User working condition construction method and device based on low-frequency RTM data

Abstract

The invention provides a user working condition construction method and device based on low-frequency RTM data, and the method comprises the steps: S100, collecting high-frequency data, and carrying out the slicing of effective data segments of the high-frequency data; s200, clustering the vehicle speed of the low-frequency RTM and the driving mileage within the sampling time, determining the center point of each cluster, and calculating the proportion of each cluster data; s300, comparing and screening the slices obtained in the step S100 according to the central point and the specific gravity determined in the step S200; and step S400, splicing the slice data screened in the step S300 to form the user working condition. According to the method, RTM low-frequency vehicle speed data is utilized, a small amount of high-frequency user data is combined, input of a large amount of high-frequency user data is not needed, and the user working condition development cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle operating condition analysis and construction, and in particular relates to a user operating condition construction method and device based on low-frequency RTM data. Background Art

[0002] When new energy vehicles are driving, the vehicle's driving direction, accumulated mileage, real-time speed and other related parameters will be collected by the RTM system, and the data will be transmitted to the relevant platform at a specific frequency to form a relevant RTM database. The RTM database is built based on the user's actual vehicle usage scenario, so it is of great significance to the construction of user working conditions. However, the national standard GB / T32960 requires that the vehicle is not less than 30s under normal conditions, and the user working condition used for bench testing needs to be accurate to within 1s. Therefore, a method is needed to utilize low-frequency RTM data to form a standard user working condition.

[0003] Invention patent content

[0004] In view of the above problems, the present invention provides a method and device for constructing user working conditions based on low-frequency RTM data, which clusters the working conditions of low-frequency RTM data and combines high-frequency user working condition slicing to realize the construction of user working conditions. The method includes a processing method for high-frequency user data and a processing method for low-frequency RTM data, and can obtain relevant user working conditions in a specified area by filtering out the orientation information in the RTM data. The purpose of the present invention is achieved through the following technical solutions:

[0005] The present invention first provides a method for constructing a user operating condition based on low-frequency RTM data, which comprises:

[0006] Step S100: collecting high-frequency data and slicing its valid data segments;

[0007] Step S200: clustering the vehicle speed of the low-frequency RTM and the mileage within the sampling time, determining the center point of each cluster, and calculating the proportion of each cluster data;

[0008] Step S300: comparing and screening the slices obtained in step S100 according to the center point and specific gravity determined in step S200;

[0009] Step S400: splicing the slice data filtered in step S300 to form a user working condition.

[0010] Further, the step S100 includes:

[0011] Step S110 - high frequency data collection: collect high frequency user data through the data collection system;

[0012] Step S120 - high-frequency data screening: using the idle point as a marker, the data between two idle points obtained in step S110 is used as a valid high-frequency data segment;

[0013] Step S130 - high-frequency data cleaning: delete outliers in valid high-frequency data segments and fill missing values ​​using interpolation;

[0014] Step S140 - high-frequency data slicing: slicing the cleaned high-frequency data segments based on the RTM data sampling period;

[0015] Step S150 - extracting features of high-frequency data slices: extracting features of the slices using the starting speed and the mileage within the sampling time as data features.

[0016] Furthermore, for the source of the high-frequency data collected in step S100, a specific implementation method is as follows: a data collection system is installed on the designated relevant vehicle to collect relevant parameter information such as driving direction, accumulated mileage, real-time vehicle speed, etc., and the collection frequency is not less than 1s; then the vehicle is handed over to the designated relevant user, and the user drives the vehicle for a specified time to obtain basic high-frequency user data for user working condition construction.

[0017] Further, the step S200 includes:

[0018] Step S210 - low frequency data collection: read the RTM database to collect low frequency user data;

[0019] Step S220 - low frequency data screening: using the idle point as a marker, the low frequency user data obtained in step S210 is processed so that the data between two idle points is taken as a valid low frequency data segment;

[0020] Step S230 - low-frequency data cleaning: delete outliers in valid low-frequency data segments and fill missing values ​​using interpolation;

[0021] Step S240 - low-frequency data clustering analysis: perform cluster analysis on the cleaned low-frequency data segments according to vehicle speed and mileage within the adopted time, so that they are clustered into four groups: low speed, medium speed, high speed, and ultra-high speed, and determine the relevant center points A, B, C, and D, and calculate the proportion of each type of data as a%, b%, c%, and d% respectively.

[0022] Further, the step S300 includes:

[0023] Step S310 - distance calculation: taking the relevant center points A, B, C, and D of the low-frequency data obtained in step S240 as a reference, calculating the distance between the slice features of the high-frequency data slice and the relevant center points;

[0024] Step S320 - ratio screening: screening out low-speed high-frequency segments, medium-speed high-frequency segments, high-speed high-frequency segments and ultra-high-speed high-frequency segments according to the ratios of a%, b%, c%, and d%.

