Vehicle supervision method and device based on OBD data

Abstract

The invention discloses a vehicle supervision method and device based on OBD data, and the method comprises the steps: marking the OBD data according to the data type of the output data of a downstream nitrogen oxide (NOx) sensor of a selective catalytic reduction (SCR) technology, carrying out the sorting of the marked OBD data according to the time sequence, comparing the data types of two adjacent OBD data, and carrying out the comparison of the data types of the two adjacent OBD data. According to the method, the first conversion proportion information of the number of the abnormal values converted from the abnormal values to the normal values accounts for the abnormal values is determined, according to the counted first conversion proportion information and the preset conversion proportion threshold value information, abnormal OBD judgment is achieved, and then vehicle abnormity supervision is achieved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to emission monitoring technology, and in particular to a vehicle supervision method and device based on OBD data. BACKGROUND

[0002] Heavy-duty diesel vehicles are the main contributor to urban nitrogen oxide (NOx) emissions, and their emissions pose a significant challenge to the environment, air quality, and human health. Therefore, it is crucial to control the NOx emissions of diesel vehicles. According to the China Mobile Source Environmental Management Annual Report, heavy-duty diesel vehicles contribute more than 70% of NOx emissions in urban transportation sources in China. In recent years, China has tightened emission limits for heavy-duty vehicles to reduce heavy-duty vehicle emissions; since July 2023, China has implemented the 6b stage of the national VI emission standard for heavy-duty diesel vehicles nationwide, requiring heavy-duty trucks to install remote online terminals for on-board diagnostics (OBD) to send OBD monitoring data to the supervision platform in real time, in order to effectively supervise the NOx emissions of heavy-duty vehicles in use. Research shows that the OBD remote emission monitoring technology has high consistency and small deviation with the emission results obtained by the vehicle test system (PEMS) in the regulatory test method, which can provide effective support for supervision. X

[0003] However, there are still problems to be solved in the practical application of remote OBD monitoring technology, mainly reflected in the quality of OBD data; in the actual operation of the vehicle, the quality of the OBD data uploaded is easily affected by system recording failures and upload failures, etc., resulting in missing or repeated key parameters of OBD data, affecting the calculation of vehicle NOx emission levels and directly affecting the effectiveness of vehicle supervision. X

[0004] In summary, remote OBD monitoring technology provides important support for the actual road emission supervision of national VI heavy-duty diesel vehicles, but system recording failures and upload failures, etc. easily affect the quality of OBD data, affecting the accuracy of vehicle supervision. SUMMARY

[0005] The embodiments of the present application provide a vehicle supervision method based on OBD data, comprising: According to the data type of the output data of the nitrogen oxide NOx sensor downstream of the selective catalytic reduction technology SCR, the on-board diagnostic system OBD data obtained in a preset time period is marked; wherein the data type includes: normal value and multiple types of abnormal value; the multiple types of abnormal value include one or any combination of the following: missing value and repeated value; ​​After sorting the marked OBD data in chronological order, the data types marked by each two adjacent OBD data are compared respectively, if the previous OBD data is marked as an abnormal value and the next OBD data is marked as a normal value, the count value corresponding to the type of abnormal value is added by 1; After comparing all OBD data, for each type of abnormal value, the count value corresponding to the type of abnormal value is divided by the total number of OBD data marked as the type of abnormal value to obtain the first transition proportion information of the type of abnormal value. According to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value, whether the OBD device is abnormal is determined. When the OBD device is determined to be abnormal, the vehicle is determined to be an abnormal vehicle.

[0006] On the other hand, the embodiment of the present application also provides a computer storage medium, the computer storage medium stores a computer program, and the computer program is executed by a processor to realize the vehicle supervision method based on OBD data.

[0007] In another aspect, the embodiment of the present application also provides a vehicle supervision device based on OBD data, comprising: a marking unit, a comparison and counting unit, a statistical proportion unit, a judgment unit and a determination unit; wherein, The marking unit is configured to mark the OBD data obtained in a preset time period according to the data type of the output data of the SCR downstream NOx sensor, wherein the data type includes normal values and a plurality of types of abnormal values; the plurality of types of abnormal values include one or any combination of the following: missing values and repeated values. The comparison and counting unit is configured to sort the marked OBD data in chronological order, compare the data types marked by each two adjacent OBD data respectively, if the previous OBD data is marked as an abnormal value and the next OBD data is marked as a normal value, the count value corresponding to the type of abnormal value is added by 1. The statistical proportion unit is configured to divide the count value corresponding to each type of abnormal value by the total number of OBD data marked as the type of abnormal value after comparing all OBD data to obtain the first transition proportion information of the type of abnormal value. The judgment unit is configured to determine whether the OBD device is abnormal according to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value. The determination unit is configured to determine the vehicle to be an abnormal vehicle when the OBD device is determined to be abnormal.

[0008] The embodiment of the present disclosure labels the OBD data according to the data type of the output data of the NOx sensor downstream of the SCR, sorts the labeled OBD data according to time, compares the data types of two adjacent OBD data, determines the first transition proportion information of each abnormal value from abnormal value to normal value, and determines the abnormal OBD according to the statistical first transition proportion information and the preset transition proportion threshold information, thereby realizing the vehicle abnormal supervision.

[0009] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. Other advantages of the present application can be realized and obtained by means of the instrumentalities and combinations pointed out in the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings are included to provide an understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the technical scheme of the present application, and do not constitute a limitation on the technical scheme of the present application.

