Commercial vehicle gear identification method based on working condition driving data

Abstract

The invention discloses a commercial vehicle gear identification method based on working condition driving data, and the method comprises the steps: a vehicle-mounted intelligent terminal collects the working condition data of vehicle driving, and transmits the working condition data to a big data management platform; finding out each cluster center calculated by all clustering algorithms by using the clustering algorithms of various modes; the ratio of the center point coordinates of the multiple cluster centers is regarded as a typical value capable of reflecting the transmission ratio level, and n-1 comparison values can be obtained through sequential ratio calculation; deducing a mode as a transmission ratio ratio according to a geometric progression rule that the transmission ratio of the gearbox is designed; reading the rotating speed / vehicle speed ratio during high-speed driving to deduce the rotating speed / vehicle speed ratio of the state; the rotating speed / vehicle speed ratio of the highest gear sequentially recurses the rotating speed / vehicle speed ratios of other gears; and the gear value under the working condition point can be obtained by combining the rotating speed / vehicle speed ratio, obtained through recursion, of each gear. Therefore, the problem of one-sided calculation result of a single algorithm can be avoided, and the gear identification precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of automobile and engine design and manufacturing, and in particular to a commercial vehicle gear position identification method based on operating condition driving data. Background Art

[0002] As an important part of intelligent connected vehicle services, the development of commercial vehicle gear recognition technology based on operating condition driving data not only greatly affects the convenience and accuracy of vehicle intelligent connected services, but also plays a vital role in research work such as driving behavior mining and user portrait description.

[0003] The traditional method of directly reading the current gear position of the vehicle using the on-board diagnostic port relies heavily on the intelligent network terminal monitoring variable acquisition protocol. When the protocol does not support this data, the gear position data label will be missing, and when the actual configuration of the commercial vehicle's gearbox is inconsistent with the preset model, it will cause the gear position reading error problem. The existing gear position identification solutions mainly include external device auxiliary identification, data classification algorithm identification, etc.

[0004] In the existing technical solutions, there are the following disadvantages:

[0005] 1. The auxiliary identification solution for external devices increases the user's usage cost, and the reliability of the device needs to be verified, which is not conducive to commercial promotion;

[0006] 2. The data classification algorithm recognition scheme lacks auxiliary markers of gearbox design rules, consumes a lot of computing resources, and has recognition accuracy errors, making it difficult to put into practical application.

[0007] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0008] The purpose of the present invention is to provide a commercial vehicle gear recognition method based on operating condition driving data, which adopts a multi-modal machine learning clustering algorithm to infer the gear speed / vehicle speed ratio of each gear in the operating condition data point, thereby avoiding the one-sided problem of calculation results of a single algorithm and improving the gear recognition accuracy.

[0009] To achieve the above-mentioned purpose, the present invention provides a commercial vehicle gear position identification method based on operating condition driving data, comprising: an on-board intelligent terminal collects the operating condition data of the vehicle, and sends the operating condition data to a big data management platform, the big data management platform reads the historical driving data file of the vehicle, and compares it with the operating condition data; clustering the rotation speed / vehicle speed data of the operating condition data by using a multi-modal clustering algorithm, and finding the centers of all clusters calculated by the clustering algorithm; the ratio of the coordinates of the center points of the multiple cluster centers is regarded as a typical value that can reflect the transmission ratio level, and the multiple typical values ​​will generate multiple Y / X coordinate values, and the multiple Y / X coordinate values ​​are arranged from small to large, and then the ratios are calculated in order. The calculation can obtain n-1 comparison values, where the rotational speed is the Y coordinate, the vehicle speed is the X coordinate, and n is the coordinate of multiple center points; according to the rule that the gearbox transmission ratio is designed as a geometric series, the mode of the n-1 comparison values ​​of Y / X at the center of each cluster is found, and the mode is inferred to be the transmission ratio ratio; the rotational speed / vehicle speed ratio when the vehicle is traveling at high speed is read, and the rotational speed / vehicle speed ratio in this state is inferred to be the rotational speed / vehicle speed ratio of the highest gear; according to the total number of gears and the transmission ratio ratio, the rotational speed / vehicle speed ratio of the highest gear is recursively deduced in turn for the rotational speed / vehicle speed ratio of the remaining gears; and the nearest neighbor matching of the operating condition data of any point is performed in combination with the rotational speed / vehicle speed ratio of each gear obtained by recursion, so that the gear value under the operating condition point can be obtained.

[0010] In a preferred embodiment, the total number of gears is the total number of gears of the current vehicle entered by the user.

[0011] In a preferred embodiment, the vehicle speed during high-speed driving is between 110 and 120 km / h.