[0025] Furthermore, the user working condition construction needs to include low-frequency RTM and high-frequency user data fields. Through time, longitude, latitude and other information, the vehicle data in the specified time and specified area can be filtered out. Through vehicle status and gear information, the relevant information of vehicle idling point can be filtered out. The filtered vehicle speed and accumulated mileage are clustered as two dimensions.

[0026] Furthermore, the clustering is implemented using the K-means method.

[0027] Furthermore, in step 300, the Euclidean distance between the data features of the high-frequency user data slices and the low-frequency RTM data cluster center points can be calculated for screening, and the Euclidean distance less than the preset threshold is the screening result.

[0028] The present invention further provides a device for constructing working conditions for implementing the above method, characterized in that it includes: a high-frequency data acquisition module, a low-frequency data acquisition module, a data processing module, a clustering module, a calculation module and a working condition construction module. The high-frequency data acquisition module is used to collect high-frequency user data; the low-frequency data acquisition module is used to collect low-frequency user data from RTM; the data processing module is used to screen, clean, slice and extract features of the collected high-frequency user data and low-frequency user data; the clustering module is used to perform cluster analysis on high-frequency data and low-frequency data; the calculation module is used to calculate the proportion of each type of data and the distance between data according to specific indicators.

[0029] The beneficial effects of the present invention are:

[0030] (1) The present invention can filter out relevant regions through the latitude and longitude information of RTM data, and construct user conditions in each region by combining high-frequency user data from each region;

[0031] (3) This solution uses RTM's low-frequency vehicle speed data combined with a small amount of high-frequency user data, eliminating the need for a large amount of high-frequency user data input and reducing user condition development costs;

[0032] (3) The user operating conditions obtained by the method of the present invention are closer to the actual user's test bench operating conditions and can better support the development of related parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The figure is a schematic diagram of the user working condition construction process of embodiment 1 of the present invention;

[0034] Figure 2The figure shows a schematic diagram of the working condition fitting process in the user working condition construction in the embodiment 1 of the present invention;

[0035] Figure 3 It is a schematic diagram of a user working condition construction device according to Embodiment 2 of the present invention; DETAILED DESCRIPTION

[0037] The technical scheme of the preferred embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Embodiment 1:

[0039] This embodiment provides a method for constructing a user operating condition based on low-frequency RTM data, which includes:

[0040] Step S100: Collect high-frequency data and slice its valid data segments. The specific process is as follows:

[0041] Step S110 - high frequency data collection: collect high frequency user data through the data collection system;

[0042] Step S120 - high-frequency data screening: using the idle point as a marker, the data between two idle points obtained in step S110 is used as a valid high-frequency data segment;

[0043] Step S130 - high-frequency data cleaning: delete outliers in valid high-frequency data segments and fill missing values ​​using interpolation;

[0044] Step S140 - high-frequency data slicing: slicing the cleaned high-frequency data segments based on the RTM data sampling period;

[0045] Step S150 - extracting features of high-frequency data slices: extracting features of the slices using the starting speed and the mileage within the sampling time as data features.

[0046] Step S200: clustering the vehicle speed of the low-frequency RTM and the mileage within the sampling time, determining the center point of each cluster, and calculating the proportion of each cluster data; the specific process is as follows:

[0047] Step S210 - low frequency data collection: read the RTM database to collect low frequency user data;

[0048] Step S220 - low frequency data screening: using the idle point as a marker, the low frequency user data obtained in step S210 is processed so that the data between two idle points is taken as a valid low frequency data segment;

[0049] Step S230 - low-frequency data cleaning: delete outliers in valid low-frequency data segments and fill missing values ​​using interpolation;

[0050] Step S240 - low-frequency data clustering analysis: perform cluster analysis on the cleaned low-frequency data segments according to vehicle speed and mileage within the adopted time, so that they are clustered into four groups: low speed, medium speed, high speed, and ultra-high speed, and determine the relevant center points A, B, C, and D, and calculate the proportion of each type of data as a%, b%, c%, and d% respectively.

[0051] Step S300: Compare and screen the slices obtained in step S100 according to the center point and specific gravity determined in step S200. The specific process is as follows:

[0052] Step S310-distance calculation: using the relevant center points A, B, C,

[0053] D is used as the benchmark to calculate the distance between the slice features of the high-frequency data slice and the relevant center point; step S320-proportional screening: filter out low-speed high-frequency segments, medium-speed high-frequency segments, high-speed high-frequency segments and ultra-high-speed high-frequency segments according to the ratios of a%, b%, c%, and d%.

[0054] Step S400: splicing the slice data filtered in step S300 to form a user working condition.

[0055] In a preferred embodiment, the implementation method for the source of the high-frequency data collected in step S100 is as follows: a data acquisition system is installed on the designated relevant vehicle to collect relevant parameter information such as driving direction, accumulated mileage, real-time vehicle speed, etc., and the acquisition frequency is not less than 1s; the vehicle is then handed over to the designated relevant user, and the user drives the vehicle for a specified time to obtain basic high-frequency user data for user working condition construction.