[0011] Figure 1 The flowchart of the vehicle supervision method based on OBD data of the embodiment of the present disclosure is shown in the figure; Figure 2 The reference transition proportion information of normal data of the embodiment of the present disclosure is shown in the figure; Figure 3 The proportion of each data type under different SCR outlet temperatures of the embodiment of the present disclosure is shown in the figure; Figure 4 The structural block diagram of the vehicle supervision device based on OBD data of the embodiment of the present disclosure is shown in the figure; Figure 5 The comparison result before and after data quality control of the embodiment of the present disclosure is shown in the figure. DETAILED DESCRIPTION

[0012] The present application describes a plurality of embodiments, but the description is exemplary rather than limiting, and it is obvious to those skilled in the art that there can be more embodiments and implementation schemes within the scope of the embodiments described in the present application. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment can be used with any other feature or element of any other embodiment, or can replace any other feature or element of any other embodiment.

[0013] The present application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features, and elements disclosed herein can also be combined with any conventional feature or element to form a unique invention. Any feature or element of any embodiment can also be combined with features or elements from other inventive solutions to form another unique inventive solution. Therefore, it should be understood that any feature shown and / or discussed in the present application can be implemented alone or in any appropriate combination. Accordingly, the embodiments are not to be restricted, except as by the appended claims and their equivalents. Moreover, various modifications and changes can be made within the scope of the claims.

[0014] Furthermore, in describing representative embodiments, the specification can have presented the method and / or process as a particular sequence of steps. However, to the extent that the method or process depends on more than one step, the method or process should not be limited to the particular sequence of steps described. Other sequences of steps can also be possible. The particular sequence of steps described should not be construed as being limiting of the claims. Furthermore, the claims should not be limited to the steps of the method and / or process in the order in which they are written, as other sequences of steps can also be possible, and the particular sequence of steps described should not be construed as limiting of the claims. The claims should not be limited by the steps of the methods and / or processes in this description, but can include any other steps or combinations of steps, and are entitled to any benefits inherent therein.

[0015] Figure 1 A flowchart of the vehicle supervision method based on OBD data according to an embodiment of the present disclosure is shown in FIG. 1, which includes the following steps. Figure 1 Step 101, marking the OBD data obtained in a preset time period according to the data type of the output data of the selective catalytic reduction (SCR) downstream nitrogen oxide (NOx) sensor, wherein the data type includes normal value and multiple types of abnormal value, and the multiple types of abnormal value include one or any combination of the following: missing value and repeated value; Step 102, after sorting the marked OBD data in chronological order, comparing the data types marked by each two adjacent OBD data respectively, if the previous OBD data is marked as abnormal value and the next OBD data is marked as normal value, then the count value corresponding to the corresponding type of abnormal value is increased by 1; Step 103, after comparing all OBD data, for each type of abnormal value, dividing the count value corresponding to the type of abnormal value by the total number of OBD data marked as the type of abnormal value to obtain the first transition proportion information of the type of abnormal value; Step 104, determining whether the OBD device is abnormal according to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value.​ Step 105, when determining that the OBD device is abnormal, determining the vehicle as an abnormal vehicle.

[0016] The embodiment of the present disclosure labels the OBD data according to the data type of the NOx sensor output data downstream of the SCR, sorts the labeled OBD data in chronological order, compares the data types of adjacent two OBD data, determines the first transition proportion information of the number of each abnormal value from abnormal value to normal value, and determines the abnormal OBD according to the statistical first transition proportion information and the preset transition proportion threshold information, thereby realizing the vehicle abnormal supervision.

[0017] In an exemplary example, the embodiment of the present disclosure determines whether the OBD device is abnormal according to the statistical first transition proportion information of each abnormal value and the preset transition proportion threshold information corresponding to each abnormal value, comprising: For each abnormal value, if the first transition proportion information of the abnormal value is greater than or equal to the transition proportion threshold information corresponding to the abnormal value, it is determined that the OBD device is abnormal.

[0018] In an exemplary example, the abnormal value of the embodiment of the present disclosure also includes the out-of-range value, and the vehicle supervision method of the embodiment of the present disclosure further comprises: After sorting the labeled OBD data in chronological order, the data types of adjacent two OBD data are compared respectively, if the previous OBD data is labeled as an out-of-range value and the next OBD data is labeled as a normal value, the counting value of the out-of-range value to normal value is increased by 1, if the previous OBD data is labeled as a normal value and the next OBD data is labeled as an out-of-range value, the counting value of the normal value to out-of-range value is increased by 1; After comparing all OBD data, the counting value of the out-of-range value to normal value is divided by the total number of OBD data labeled as an out-of-range value to obtain the second transition proportion information of the out-of-range value to normal value, and the counting value of the normal value to out-of-range value is divided by the total number of OBD data labeled as a normal value to obtain the third transition proportion information of the normal value to out-of-range value; According to the statistical second transition proportion information and the third transition proportion information, it is determined whether the OBD device is abnormal.

[0019] In an exemplary example, the embodiment of the present disclosure determines whether the OBD device is abnormal according to the statistical second transition proportion information and the third transition proportion information, comprising: If the second transition proportion information is greater than or equal to the preset transition proportion threshold information corresponding to the out-of-range value to normal value, it is determined that the OBD data is not the data recorded by the OBD device in the cold start stage, and the OBD device is determined to be abnormal when the OBD data is not the data recorded by the OBD device in the cold start stage. When the third transition proportion information is greater than or equal to the transition proportion threshold information corresponding to the transition from the normal value to the out-of-range value, it is determined that the OBD device is abnormal.

[0020] When the second transition proportion information is greater than or equal to the transition proportion threshold information corresponding to the transition from the out-of-range value to the normal value, it is determined that the OBD data is not the data recorded by the OBD device in the cold start phase, which indicates that the out-of-range value appears when the vehicle is normally running. At this time, it is determined that the OBD device is abnormal.