[0012] In a preferred embodiment, the multi-modal clustering algorithms include K-Means, mean shift clustering, DBSCAN, GMM EM clustering, agglomerative hierarchical clustering or graph group detection.

[0013] In a preferred embodiment, DBSCAN includes: loading data; the user enters the known total number of gears; only retains the rows with the value of the Epm_nEng (rpm) column between 1200-1500; converts the nested list into a two-dimensional NumPy array; calculates the mean of each column; calculates the standard deviation of each column; standardizes each column of data; uses DBSCAN for clustering; obtains cluster labels and core sample indexes; outputs core point data; calculates possible transmission ratios; outputs the mode as the inferred transmission ratio; extracts rows with vehicle speeds between certain values ​​and calculates the minimum transmission ratio; converts the nested list into a two-dimensional NumPy array; processes the array, divides the values ​​of the first column by the values ​​of the second column, and retains 2 decimal places to calculate the mode; and draws the clustering results.

[0014] In a preferred embodiment, the core point data includes: ensuring that the array has at least two columns; extracting the first column and the second column data; calculating the result of dividing the first column by the second column; sorting the results; creating an empty list to store the division results; traversing the sorted array, starting from the second element, dividing the current element by the previous element, and adding the result to the list; and filtering out the results in the list whose values ​​are between 1.1 and 1.6.

[0015] In a preferred embodiment, calculating possible transmission ratio values ​​includes: retaining two decimal places for each result; and finding the mode using the mode function in the scipy library.

[0016] In a preferred embodiment, agglomerative hierarchical clustering includes: loading data; extracting data; retaining only rows whose Epim_nEng (rpm) column values ​​are between 1200-1500; converting nested lists into two-dimensional NumPy arrays; generating simulated data; calculating silhouette coefficients for different numbers of clusters; selecting the number of clusters with the highest silhouette coefficient; clustering using the optimal number of clusters; calculating the centroid of each cluster; drawing the clustering results; and drawing the cluster centers.

[0017] Compared with the prior art, the commercial vehicle gear recognition method based on operating condition driving data of the present invention has the following beneficial effects: the present scheme adopts a multi-modal machine learning clustering algorithm to infer the speed / vehicle speed ratio of each gear in the operating condition data points respectively, and a plurality of clustering algorithms jointly cluster the operating condition point clusters, thereby avoiding the one-sided problem of the calculation results of a single algorithm and improving the gear recognition accuracy; the commercial vehicle gear recognition and judgment method introduces the step of inferring and reconstructing the calculation results based on the total number of vehicle gears input by the user and the design rules of the gearbox. Only one round of clustering calculation is required to derive the speed / vehicle speed ratio of all gears, without the need for multiple rounds of calculation, which greatly saves the consumption of computing resources and improves the recognition accuracy of the scheme at the same time. There is no need to install external equipment and there is no additional increase in the user's usage cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic flow chart of a commercial vehicle gear position identification method according to an embodiment of the present invention;

[0019] Figure 2 is a screenshot of sample data of a csv file of a dbscan clustering algorithm according to an embodiment of the present invention;

[0020] Figure 3 is a comparison chart of calculation data and official parameters of a commercial vehicle gear position identification method according to an embodiment of the present invention;

[0021] Figure 4It is a schematic diagram comparing the results of the dbscan clustering algorithm and the agglomerative hierarchical clustering algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings, but it should be understood that the protection scope of the present invention is not limited by the specific implementation modes.

[0023] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.

[0024] like Figure 1 As shown, according to a preferred embodiment of the present invention, a commercial vehicle gear position identification method based on operating condition driving data includes: an on-board intelligent terminal collects the operating condition data of the vehicle, and sends the operating condition data to a big data management platform, the big data management platform reads the historical driving data file of the vehicle, and compares it with the operating condition data; a multi-modal clustering algorithm is used to perform point clustering on the rotation speed / vehicle speed data of the operating condition data, and the centers of all clusters calculated by the clustering algorithms are found; the ratio of the coordinates of the center points of the multiple cluster centers is regarded as a typical value that can reflect the transmission ratio level, and the multiple typical values ​​will generate multiple Y / X coordinate values, and the multiple Y / X coordinate values ​​are arranged from small to large, and then the ratios are calculated in order. The calculation can obtain n-1 comparison values, where the rotational speed is the Y coordinate, the vehicle speed is the X coordinate, and n is the coordinate of multiple center points; according to the rule that the gearbox transmission ratio is designed as a geometric series, the mode of the n-1 comparison values ​​of Y / X at the center of each cluster is found, and the mode is inferred to be the transmission ratio ratio; the rotational speed / vehicle speed ratio when the vehicle is traveling at high speed is read, and the rotational speed / vehicle speed ratio in this state is inferred to be the rotational speed / vehicle speed ratio of the highest gear; according to the total number of gears and the transmission ratio ratio, the rotational speed / vehicle speed ratio of the highest gear is recursively deduced in turn for the rotational speed / vehicle speed ratio of the remaining gears; and the nearest neighbor matching of the operating condition data of any point is performed in combination with the rotational speed / vehicle speed ratio of each gear obtained by recursion, so that the gear value under the operating condition point can be obtained.

