A Method and Device for Identifying Abnormal Values of Odometer in a Connected Vehicle Instrument Panel

Abstract

A method and device for identifying abnormal values ​​of dashboard mileage of a connected vehicle, the method comprising: step 1, collecting a group of dashboard mileage data, arranging the dashboard mileage data in this group in order from small to large, and obtaining the interquartile range IQR of the dashboard mileage data in this group; step 2, comparing the size between the IQR and a preset threshold M, taking the maximum value between the two, recorded as delta; step 3, based on the obtained maximum value, calculating the reasonable upper bound upper and lower bound lower of the dashboard mileage data in this group; step 4, comparing each dashboard mileage data in this group of dashboard mileage data with the upper bound upper and lower bound lower calculated in step 4, respectively, if the dashboard mileage data is greater than the upper bound upper or less than the lower bound lower, then the dashboard mileage data is determined to be abnormal data and is removed from the dashboard mileage data.

Description

Technical Field

[0001] The present invention relates to the field of vehicle data processing, and particularly to a method and device for identifying abnormal values of instrument panel mileage of connected vehicles. Background Art

[0002] With the development of vehicle networking and intelligence, in-vehicle data collection devices (such as Tbox) can collect various vehicle signals from vehicles and upload them to the cloud for storage and analysis.

[0003] Vehicle instrument panel mileage data, as a key basic data collected by data collection devices, is widely used in vehicle big data analysis. In the fields of vehicle R & D tests, vehicle user portraits, vehicle itinerary analysis, vehicle health assessment, battery life attenuation, after-sales maintenance and marketing, etc., extracting accurate instrument panel mileage data is of great significance.

[0004] However, due to various reasons, various abnormal values will appear in the vehicle instrument panel mileage data during the collection and transmission process, such as Figure 1 and Figure 2 As shown, since the values at the abnormal points are relatively large, the values at other places are compressed together in the figure. This type of abnormal phenomenon has a serious impact on the analysis and application of vehicle big data. Summary of the Invention

[0005] In view of the technical defects and drawbacks existing in the prior art, embodiments of the present invention provide a method and device for identifying abnormal values of instrument panel mileage of connected vehicles to overcome or at least partially solve the above problems. The specific solutions are as follows:

[0006] As a first aspect of the present invention, a method for identifying abnormal values of instrument panel mileage of connected vehicles is provided. The method includes:

[0007] Step 1: Collect a set of instrument panel mileage data, arrange the set of instrument panel mileage data in ascending order, and obtain the interquartile range IQR of the set of instrument panel mileage data;

[0008] Step 2: Compare the size of IQR and a preset threshold M, and take the maximum value of the two, denoted as delta;

[0009] Step 3: Based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of the set of instrument panel mileage data;

[0010] Step 4: Compare each piece of instrument panel mileage data in the set of instrument panel mileage data with the calculated upper bound upper and lower bound lower respectively. If the instrument panel mileage data is greater than the upper bound upper or less than the lower bound lower, it is determined that this piece of instrument panel mileage data is abnormal data and is excluded from the set of instrument panel mileage data.

[0011] Further, step 1 further includes: filtering out the dashboard mileage data that exceeds the effective range based on the effective range of the dashboard mileage of the connected vehicle, so as to obtain a set of effective dashboard mileage data.

[0012] Further, in step 1, obtaining the interquartile range IQR of this set of dashboard mileage data specifically includes:

[0013] Denote the lower quartile of this set of dashboard mileage data as Q1, and denote the upper quartile of this set of dashboard mileage data as Q3, so as to obtain the interquartile range IQR of this set of dashboard mileage data = Q3 - Q1.

[0014] Further, in step 2, compare the size between 1.5 times of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta, delta = max(1.5 * IQR, M).

[0015] Further, in step 3, based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of this set of dashboard mileage data specifically as:

[0016] Upper bound upper = Q3 + delta;

[0017] Lower bound lower = Q1 - delta;

[0018] Wherein, Q1 is the lower quartile of this set of dashboard mileage data, and Q3 is the upper quartile of this set of dashboard mileage data.

[0019] As the second aspect of the present invention, there is provided a device for identifying outliers in the dashboard mileage of a connected vehicle, and the device includes: a data acquisition module, an interquartile range calculation module, a comparison module, an upper and lower bound value calculation module, and an outlier analysis module;

[0020] The data acquisition module is used to acquire a set of dashboard mileage data;

[0021] The interquartile range calculation module is used to arrange the acquired set of dashboard mileage data in ascending order and obtain the interquartile range IQR of this set of dashboard mileage data;

[0022] The comparison module is used to compare the size between IQR and the preset threshold M, and take the maximum value of the two, denoted as delta;

[0023] The upper and lower bound value calculation module is used to calculate the reasonable upper bound upper and lower bound lower of this set of dashboard mileage data based on the obtained maximum value;

[0024] The abnormal analysis module is used to compare each piece of dashboard mileage data in this group of dashboard mileage data with the calculated upper bound upper and lower bound lower respectively. If the dashboard mileage data is greater than the upper bound upper or less than the lower bound lower, it is determined that this piece of dashboard mileage data is abnormal data and is excluded from this group of dashboard mileage data.