[0021] In an exemplary example, the embodiments of the present disclosure include: When the data type includes a missing value, the missing value is the output of the SCR downstream NOx sensor being empty; When the data type includes an out-of-range value, the out-of-range value is that the output of the SCR downstream NOx sensor exceeds the actual normal range. Preliminary analysis of the sample vehicle data shows that there are some data records close to or equal to the threshold value (such as NOx = 3012.75 ppm or -200 ppm), which deviates significantly from the normal working condition of the vehicle and belongs to data anomaly. Therefore, the present disclosure tightens the effective range to -100-2500 ppm; When the data type includes a repeated value, the repeated value is that the output of the SCR downstream NOx sensor is within the effective range (-100-2500 ppm), but the data record value remains unchanged for 10 times or more in a row. This phenomenon is caused by sensor failure or decoding anomaly, and is not the real emission situation of the heavy-duty vehicle, so it is classified as abnormal. The above three types of data are collectively referred to as abnormal values, which cannot reflect the real emission situation of the vehicle. In addition, the remaining data records are referred to as normal values. In the use of OBD data to calculate the vehicle NOx emission factor, the present disclosure first eliminates the abnormal value data and only retains the normal value for calculation to ensure the accuracy of the result.

[0022] In an exemplary example, before determining whether the OBD device is abnormal according to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value, the method of the present disclosure further includes: Statistically determining the reference transition proportion information of the mutual transition between different data types of the pre-determined normal data, wherein the reference transition proportion information of the following transition paths of the normal data is all greater than or equal to the pre-set reference transition proportion threshold: missing value-repeated value-out-of-range value-normal value-missing value; Based on the determined reference transition proportion information, set the transition proportion threshold information of the OBD device abnormality; The reference transition proportion information includes: transition proportion information of missing values transitioning to normal values, transition proportion information of repeated values transitioning to normal values, transition proportion information of out-of-range values transitioning to normal values, and transition proportion information of normal values transitioning to out-of-range values.

[0023] In an exemplary instance, the reference transition proportion threshold value can be 10%; the reference transition proportion information of the present embodiment is the same as the statistical manner of the first transition proportion information, and the present embodiment does not repeat the description. Figure 2 A reference transition proportion information diagram of normal data of the present embodiment is shown in the figure, which shows the reference transition proportion information on the transition path of missing values-repeated values-out-of-range values-normal values-missing values, and also shows part of other reference transition proportion information. The present embodiment takes normal data as a reference to analyze the reference transition proportion information of mutual transition between different data types of normal data, and sets the above transition proportion threshold value information. Based on the time sequence correlation of continuous data types, the present embodiment can analyze the transformation rule of different types of data by constructing a directed graph, perform analysis and statistics, and determine the OBD data scanned piece by piece; if the data type (i) of the current OBD data is different from the data type (j) of the next data, it is determined that the data type has transitioned, and the subsequent record matrix F[i, j]+1; after all the trips of the vehicle sample are scanned, for each data type, the proportion of each subsequent data type is counted: for data type i, the proportion of its successor j_0 is calculated as: G[i, j_0]=F[i, j_0] / ∑_j(F[i, j]); all G[i, j]>10% are screened out, and a directed graph with proportions is drawn; if only the transformation rule between different types of data is concerned, when the two continuous data types are different, the next type is defined as the successor of the previous type; for a certain data type, the successor distribution can reflect the tendency of the type transitioning to other types; the present embodiment pre-selects normal data as sample data, counts the type proportion of each successor of different data types in all sample data, and extracts the main transformation path with a proportion exceeding 10%, as shown in Figure 2For example, 70% of the successors of the missing value are repeated values, and 30% of the successors are normal values. Except that the main successor of the normal value accounts for 54%, the maximum proportion of other types accounts for 70%. By retaining only the maximum proportion of the successor (solid arrow) and ignoring other secondary paths, a closed-loop path of "missing value-repeated value-out-of-range value-normal value-missing value" can be formed, reflecting the fixed form of transformation between different data types. Among the repeated values, since the proportion of 0 values is as high as 99.8%, the determination method can be optimized to include 0 values with less than 10 consecutive repeated times after the missing value x into the repeated value category. After inspection, 65.7% of the normal trips in the entire data after the determination optimization conform to the above definition, indicating that the transformation rule has significant universality in the sample vehicle. Data records that do not conform to the transformation rule are more likely to occur in the case where the vehicle data record itself is abnormal, so the transformation rule can be used as an auxiliary judgment basis for judging the normality of the vehicle OBD data record.

