radius e where the number of data points exceeds the minimum number MinPts as core objects, and demarcates all density-reachable samples of the core objects into a cluster. The algorithm continuously determines whether unprocessed data points are core objects until all data points are divided. In an embodiment of the present invention, most vehicles have low emission values, so they can be divided into a cluster using the DBSCAN clustering algorithm, and vehicles with excessive emissions are divided into other clusters or edge points. In this example, the neighborhood radius e is set to 20 and the minimum number of data points required to become a core object, MinPts, is set to 10 during the screening process.

[0148] Figure 4 (a) shows the NO fuel emission factor data distribution of all vehicles. The red points are the intersection of the outliers identified by the two machine learning algorithms. These data points are marked as unqualified data points and will undergo further threshold judgment. The blue points are qualified data points.

[0149] Figure 4 (b) shows the Figure 4The result graph after filtering out the unqualified data points in (a). It should be noted that before making the threshold determination, this example first converts the fuel emission factor data back into concentration data based on the ID of the unqualified data point. Then, twice the NO emission limit for diesel vehicles in the relevant industry standard (3000ppm) is set as the threshold and determination is made. Data points above the threshold line will be determined as the final vehicle data with excessive exhaust emissions.

[0150] Site 1 had 6,697 raw data points, 1,220 of which were valid for diesel vehicles. After data correction and screening, eight vehicles were found to have excessive exhaust emissions. Site 2 had 28,482 raw data points, 2,357 of which were valid for diesel vehicles. After data correction and screening, 58 vehicles were found to have excessive exhaust emissions. The screening results at both sites reflect the reality that most vehicles have normal emissions, while a small number have abnormal emissions.

[0151] In summary, the embodiment of the present invention proposes a method for correcting and screening drift data of motor vehicle exhaust telemetry. First, the concentration ratio of each type of exhaust pollutants to CO2 is calculated; then, the fuel emission factor of each type of exhaust pollutants is further calculated; then, the STL time series decomposition algorithm is used to divide the fuel emission factor data into three parts: trend item, period item, and fluctuation item, and the trend item data and period item data are used to correct the drift data; then, several machine learning algorithms are used to determine the abnormal points, and the data points determined as abnormal points by each machine learning algorithm are marked for threshold determination; finally, the marked data points are determined to have excessive emissions according to a pre-set threshold, and the final vehicle data with excessive emissions is obtained. The present invention combines the existing research on the calibration of exhaust telemetry data and the determination of vehicles with excessive emissions, and proposes a method for correcting and screening motor vehicle exhaust telemetry drift data, which can reasonably and effectively solve the data drift problem that restricts the application of motor vehicle exhaust telemetry technology and realize the automated determination of vehicles with excessive exhaust emissions.

[0152] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0153] Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise indicated, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It will also be understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the ordinary skill of an engineer. Therefore, a person skilled in the art using ordinary skill will be able to implement the present invention set forth in the claims without undue experimentation. It will also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0154] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0155] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0156] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0157] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0158] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0159] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0160] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for processing motor vehicle exhaust telemetry drift data, characterized in that: include: Calculate the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide; Calculating fuel emission factors for different types of motor vehicle exhaust pollutants based on the concentration ratios; Using the STL time series decomposition algorithm to divide the fuel emission factor into a trend item, a period item, and a fluctuation item, and correcting the data drift of the trend item and the period item to obtain corrected fuel emission factor data; Screening the corrected fuel emission factor data for outliers using several different machine learning models, and marking each outlier screened by the different machine learning models; According to the preset screening threshold, a threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

2. The method for processing vehicle exhaust telemetry drift data according to claim 1, characterized in that: The step of calculating the concentration ratios of different types of motor vehicle exhaust pollutants to carbon dioxide includes: Calculating a first concentration ratio of nitrogen monoxide to carbon dioxide in exhaust pollutants from a motor vehicle; calculating a second concentration ratio of hydrocarbons to carbon dioxide in exhaust pollutants from motor vehicles; Calculating a third concentration ratio between carbon monoxide and carbon dioxide in exhaust pollutants from motor vehicles; Wherein, the calculation formula of the first concentration ratio is: The calculation formula of the second concentration ratio is: The calculation formula of the third concentration ratio is: in, Represents the concentration ratio of NO to CO2; Represents the concentration ratio of hydrocarbons to CO2; Represents the concentration ratio of CO to CO2; C NO represents the measured concentration of NO; C HC Represents the measured concentration of hydrocarbons; C CO represents the measured concentration of CO; Represents the measured concentration of CO2.