[0025] In some implementations, the total number of gears is the total number of gears of the current vehicle entered by the user.

[0026] In some embodiments, the vehicle speed during high-speed driving is between 110 and 120 km / h.

[0027] In some embodiments, the multi-modal clustering algorithms mainly include K-Means, mean shift clustering, DBSCAN clustering, GMM EM clustering, agglomerative hierarchical clustering or graph group detection clustering, etc.

[0028] In some embodiments, DBSCAN includes: loading data; the user enters a known total number of gears; only retains rows with values ​​in the Epm_nEng (rpm) column between 1200-1500; converts the nested list into a two-dimensional NumPy array; calculates the mean of each column; calculates the standard deviation of each column; standardizes each column of data; uses DBSCAN for clustering; obtains cluster labels and core sample indexes; outputs core point data; calculates possible transmission ratios; outputs the mode as an inferred transmission ratio; extracts rows with vehicle speeds between certain values ​​and calculates the minimum transmission ratio; converts the nested list into a two-dimensional NumPy array; processes the array, divides the values ​​in the first column by the values ​​in the second column, and retains 2 decimal places to calculate the mode; and plots the clustering results.

[0029] In some embodiments, outputting the core point data also includes ensuring that the array has at least two columns, extracting data from the first column and the second column, checking whether the second column has zero to avoid division by zero errors, calculating the result of dividing the first column by the second column, sorting the results, creating an empty list to store the division results, traversing the sorted array, starting from the second element, dividing the current element by the previous element, adding the result to the list, and filtering out the results in the list with values ​​between 1.1 and 1.6.

[0030] In some implementations, calculating possible transmission ratio values ​​further includes calculating steps such as retaining two decimal places for each result and finding the mode using the mode function in the scipy library.

[0031] like Figure 2 As shown, it is a csv file based on the dbscan clustering algorithm. It is a screenshot of the sample data input by the script; the total number of gears input in the above embodiment is 12, and the order ratio of each gear ratio is inferred to be 1.29, the speed / speed ratio of the highest gear is 24.67, and the subsequent gears are as follows Figure 2 The string of numbers below the screenshot is the recursive transmission ratio of each gear.

[0032] In some embodiments, agglomerative hierarchical clustering includes: loading data; extracting data; retaining only rows whose Epm_nEng (rpm) column values ​​are between 1200-1500; converting nested lists into two-dimensional NumPy arrays; generating simulated data; calculating silhouette coefficients for different numbers of clusters; selecting the number of clusters with the highest silhouette coefficient; clustering using the optimal number of clusters; calculating the centroid of each cluster; drawing the clustering results; and drawing the cluster centers.

[0033] like Figure 3As shown, it is a script based on the hierarchical clustering algorithm. The principle of recursive gear is similar, but the clustering algorithm is different. According to the sample data of the csv file run by the algorithm, the actual inspection results are used, and then according to the parameters provided by the official, the ratio of the transmission ratio of each gear is very close to the actual inferred result of 1.29.

[0034] like Figure 4 See also Figure 2 In some implementations, the above two clustering algorithms differ in the mechanism of searching for clusters. The difference in effect is that the clustering effects of scattered points with different distribution shapes are different. The above two clustering algorithms are relatively suitable for the distribution of strip-shaped scattered points. If the operating condition data is drawn, the distribution of the gear operating condition points is as follows: Figure 2 Like that, one by one. Figure 4 The two comparison graphs show that different clustering algorithms have different applicability when facing different scattered point distributions.

[0035] In some embodiments, the commercial vehicle gear recognition method based on operating condition driving data of the present invention cannot be completely implemented by relying on a certain clustering algorithm. More precisely, it should be a gear calculation mode that combines "design rule experience + clustering algorithm". The clustering algorithm includes but is not limited to the aforementioned clustering algorithms.

[0036] In summary, the commercial vehicle gear recognition method based on operating condition driving data of the present invention has the following advantages: the present scheme adopts a multi-modal machine learning clustering algorithm to infer the speed / vehicle speed ratio of each gear in the operating condition data points respectively, and a plurality of clustering algorithms jointly cluster the operating condition point clusters, thereby avoiding the one-sided problem of the calculation results of a single algorithm and improving the gear recognition accuracy; the commercial vehicle gear recognition and judgment method introduces the step of inferring and reconstructing the calculation results based on the total number of vehicle gears input by the user and the design rules of the gearbox. Only one round of clustering calculation is required to derive the speed / vehicle speed ratio of all gears, without the need for multiple rounds of calculation, which greatly saves the consumption of computing resources and improves the recognition accuracy of the scheme at the same time. There is no need to install external equipment and there is no additional increase in the user's usage cost.