[0024] In an exemplary example, the embodiments of the present disclosure: When the abnormal value includes the missing value, the reference transition proportion information A_0 is the transition proportion information of the missing value of the normal data to the normal value, and the transition proportion threshold information corresponding to the missing value is A_1: A_1=A_0+(20±5)%;The embodiments of the present disclosure assume that the reference transition proportion information A_0 is 30%, and on this basis, the formula (20±5)% is set to calculate the transition proportion threshold information as: A_1=30%+(20±5)%; When the abnormal value includes the repeated value, the reference transition proportion information B_0 is the transition proportion information of the repeated value of the normal data to the normal value, and the transition proportion threshold information corresponding to the repeated value is B_1: B_1=B_0+(10±5)%;The embodiments of the present disclosure assume that the reference transition proportion information B_0 is close to 0%, for example, 1%, and on this basis, the formula (10±5)% is calculated to obtain the transition proportion threshold information as: B_1=1%+(10±5)%; When the abnormal value includes the out-of-range value, the reference transition proportion information C_0 is the transition proportion information of the out-of-range value of the normal data to the normal value, and the transition proportion threshold information corresponding to the out-of-range value to the normal value is C_1: C_1=C_0-(15±5)%;The embodiments of the present disclosure assume that the reference transition proportion information C_0 is close to 100%, for example, 100%, and on this basis, the formula (15±5)% is calculated to obtain the transition proportion threshold information as: C_1=100%-(15±5)%; The abnormal value includes an out-of-bound value. When the reference transition proportion information D_0 is normal value transition proportion information of a normal value of normal data transitioning to an out-of-bound value, the transition proportion threshold information corresponding to the transition of the normal value to the out-of-bound value is D_1: D_1=D_0+(10±5)%. The embodiment of the present disclosure assumes that the reference transition proportion information D_0 is close to 10%, and takes 10% as an example. On this basis, the transition proportion threshold information is calculated according to the above formula: D_1=10%+(10±5)%.

[0025] When the first transition proportion information of the missing value transitioning to the normal value is greater than or equal to A_1, the OBD device is determined to be abnormal. The transition process may occur in a non-cold start link, that is, the missing value appears during normal driving, and the OBD device has a problem. When the first transition proportion information of the repeated value transitioning to the normal value is greater than or equal to B_1, the OBD device is determined to be abnormal. The transition process may occur in a non-cold start link, that is, the missing value appears during normal driving, and the OBD device has a problem. When the second transition proportion information is greater than or equal to C_1, it is determined that the OBD data is not data recorded by the OBD device in the cold start stage, indicating that the out-of-bound value appears during normal driving of the vehicle. At this time, it is determined that the OBD device has a problem. When the third transition proportion information is greater than or equal to D_1, it indicates that the proportion of the transition from the normal value to the out-of-bound value is too high, that is, the out-of-bound value appears during normal driving of the vehicle. At this time, the OBD device has a problem.

[0026] In an exemplary example, when the OBD device is determined to be abnormal, the OBD data of the OBD device determined to be abnormal is determined to be invalid data.

[0027] In an exemplary example, the method of the embodiment of the present disclosure further includes: OBD data with an interval between adjacent two OBD data less than or equal to a pre-set continuous determination threshold is divided into a trip; When the number of OBD data contained in the divided trip is less than or equal to a pre-set number threshold or the total duration of the trip is less than or equal to a pre-set duration threshold, the trip is determined to be a short trip; The OBD data determined to be a short trip is deleted.

[0028] The disclosure embodiment can set the trip continuity determination time threshold to Tc=300s, if the collection time interval At of the adjacent two OBD data records of the same vehicle is less than or equal to Tc, the two OBD data records are divided into one trip; the collection time interval of the adjacent two OBD data records is constant at 10s when the sample vehicle OBD system works normally, and the actual operation may be affected by factors such as equipment failure or transmission anomaly, which may cause At to increase significantly. The purpose of setting Tc=300s in the disclosure is to improve the fault tolerance of the determination, to include the interruption caused by data failure / transmission anomaly, and to avoid misjudging it as a parking event, so as to distinguish the vehicle driving trip from the parking event. The disclosure embodiment sets the number threshold to 19 and the time threshold to 180 seconds. If the number of OBD data records in the trip is less than 19 or the total duration is less than 180s, the trip is determined to be a short trip, the short trip is deleted, and is excluded in subsequent analysis to avoid interference of the short-time behavior of the vehicle; otherwise, it is determined to be a normal trip and is retained. The disclosure embodiment marks the starting point and the ending point of the normal trip in the trip, so as to process the subsequent processing in the unit of the normal trip, that is, a segment of the actual driving process of the vehicle. Each data record defines the record order in the unit of the trip, for example, the record order of the first data record of all trips is 1.

[0029] According to the above closed-loop conversion formula, the state of the vehicle OBD device is associated with the type of data recorded by it. For example, at the beginning of the trip, after the OBD device of the vehicle is started, the type of data recorded will gradually change from missing value to normal value under normal circumstances. In addition to time, the starting state of the vehicle is also reflected in the SCR outlet temperature of the OBD data record, especially in the cold start link. Therefore, the disclosure embodiment analyzes the data quality in combination with the SCR outlet temperature, as shown in Figure 3 , the proportion of each data type under different SCR outlet temperatures is counted; on the overall trend, the normal value gradually increases and the repeated value gradually decreases as the SCR outlet temperature increases, and the out-of-range value mainly appears at 100-200℃. Specifically, it can be divided into three stages: stage 1): when the SCR outlet temperature is less than 75℃, the repeated value remains at a high level, and there is basically no out-of-range value, indicating that the vehicle has just started; stage 2): when the SCR outlet temperature exceeds 75℃, the out-of-range value begins to appear, the repeated value rapidly decreases to below 20%, and the normal value rapidly rises to above 80%, indicating that the OBD device starts to record data normally; stage 3): when the SCR outlet temperature exceeds 200℃, the normal value rapidly rises to nearly 100%, indicating that the vehicle has completely exited the cold start stage. Because the temperature of the SCR system is low during the cold start of the vehicle, the sensors in the system have not reached the working condition, resulting in a large number of abnormal values. The change of "repeated value-out-of-range value-normal value" is the sign of the completion of the cold start of the vehicle, which can provide a basis for monitoring the cold start stage of the vehicle.