[0025] Further, the data acquisition module is also used to: filter out the dashboard mileage data that exceeds the effective range based on the effective range of the dashboard mileage of the connected vehicle, so as to obtain a group of valid dashboard mileage data.

[0026] Further, the specific process for the interquartile range calculation module to obtain the interquartile range IQR of this group of dashboard mileage data is as follows: denote the lower quartile of this group of dashboard mileage data as Q1, and denote the upper quartile of this group of dashboard mileage data as Q3, so as to obtain the interquartile range IQR of this group of dashboard mileage data as IQR = Q3 - Q1;

[0027] Further, the comparison module is specifically used to: compare the size between 1.5 times of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta, where delta = max(1.5 * IQR, M).

[0028] Further, the abnormal analysis module is specifically used to:

[0029] Based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of this group of dashboard mileage data specifically as:

[0030] Upper bound upper = Q3 + delta;

[0031] Lower bound lower = Q1 - delta;

[0032] Wherein, Q1 is the lower quartile of this group of dashboard mileage data, and Q3 is the upper quartile of this group of dashboard mileage data.

[0033] The present invention has the following beneficial effects:

[0034] The present invention obtains the interquartile range of this group of dashboard mileage data, compares it with the preset threshold M to obtain the maximum value, and based on the obtained maximum value, obtains the upper and lower bounds of the dashboard mileage data. Then, each piece of dashboard mileage data is compared with the upper bound upper and lower bound lower calculated in step 4 respectively, so as to be able to quickly, effectively and automatically identify such data anomalies. Description of the Drawings

[0035] Figure 1 and Figure 2 show various outlier situations provided by the embodiments of the present invention;

[0036] Figure 3 Schematic flowchart of a method for identifying abnormal values of the instrument panel mileage of a connected vehicle provided by an embodiment of the present invention;

[0037] Figure 4 and Figure 5 Comparison diagram before and after effectively removing abnormal data provided by an embodiment of the present invention. Detailed implementation manners

[0038] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0039] As Figure 3 shown, as the first embodiment of the present invention, a method for identifying abnormal values of the instrument panel mileage of a connected vehicle is provided. The method includes:

[0040] Step 1: Collect a set of instrument panel mileage data. Based on the effective range of the instrument panel mileage of the connected vehicle, filter out the instrument panel mileage data that exceeds this effective range, so as to obtain a set of effective instrument panel mileage data. Then, arrange the mileage data of this group in ascending order, and obtain the interquartile range IQR of the mileage data of this group.

[0041] Step 2: Compare the size of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta.

[0042] Step 3: Based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of the mileage data of this group of instrument panels.

[0043] Step 4: Compare each piece of instrument panel mileage data in this group of instrument panel mileage data with the calculated upper bound upper and lower bound lower respectively. If the instrument panel mileage data is greater than the upper bound upper or less than the lower bound lower, it is determined that this piece of instrument panel mileage data is abnormal data and is removed from this group of instrument panel mileage data.

[0044] The present invention obtains the interquartile range of the mileage data of this group of instrument panels, compares it with the preset threshold M to obtain the maximum value, and based on the obtained maximum value, obtains the upper and lower bounds of the mileage data of the instrument panel. Each piece of instrument panel mileage data is compared with the upper bound upper and lower bound lower calculated in Step 4 respectively, so as to be able to quickly, effectively and automatically identify such data anomalies, such as Figure 4 and 5As shown in the figure, it is a comparison schematic diagram before and after effectively eliminating abnormal data using the present invention. It should be noted that due to the relatively large value at the abnormal point, the Y-axis is relatively long, so the values at other places in the figure are compressed together. After deleting the abnormal point, it can be seen that the dashboard mileage data values at other places are different.

[0045] Among them, the lower quartile of this group of dashboard mileage data is denoted as Q1, and the upper quartile of this group of dashboard mileage data is denoted as Q3, so as to obtain the interquartile range IQR = Q3 - Q1 of this group of dashboard mileage data.

[0046] Among them, in step 2, compare the size of 1.5 times of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta, delta = max(1.5 * IQR, M). In this embodiment, the preset threshold M is 800.

[0047] Among them, in step 3, based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of this group of dashboard mileage data specifically as:

[0048] Upper bound upper = Q3 + delta;

[0049] Lower bound lower = Q1 - delta;

[0050] Among them, Q1 is the lower quartile of this group of dashboard mileage data, and Q3 is the upper quartile of this group of dashboard mileage data.

[0051] As a second embodiment of the present invention, a device for identifying abnormal values of a connected vehicle dashboard mileage is characterized in that the device includes: a data acquisition module, an interquartile range calculation module, a comparison module, an upper and lower bound value calculation module, and an abnormal analysis module;

[0052] The data acquisition module is used to acquire a group of dashboard mileage data;

[0053] The interquartile range calculation module is used to arrange the acquired group of dashboard mileage data in ascending order and obtain the interquartile range IQR of this group of dashboard mileage data;

[0054] The comparison module is used to compare the size of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta;

[0055] The upper and lower bound value calculation module is used to calculate the reasonable upper bound upper and lower bound lower of this group of dashboard mileage data based on the obtained maximum value;

[0056] The abnormal analysis module is used to compare each piece of dashboard mileage data in this group of dashboard mileage data with the calculated upper bound "upper" and lower bound "lower" respectively. If the dashboard mileage data is greater than the upper bound "upper" or less than the lower bound "lower", it is determined that this piece of dashboard mileage data is abnormal data and is excluded from this group of dashboard mileage data.