[0056] In a preferred embodiment, the low-frequency RTM and high-frequency user data fields that need to be included in the user working condition construction can filter out vehicle data in a specified area within a specified time through information such as time, longitude, and latitude, and can filter out information related to vehicle idle points through vehicle status and gear information. The filtered vehicle speed and accumulated mileage are clustered as two dimensions for clustering.

[0057] In a preferred embodiment, the clustering is implemented by K-means method. In step 300, the Euclidean distance between the data features of the high-frequency user data slices and the low-frequency RTM data cluster center points can be calculated for screening. If the Euclidean distance is less than a preset threshold, it is the screening result.

[0058] Embodiment 2:

[0059] This embodiment provides a user operating condition construction device for implementing the method of embodiment 1, such as Figure 2 As shown, the device includes: a high-frequency data acquisition module, a low-frequency data acquisition module, a data processing module, a clustering module, a calculation module and a working condition construction module. Among them:

[0060] High-frequency data collection module, used to collect high-frequency user data;

[0061] A low-frequency data collection module is used to collect low-frequency user data from RTM;

[0062] The data processing module is used to screen, clean, slice and extract features of the collected high-frequency user data and low-frequency user data;

[0063] Clustering module, used to perform clustering processing on high-frequency data and low-frequency data;

[0064] The calculation module is used to calculate the proportion of various data and the distance between data according to specific indicators;

[0065] The working condition construction module is used to fit and splice data to form user working conditions.

[0066] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention. The dimensions described in the drawings and embodiments are not related to the specific objects and are not used to limit the protection scope of the present invention. The dimensions of the objects can be selected and changed according to actual needs.

Claims

1. A method for constructing user operating conditions based on low-frequency RTM data, comprising: Step S100: collecting high-frequency data and slicing its valid data segments; Step S200: clustering the vehicle speed of the low-frequency RTM and the mileage within the sampling time, determining the center point of each cluster, and calculating the proportion of each cluster data; Step S300: comparing and screening the slices obtained in step S100 according to the center point and specific gravity determined in step S200; Step S400: splicing the slice data filtered in step S300 to form a user working condition.

2. A method for constructing user operating conditions based on low-frequency RTM data according to claim 2, characterized in that: The step S100 includes: Step S110 - high frequency data collection: collect high frequency user data through the data collection system; Step S120 - high-frequency data screening: using the idle point as a marker, the data between two idle points obtained in step S110 is used as a valid high-frequency data segment; Step S130 - high-frequency data cleaning: delete outliers in valid high-frequency data segments and fill missing values ​​using interpolation; Step S140 - high-frequency data slicing: slicing the cleaned high-frequency data segments based on the RTM data sampling period; Step S150 - extracting features of high-frequency data slices: extracting features of the slices using the starting speed and the mileage within the sampling time as data features.

3. The method for constructing user operating conditions based on low-frequency RTM data according to claim 1, characterized in that: The step S200 includes: Step S210 - low frequency data collection: read the RTM database to collect low frequency user data; Step S220 - low frequency data screening: taking the idle point as a marker, the low frequency user data obtained in step S210 is processed to take the data between two idle speed sections as a valid low frequency data segment; Step S230 - low-frequency data cleaning: delete outliers in valid low-frequency data segments and fill missing values ​​using interpolation; Step S240 - low-frequency data clustering analysis: perform cluster analysis on the cleaned low-frequency data segments according to vehicle speed and mileage within the adopted time, so that they are clustered into four groups: low speed, medium speed, high speed, and ultra-high speed, and determine the relevant center points A, B, C, and D, and calculate the proportion of each type of data as a%, b%, c%, and d% respectively.

4. The method for constructing user operating conditions based on low-frequency RTM data according to claim 3, characterized in that: The step S300 includes: Step S310 - distance calculation: taking the relevant center points A, B, C, and D of the low-frequency data obtained in step S240 as a reference, calculating the distance between the slice features of the high-frequency data slice and the relevant center points; Step S320 - ratio screening: screening out low-speed high-frequency segments, medium-speed high-frequency segments, high-speed high-frequency segments and ultra-high-speed high-frequency segments according to the ratios of a%, b%, c%, and d%.

5. A method for constructing user operating conditions based on low-frequency RTM data according to any one of claims 1 to 4, characterized in that: Step S100 and step S200 include vehicle data of a specified time and region formed based on filtering low-frequency data in terms of time, longitude, and latitude, and / or idle point information based on filtering vehicle status and gear information.

6. A method for constructing user operating conditions based on low-frequency RTM data according to any one of claims 1 to 4, characterized in that: The clustering is achieved using the K-means method.

7. The method for constructing user operating conditions based on low-frequency RTM data according to claim 6, characterized in that: In step 300 , comparison and screening are performed using Euclidean distance.

8. A user working condition construction device, used to implement the method according to any one of claims 1 to 7, characterized in that: include: High-frequency data acquisition module, low-frequency data acquisition module, data processing module, clustering module, calculation module and working condition construction module.

Images