[0030] In one exemplary instance, this disclosure embodiment determines whether OBD data is data recorded by the OBD device during the cold start phase, including: For the OBD data after the trip is divided, filter out the data segments that meet the transition pattern of duplicate value - out-of-bounds value - normal value. Among them, the number of OBD data marked as duplicate value before the OBD data marked as out-of-bounds value in the data segment is greater than or equal to the first preset number, and the number of OBD data marked as normal value after the OBD data marked as out-of-bounds value is greater than or equal to the second preset number. The NOx outlet temperature corresponding to all out-of-bounds values ​​in all data segments is statistically analyzed to obtain the temperature threshold range for the NOx sensor downstream of the vehicle's SCR to start working normally; the time taken from the start of the trip to the first normal value in the cold start data segment is used to obtain the vehicle's cold start time range. OBD data that meets the defined temperature threshold range and consumption time range are identified as OBD data recorded by the OBD device during the cold start phase.

[0031] For the vehicle's divided routes, this embodiment of the disclosure takes an example where both the first preset number of routes and the second preset number of routes are 5, and selects the data segments that satisfy "repetition value - out-of-bounds value - normal value". In this embodiment of the disclosure, the data segment with the most OBD entries marked as out-of-bounds values ​​can also be selected from the filtered data segments as the data segment for statistical temperature threshold range. The statistical data range can be analyzed and set by technicians. Within the stroke, for a whole segment of the maximum consecutive out-of-bounds values ​​(maximum consecutive means: if the length is n and the data sequence number is k to k+n-1, it must satisfy that [NOx]_k to [NOx]_{k+n-1} are all out-of-bounds values, and [NOx]_{k-1} and [NOx]_{k+n} are not out-of-bounds values), the five consecutive records ([NOx]_{k-5} to [NOx]_{k-1}) preceding the first record ([NOx]_k) of this segment are all repeated values, and the five consecutive records ([NOx]_{k+n} to [NOx]_{k+n+4}) following the last record ([NOx]_{k+n-1}) of this segment are all normal values. After confirming all the data segments of "repeated value - out-of-bounds value - normal value" for the vehicle, the NOx outlet temperature corresponding to the out-of-bounds value in all segments can be counted to obtain the temperature threshold range for the NOx sensor downstream of the vehicle's SCR to start working normally; the time taken from the start of each stroke to the first normal value in the segment can be counted to obtain the time range for the vehicle's cold start.

[0032] This embodiment of the disclosure uses two parameters—data type (whether it is an out-of-bounds value in the normal conversion path) and SCR outlet temperature—to jointly determine whether the vehicle is in a cold start state, thereby improving the accuracy and robustness of vehicle cold start judgment.

[0033] This disclosure also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the above-described vehicle monitoring method based on OBD data.

[0034] This disclosure also provides a terminal, including: a memory and a processor, wherein the memory stores a computer program; wherein, The processor is configured to execute computer programs in memory; When the computer program is executed by the processor, it implements the vehicle monitoring method based on OBD data as described above.

[0035] Figure 4 This is a structural block diagram of a vehicle monitoring device based on OBD data according to an embodiment of this disclosure, as shown below. Figure 4 As shown, it includes: a marking unit, a comparison and counting unit, a statistical proportion unit, a judgment unit, and a determination unit; wherein, The marking unit is set to mark the on-board diagnostic system (OBD) data acquired within a preset time period according to the data type of the NOx sensor output data downstream of the selective catalytic reduction (SCR) technology, which is pre-divided; wherein, the data type includes: normal values ​​and various types of abnormal values; the various types of abnormal values ​​include one or any combination of the following: missing values ​​and duplicate values; The comparison counting unit is set as follows: after sorting the marked OBD data in chronological order, the data types marked on each of the two adjacent OBD data are compared. If the previous OBD data is marked as an outlier and the next OBD data is marked as a normal value, the count value corresponding to the outlier type is incremented by 1. The statistical proportion unit is set as follows: After comparing all OBD data, for each type of outlier, the count value corresponding to the outlier is divided by the total number of OBD data entries marked as that outlier to obtain the first change proportion information of that outlier. The judgment unit is set to determine whether the OBD device is abnormal based on the first conversion percentage information of each type of abnormal value and the pre-set conversion percentage threshold information corresponding to each type of abnormal value. The determination unit is set to: when it is determined that the OBD device is abnormal, the vehicle is determined to be an abnormal vehicle.

[0036] In one exemplary instance, the determination unit of this disclosure embodiment is configured as follows: For each type of outlier, if the first change percentage information of the outlier is greater than or equal to the change percentage threshold information corresponding to the outlier, the OBD device is determined to be abnormal.

[0037] In one exemplary instance, outliers in this disclosure also include out-of-bounds values; The comparison counting unit is also set to: after sorting the marked OBD data in chronological order, compare the data types marked on each of the two adjacent OBD data. If the previous OBD data is marked as an out-of-bounds value and the next OBD data is marked as a normal value, then the count value for the out-of-bounds value to the normal value is incremented by 1. If the previous OBD data is marked as a normal value and the next OBD data is marked as an out-of-bounds value, then the count value for the normal value to the out-of-bounds value is incremented by 1. The statistical proportion unit is also set as follows: after comparing all OBD data, the count of out-of-bounds values ​​turning into normal values ​​is divided by the total number of OBD data entries marked as out-of-bounds values ​​to obtain the second proportion information of out-of-bounds values ​​turning into normal values; the count of normal values ​​turning into out-of-bounds values ​​is divided by the total number of OBD data entries marked as normal values ​​to obtain the third proportion information of normal values ​​turning into out-of-bounds values. The judgment unit is also set to determine whether the OBD device is malfunctioning based on the statistical information of the second and third transformation ratios.