[0057] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for identifying abnormal values of the odometer of a connected vehicle, characterized in that, The method includes: Step 1: Collect a set of instrument panel mileage data, sort this set of instrument panel mileage data in ascending order, and obtain the interquartile range (IQR) of this set of instrument panel mileage data. Step 2: Compare the size of the IQR with a preset threshold M, and take the maximum value of the two, denoted as delta. Step 3: Based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of this set of instrument panel mileage data. Step 4: Compare each piece of instrument panel mileage data in this set of instrument panel mileage data with the calculated upper bound upper and lower bound lower respectively. If the instrument panel mileage data is greater than the upper bound upper or less than the lower bound lower, then determine that this piece of instrument panel mileage data is abnormal data and remove it from this set of instrument panel mileage data. Among them, in Step 3, based on the obtained maximum value, calculating the reasonable upper bound upper and lower bound lower of this set of instrument panel mileage data is specifically as follows: Upper bound upper = Q3 + delta; Lower bound lower = Q1 - delta; Among them, Q1 is the lower quartile of this set of instrument panel mileage data, and Q3 is the upper quartile of this set of instrument panel mileage data.

2. The method for identifying abnormal mileage values of a connected vehicle dashboard according to claim 1, wherein Step 1 further includes: Based on the effective range of the instrument panel mileage of the connected vehicle, filter out the instrument panel mileage data that exceeds this effective range, so as to obtain a set of valid instrument panel mileage data.

3. The method for identifying abnormal values of the odometer of a connected vehicle according to claim 1, characterized in that, In Step 1, obtaining the interquartile range (IQR) of this set of instrument panel mileage data specifically includes: Take the lower quartile of this set of instrument panel mileage data and denote it as Q1, take the upper quartile of this set of instrument panel mileage data and denote it as Q3, so as to obtain the interquartile range (IQR) of this set of instrument panel mileage data = Q3 - Q1.

4. The method for identifying abnormal mileage values of a connected vehicle dashboard according to claim 1, characterized in that, In Step 2, compare the size of 1.5 times the IQR with the preset threshold M, take the maximum value of the two, denoted as delta, and delta = max(1.5 * IQR, M).

5. An abnormal value recognition device for the odometer of a connected vehicle, characterized in that, The device includes: a data acquisition module, an interquartile range calculation module, a comparison module, an upper and lower bound value calculation module, and an abnormal analysis module; The data acquisition module is used to collect a set of instrument panel mileage data; The interquartile range calculation module is used to sort the collected set of instrument panel mileage data in ascending order and obtain the interquartile range (IQR) of this set of instrument panel mileage data; The comparison module is used to compare the size of the IQR with the preset threshold M, and take the maximum value of the two, denoted as delta; The upper and lower bound value calculation module is used to calculate the reasonable upper bound upper and lower bound lower of this set of instrument panel mileage data based on the obtained maximum value; The abnormal analysis module is used to compare each piece of instrument panel mileage data in this set of instrument panel mileage data with the calculated upper bound upper and lower bound lower respectively. If the instrument panel mileage data is greater than the upper bound upper or less than the lower bound lower, then determine that this piece of instrument panel mileage data is abnormal data and remove it from this set of instrument panel mileage data; Among them, the abnormal analysis module is specifically used for: Based on the obtained maximum value, calculate the reasonable upper bound upper and lower bound lower of the dashboard mileage data of this group specifically as follows: Upper bound upper = Q3 + delta; Lower bound lower = Q1 - delta; Wherein, Q1 is the lower quartile of the dashboard mileage data of this group, and Q3 is the upper quartile of the dashboard mileage data of this group.

6. The device for identifying abnormal values of the odometer of a connected vehicle according to claim 5, characterized in that, The data acquisition module is further configured to: filter out the dashboard mileage data that exceeds the effective range based on the effective range of the dashboard mileage of the connected vehicle, so as to obtain a set of valid dashboard mileage data.

7. The device for identifying abnormal mileage values of the dashboard of a connected vehicle according to claim 5, characterized in that, The interquartile range calculation module obtains the interquartile range IQR of the dashboard mileage data of this group, specifically including: denoting the lower quartile of the dashboard mileage data of this group as Q1, and denoting the upper quartile of the dashboard mileage data of this group as Q3, so as to obtain the interquartile range IQR of the dashboard mileage data of this group = Q3 - Q1.

8. The networked vehicle instrument panel mileage outlier recognition device according to claim 5, wherein, The comparison module is specifically configured to: compare the size between 1.5 times of IQR and the preset threshold M, and take the maximum value of the two, denoted as delta, delta = max(1.5 * IQR, M).

Images