3. The method for processing vehicle exhaust telemetry drift data according to claim 2, characterized in that: Calculating the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratios includes: Calculating a fuel emission factor of NO based on the first concentration ratio; calculating a fuel emission factor of hydrocarbons based on the second concentration ratio; Calculating a fuel emission factor for CO based on the third concentration ratio; The calculation formula of the NO fuel emission factor is: The calculation formula for the hydrocarbon fuel emission factor is: The calculation formula of the CO fuel emission factor is: Among them, EF NO represents the fuel emission factor of NO; EF HC represents the fuel emission factor of the hydrocarbon in question; EF CO represents the fuel emission factor of CO; constants 30, 44, and 28 are the molar masses of various pollutants respectively; M fuel represents the molar mass of the fuel type used; the constant 0.493 is the carbon mass conversion factor calculated based on propane.

4. The method for processing vehicle exhaust telemetry drift data according to claim 1, characterized in that: The STL time series decomposition algorithm is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term, and the data drift of the trend term and the period term is corrected to obtain corrected fuel emission factor data, including: According to the additive splitting rule, the STL time series decomposition algorithm based on Fourier transform is used to divide the fuel emission factor data into three parts: trend term, period term, and fluctuation term. The trend term and period term are used to characterize the degree of data drift caused by different calibration conditions and different operating time of the exhaust gas telemetry equipment. performing correction processing on the drift data using the trend item and the period item; The expression of the correction process is: IF A =EF O -IF T -IF S Among them, EF A For the corrected data; EF O is the drift data before correction; EF T is the trend item data; EF S It is the periodic item data.

5. The method for processing vehicle exhaust telemetry drift data according to claim 1, characterized in that: The method of screening the corrected fuel emission factor data for abnormal points using several different machine learning models and marking the abnormal points obtained by screening using different machine learning models includes: Using a first model constructed by an isolation forest algorithm and a second model constructed by a DBSCAN clustering algorithm, the corrected fuel emission factor data is screened for outliers to obtain outliers; The outliers obtained by screening the first model and the second model are marked respectively.

6. The method for processing vehicle exhaust telemetry drift data according to claim 1, characterized in that: The method of performing threshold determination on each of the abnormal points according to a preset screening threshold to determine the target motor vehicle with excessive emissions includes: Determine the screening threshold based on the fuel type and exhaust pollutant type of the vehicle to be screened, combined with the exhaust emission standard documents; According to the screening threshold, threshold determination is performed on each of the abnormal points to determine the target motor vehicles with excessive emissions.

7. A motor vehicle exhaust telemetry drift data processing device, characterized in that: include: The first module is used to calculate the concentration ratio of different types of motor vehicle exhaust pollutants to carbon dioxide; The second module is used to calculate the fuel emission factors of different types of motor vehicle exhaust pollutants based on the concentration ratio; The third module is used to divide the fuel emission factor into a trend term, a period term, and a fluctuation term using an STL time series decomposition algorithm, and correct the data drift of the trend term and the period term to obtain corrected fuel emission factor data; A fourth module is configured to screen the corrected fuel emission factor data for outliers using several different machine learning models, and mark each outlier screened by different machine learning models; The fifth module is used to perform threshold determination on each of the abnormal points according to a preset screening threshold, and determine the target motor vehicles with excessive emissions.

8. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The storage medium stores a program, and the program is executed by a processor to implement the method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images