[0038] In one exemplary instance, the determination unit of this disclosure embodiment is further configured to determine whether the OBD device is malfunctioning based on the statistical second transformation ratio information and the third transformation ratio information, including: If the second conversion ratio information is greater than or equal to the pre-set conversion ratio threshold information corresponding to the conversion from the out-of-bounds value to the normal value, it is determined whether the OBD data is the data recorded by the OBD device during the cold start phase. If the OBD data is not the data recorded by the OBD device during the cold start phase, it is determined that the OBD device is abnormal. When the third transformation ratio information is greater than or equal to the preset normal value transformation to the transformation ratio threshold information corresponding to the out-of-bounds value, the OBD device is determined to be abnormal.

[0039] In one exemplary instance, the determining unit of this disclosure embodiment is further configured to determine whether the OBD data is data recorded by the OBD device during the cold start phase, including: For the OBD data after the trip is divided, filter out the data segments that meet the transition pattern of duplicate value - out-of-bounds value - normal value. Among them, the number of OBD data marked as duplicate value before the OBD data marked as out-of-bounds value in the data segment is greater than or equal to the first preset number, and the number of OBD data marked as normal value after the OBD data marked as out-of-bounds value is greater than or equal to the second preset number. The NOx outlet temperature corresponding to all out-of-bounds values ​​in all data segments is statistically analyzed to obtain the temperature threshold range for the NOx sensor downstream of the vehicle's SCR to start working normally; the time taken from the start of the trip to the first normal value in the cold start data segment is used to obtain the vehicle's cold start time range. OBD data that meets the defined temperature threshold range and consumption time range are identified as OBD data recorded by the OBD device during the cold start phase.

[0040] In one exemplary instance, the vehicle monitoring device of this disclosure further includes a threshold setting unit, configured as follows: The reference transformation percentage information for the mutual transformation between different data types of normal data is statistically determined. Among them, the reference transformation percentage information for the following transformation paths of normal data is greater than or equal to the pre-set reference transformation percentage threshold: missing value - duplicate value - out-of-bounds value - normal value - missing value. Based on the determined reference conversion ratio information, set the conversion ratio threshold information for OBD device anomalies; The reference transformation percentage information includes: the percentage of missing values ​​that have been transformed into normal values, the percentage of duplicate values ​​that have been transformed into normal values, the percentage of out-of-bounds values ​​that have been transformed into normal values, and the percentage of normal values ​​that have been transformed into out-of-bounds values.

[0041] In one exemplary instance, this disclosure embodiment: When outliers include missing values, the reference transformation percentage information A_0 is the percentage of missing values ​​that have been transformed into normal values. The threshold information for the transformation percentage of missing values ​​is A_1: A_1 = A_0 + (20 ± 5)% %. When outliers include duplicate values, the reference transformation percentage information B_0 represents the percentage of duplicate values ​​that have transformed into normal values. The threshold information for the transformation percentage of duplicate values ​​is B_1: B_1 = B_0 + (10 ± 5)% %. Outliers include out-of-bounds values. Referring to the transformation percentage information C_0, which represents the percentage of out-of-bounds values ​​that transform into normal values, the threshold information for the transformation percentage from out-of-bounds to normal values ​​is C_1: C_1 = C_0 - (15 ± 5)%; Outliers include out-of-bounds values. When referring to the transformation ratio information D_0, which is the transformation ratio information of normal data from normal values ​​to out-of-bounds values, the transformation ratio threshold information corresponding to the transformation from normal values ​​to out-of-bounds values ​​is D_1: D_1=D_0+(10±5)%.

[0042] In one exemplary instance, the vehicle monitoring device of this disclosure further includes a trip screening unit, configured as follows: OBD data with an acquisition time interval of less than or equal to a pre-set continuous judgment threshold are divided into one run. If the number of OBD data entries in a segmented trip is less than or equal to a preset threshold, or if the total duration of the trip is less than or equal to a preset duration threshold, the trip is determined to be a short trip. Delete OBD data that was identified as a short trip.

[0043] The following application examples briefly illustrate the embodiments of this disclosure. These application examples are only used to illustrate the embodiments of this disclosure and are not intended to limit the scope of protection of the embodiments of this disclosure.

[0044] Application Examples This disclosure, based on OBD remote emission monitoring data from 98 China VI heavy-duty diesel vehicles, analyzes and finds that the operating status of the OBD device affects the frequency of data quality issues. Problematic data exhibits a closed-loop transformation path of "missing values ​​- duplicate values ​​- out-of-bounds values ​​- normal values ​​- missing values," resulting in a higher frequency of data quality problems during the cold start phase. Compared to the OBD data of abnormal vehicles identified as invalid in this disclosure, the NOx emission factor of high-emission heavy-duty vehicles that were not processed for abnormal vehicle identification was underestimated by an average of 25.5%, demonstrating the necessity of this disclosure for assessing the actual road emission levels of heavy-duty vehicles and conducting precise supervision.

[0045] This embodiment is based on remote OBD monitoring data collected from 98 China VI heavy-duty diesel vehicles. All vehicles involved have pre-installed OBD systems obtained from the vehicle manufacturers. Specific information is shown in Table 1.

[0046] Table 1 After calculating the emission rate information in this embodiment, the unit fuel emission factor and the unit distance emission factor are calculated with reference to related technologies; wherein, the unit power emission factor is calculated by combining the emission rate information with Ne and R.