[0037] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.

Claims

1. A commercial vehicle gear position identification method based on driving data, characterized in that: include: The vehicle-mounted intelligent terminal collects the operating condition data of the vehicle and sends the operating condition data to the big data management platform. The big data management platform reads the historical driving data file of the vehicle and compares it with the operating condition data; Using a multi-modal clustering algorithm to perform point clustering on the rotation speed / vehicle speed data of the operating condition data, and finding out the centers of all clusters calculated by the clustering algorithms; The ratio of the center point coordinates of the plurality of cluster centers is regarded as a typical value that can reflect the transmission ratio level. The plurality of typical values ​​will generate a plurality of Y / X coordinate values. The plurality of Y / X coordinate values ​​are arranged from small to large, and then the ratios are calculated in sequence to obtain n-1 comparison values, where the rotation speed is the Y coordinate, the vehicle speed is the X coordinate, and n is the coordinates of the plurality of center points; According to the rule that the gearbox transmission ratio is designed as a geometric series, find the mode of the n-1 comparison values ​​of Y / X at the center of each cluster, and infer that the mode is the transmission ratio ratio; Read the speed / vehicle speed ratio when the vehicle is traveling at a high speed, and infer that the speed / vehicle speed ratio in this state is the speed / vehicle speed ratio of the highest gear; According to the total number of gears and the transmission ratio, the speed / vehicle speed ratios of the remaining gears are recursively deduced from the speed / vehicle speed ratio of the highest gear; as well as By combining the recursively derived speed / vehicle speed ratios of each gear, a nearest neighbor match is performed on the operating condition data at any point to obtain the gear value at that operating condition point.

2. The commercial vehicle gear position identification method based on operating condition driving data according to claim 1, characterized in that: The total number of gears is the total number of gears of the current vehicle entered by the user.

3. The commercial vehicle gear position identification method based on operating condition driving data according to claim 1, characterized in that: The vehicle speed during the high-speed driving is between 110 and 120 km / h.

4. The commercial vehicle gear position identification method based on operating condition driving data according to claim 1, characterized in that: The multi-modal clustering algorithms include K-Means, mean shift clustering, DBSCAN, GMM EM clustering, agglomerative hierarchical clustering or graph group detection.

5. The commercial vehicle gear position identification method based on operating condition driving data according to claim 4, characterized in that: The DBSCAN includes: Load data; The user enters the total number of gears known; Only keep the rows where the value of the Epm_nEng (rpm) column is between 1200 and 1500; Convert a nested list into a two-dimensional NumPy array; Calculate the mean of each column; Calculate the standard deviation of each column; Standardize each column of data; Clustering using DBSCAN; Get cluster labels and core sample indexes; Output core point data; Calculate possible transmission ratios; Output the mode as the inferred transmission ratio; Extract the rows where the vehicle speed is between certain values ​​and calculate the minimum transmission ratio; Convert a nested list to a two-dimensional NumPy array; Process the array, divide the value of the first column by the value of the second column, and retain 2 decimal places Calculate the mode; and Plot the clustering results.

6. The commercial vehicle gear position identification method based on operating condition driving data according to claim 5, characterized in that: The core point data includes: Make sure the array has at least two columns; Extract the first and second columns of data; Calculate the result of dividing the first column by the second column; Sort the results; Create an empty list to store the division results; Iterate over the sorted array, starting at the second element, dividing the current element by the previous element and appending the result to the list; and Filter out the results in the list whose values ​​are between 1.1 and 1.

6.

7. The commercial vehicle gear position identification method based on operating condition driving data according to claim 5, characterized in that: The calculation of possible transmission ratio values ​​includes: Round each result to two decimal places; and Use the mode function from the scipy library to find the mode.

8. The commercial vehicle gear position identification method based on operating condition driving data according to claim 4, characterized in that: The agglomerative hierarchical clustering includes: Load data; Extract data; Only keep the rows where the value of the Epm_nEng (rpm) column is between 1200 and 1500; Convert a nested list to a two-dimensional NumPy array; Generate simulated data; Calculate the silhouette coefficient for different numbers of clusters; Select the number of clusters with the highest silhouette coefficient; Clustering is performed using the optimal number of clusters; Calculate the centroid of each cluster; plot the clustering results; and Plot the cluster centers.

Images