[0047] Calculate NOx emission rate NOx emission rate is the basis for calculating emission factor. The method for converting the second-by-second NOx concentration in the OBD data stream into NOx emission rate is as follows: ; ; ; in, NOx emission rate, in g / s; The exhaust gas density can be replaced by the air density, which is taken as 1.293. [NOx] represents the NOx sensor output value downstream of SCR post-processing in OBD data, which is the NOx volume concentration in ppmv; Q exhaust This refers to the exhaust gas flow rate, expressed in kg / h. Fuel flow rate is denoted by MAF, and intake air flow rate is denoted by MAF; both are measured in kg / h. Fuel flow rate (engine fuel volumetric flow rate), unit: ; The fuel density is taken as 0.85 kg / L for diesel fuel.

[0048] Unit fuel emission factor is (Unit: g / kg-fuel), the calculation formula is as follows: ; Emission factor per unit distance (Unit: g / km), the calculation formula is as follows: ; Where d= d represents the total distance traveled, in km; v represents the vehicle speed, in km / h. It is a time interval, measured in seconds (s). The emission factor per unit power is The calculation formula is as follows: ； ; Where P represents the work done by the engine, measured in kWh; Engine speed, unit: ; R is the engine's rated torque, in N·m; R is the engine's net output torque, in %.

[0049] To verify the impact of the method of this disclosure embodiment on the calculated emission factor, the NOx unit fuel emission factor (g / kg-fuel) of heavy-duty vehicles before and after data quality control of this disclosure embodiment (before and after processing by the method of this disclosure embodiment) is compared. Figure 5 This is a schematic diagram showing the comparison results of data quality control before and after the present disclosure embodiments, as shown below. Figure 5As shown, the overall average deviation level of the vehicles was 8.3%. The NOx emission factors of most vehicles were relatively similar, while the NOx emission factors of some low-emission vehicles and high-emission vehicles showed significant deviations before and after quality control. Compared to the emission factors after data quality control, without a quality control process, the emission factors of low-emission vehicles with emission factors below 5 g / kg fuel were mostly overestimated, with an average overestimation of 35.6%. Meanwhile, emission factors greater than or equal to 19 g / kg fuel were significantly overestimated. The emission factors of high-emission vehicles (g / kg fuel) are underestimated, with an average underestimation of 25.5%. Due to the aforementioned closed-loop transformation formula, a large number of duplicate zero values ​​and out-of-bounds values ​​may exist before the vehicle's cold start phase is complete. Failure to filter these out will lead to incorrect calculations of vehicle emission levels. Specifically, the emission factors of low-emission vehicles are mainly overestimated due to out-of-bounds values, while the emission factors of high-emission vehicles are mainly underestimated due to duplicate values. Using the method described in this disclosure will improve the sensitivity of identifying high-emission vehicles, effectively enhancing the regulatory effect of high-emission vehicles. Simultaneously, it will reduce misjudgments of low-emission vehicles meeting the China VI emission standard, improve the accuracy of vehicle emission level calculations, and thus achieve precise assessment of vehicle emissions.

[0050] In this embodiment, the remote OBD data type exhibits a closed-loop transformation path of "missing value - duplicate value - out-of-bounds value - normal value - missing value" for most of the journey, which can be used as a diagnostic basis for the OBD device status. During the cold start phase of the vehicle, the data quality is poor due to the sensors not reaching their working conditions. It only becomes normal after completing the transformation of "duplicate value - out-of-bounds value - normal value", which can be used as a judgment basis for the cold start phase. When the remote OBD data is not quality controlled, the interference of duplicate 0 values ​​during the cold start phase causes the NOx emission factor of high-emission vehicles to be underestimated by an average of 25.5%, while the interference of out-of-bounds values ​​causes the NOx emission factor of low-emission vehicles to be overestimated by an average of 35.6%. Therefore, data quality control is necessary when calculating vehicle emission levels in actual supervision.

[0051] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term "computer storage medium" includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

Claims

1. A vehicle supervision method based on OBD data, characterized by, The method comprises the following steps: According to the pre-divided data type of the output data of the nitrogen oxide NOx sensor downstream of the selective catalytic reduction technology SCR, mark the on-board diagnostic system OBD data obtained in a preset time period; wherein the data type includes: normal value and multiple types of abnormal value; the multiple types of abnormal value include one or any combination of the following: missing value and repeated value; After sorting the marked OBD data in chronological order, compare the data types marked by each two adjacent OBD data respectively, if the previous OBD data is marked as an abnormal value and the next OBD data is marked as a normal value, the count value corresponding to the corresponding type of abnormal value is increased by 1; After comparing all OBD data, for each type of abnormal value, divide the count value corresponding to the type of abnormal value by the total number of OBD data marked as the type of abnormal value to obtain the first transition proportion information of the type of abnormal value; According to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value, determine whether the OBD device is abnormal; When it is determined that the OBD device is abnormal, the vehicle is determined as an abnormal vehicle.

2. The vehicle supervision method according to claim 1, characterized by, According to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value, determine whether the OBD device is abnormal, comprising: For each type of abnormal value, if the first transition proportion information of the abnormal value is greater than or equal to the transition proportion threshold information corresponding to the abnormal value, it is determined that the OBD device is abnormal.

3. The vehicle supervision method according to claim 1, characterized by, The abnormal value also includes an out-of-range value, and the vehicle monitoring method further comprises: After sorting the marked OBD data in chronological order, compare the data types marked by each two adjacent OBD data respectively, if the previous OBD data is marked as an abnormal value and the next OBD data is marked as a normal value, the count value corresponding to the corresponding type of abnormal value is increased by 1; if the previous OBD data is marked as a normal value and the next OBD data is marked as an out-of-range value, the count value of the normal value turning into an out-of-range value is increased by 1; After comparing all OBD data, divide the count value of the out-of-range value turning into a normal value by the total number of OBD data marked as an out-of-range value to obtain the second transition proportion information of the out-of-range value turning into a normal value; divide the count value of the normal value turning into an out-of-range value by the total number of OBD data marked as a normal value to obtain the third transition proportion information of the normal value turning into an out-of-range value; According to the second transition proportion information and the third transition proportion information, determine whether the OBD device is abnormal.

4. The vehicle supervision method according to claim 3, characterized by, According to the second transition proportion information and the third transition proportion information, determine whether the OBD device is abnormal, comprising: If the second transition proportion information is greater than or equal to the pre-set transition proportion threshold information corresponding to the transition of the out-of-range value to the normal value, it is determined that the OBD data is not the data recorded by the OBD device in the cold start stage, and the OBD device is determined to be abnormal when the OBD data is not the data recorded by the OBD device in the cold start stage. When the third transition proportion information is greater than or equal to a transition proportion threshold information corresponding to a transition from the normal value to the out-of-range value, it is determined that the OBD device is abnormal.

5. The vehicle supervision method according to claim 4, characterized by, The determination of whether the OBD data is recorded by the OBD device in the cold start phase comprises: For each of the OBD data, a data segment satisfying a transition rule of repeated value-out-of-range value-normal value is screened out, wherein the number of OBD data marked as repeated value before the OBD data marked as out-of-range value in the data segment is greater than or equal to a first preset number, and the number of OBD data marked as normal value after the OBD data marked as out-of-range value is greater than or equal to a second preset number; The NOx outlet temperature corresponding to all out-of-range values in all the data segments is counted to obtain a temperature threshold range at which the NOx sensor downstream of the vehicle SCR starts to work normally; and the time used from the start of the trip to the first normal value in the cold start data segment is obtained to obtain a consumption time range of the cold start of the vehicle. The OBD data meeting the determined temperature threshold range and consumption time range is determined as the OBD data recorded by the OBD device in the cold start phase.

6. The vehicle monitoring method according to claim 4, characterized by, Before determining whether the OBD device is abnormal according to the first transition proportion information of each abnormal value and the transition proportion threshold information corresponding to each abnormal value, the vehicle supervision method further comprises: The reference transition proportion information of mutual transition between different data types of the pre-determined normal data is counted, wherein the reference transition proportion information of the following transition paths of the normal data is all greater than or equal to a pre-set reference transition proportion threshold: missing value-repeated value-out-of-range value-normal value-missing value; Based on the determined reference transition proportion information, the transition proportion threshold information of the OBD device abnormality is set; The reference transition proportion information comprises: missing value transition to normal value transition proportion information, repeated value transition to normal value transition proportion information, out-of-range value transition to normal value transition proportion information, and normal value transition to out-of-range value transition proportion information.

7. The vehicle supervision method according to claim 6, wherein: When the abnormal value comprises a missing value, the reference transition proportion information A_0 is the missing value transition to normal value transition proportion information of the normal data, and the transition proportion threshold information corresponding to the missing value is A_1, A_1=A_0+(20±5)%; When the abnormal value comprises a repeated value, the reference transition proportion information B_0 is the repeated value transition to normal value transition proportion information of the normal data, and the transition proportion threshold information corresponding to the repeated value is B_1, B_1=B_0+(10±5)%; When the abnormal value comprises an out-of-range value, the reference transition proportion information C_0 is the out-of-range value transition to normal value transition proportion information of the normal data, and the transition proportion threshold information corresponding to the transition from the out-of-range value to the normal value is C_1, C_1=C_0-(15±5)%. The abnormal value includes an out-of-bound value, the reference transition proportion information D_0 is the transition proportion information of the normal value of the normal data to the out-of-bound value, the transition proportion threshold information corresponding to the normal value to the out-of-bound value is D_1, D_1=D_0+(10±5)%.

8. The vehicle supervision method according to any one of claims 1 to 7, characterized by, The vehicle supervision method further comprises: The collection time interval of adjacent two pieces of OBD data, and OBD data less than or equal to a pre-set continuous determination threshold are divided into a trip; When the number of OBD data contained in the divided trip is less than or equal to a pre-set number threshold, or the total duration of the trip is less than or equal to a pre-set duration threshold, the trip is determined as a short trip; The OBD data determined as the short trip is deleted. 9.A computer storage medium, the computer storage medium storing a computer program, the computer program being executed by a processor to implement the vehicle supervision method based on OBD data according to any one of claims 1 to 8.

10. A vehicle monitoring device based on OBD data, characterized by, Comprise: A marking unit, a comparison counting unit, a statistical proportion unit, a judgment unit and a determination unit; wherein, The marking unit is configured to mark the OBD data acquired in a preset time period according to the data type of the SCR downstream NOx sensor output data pre-divided; wherein, the data type includes normal values and a plurality of types of abnormal values; the plurality of types of abnormal values include one or any combination of the following: missing values and repeated values; The comparison counting unit is configured to sort the marked OBD data in chronological order, and then compare the data types marked by each two adjacent OBD data respectively, if the previous OBD data is marked as an abnormal value and the next OBD data is marked as a normal value, the count value corresponding to the corresponding type of abnormal value is increased by 1; The statistical proportion unit is configured to, after comparing all OBD data, for each type of abnormal value, divide the count value corresponding to the type of abnormal value by the total number of OBD data marked as the type of abnormal value to obtain the first transition proportion information of the type of abnormal value; The judgment unit is configured to determine whether the OBD device is abnormal according to the first transition proportion information of each type of abnormal value and the pre-set transition proportion threshold information corresponding to each type of abnormal value; The determination unit is configured to determine the vehicle as an abnormal vehicle when the OBD device is determined to be abnormal.

Images