Dynamic low-pressure identification method based on tire pressure data

Abstract

The present invention relates to the technical field of pressure recognition, and specifically discloses a dynamic low-pressure recognition method based on tire pressure data, including: obtaining the tire air pressure value, obtaining the low-pressure approach ratio through numerical analysis, constructing an approach ratio data group and analyzing the continuity of the low-pressure approach ratio; if it appears continuously, then draw a time-low-pressure approach ratio change curve, use the local outlier factor algorithm to determine the mutation point, and then obtain the mutation point of the vehicle driving data, and construct a low-pressure time group and a driving time group; by calculating the distance between the two groups of times to construct a time distance matrix and through normalization processing, a normalized time distance matrix is obtained, combined with the vehicle historical data, use the support vector machine algorithm to identify the low-pressure state related to the vehicle driving data, and feedback it to the vehicle control system; this method can effectively identify dynamic low pressure, which helps the vehicle to monitor the tire low-pressure state in a timely manner and is beneficial to ensuring driving safety.

Description

Technical Field

[0001] The present invention relates to the technical field of pressure recognition, and particularly to a dynamic low-pressure recognition method based on tire pressure data. Background Art

[0002] A Chinese patent application with the publication number CN103085612B discloses a tire low-pressure alarm device and method, including: measuring the pressure of each of the plurality of tires respectively, and sending a signal corresponding to the pressure; identifying whether the pressure corresponding to the signal is the pressure of the front tire or the rear tire, and if it is the pressure of the front tire, comparing it with a first reference pressure, and if it is the pressure of the rear tire, comparing it with a second reference pressure to determine whether the tire corresponding to the signal is in a low-pressure state.

[0003] The current technology only simply sets different reference pressures according to the tire position for comparison and judgment, without considering the dynamic influence of various complex factors on the tire pressure during vehicle driving. For example, when the vehicle is driving, changes in vehicle speed, passing through bumpy roads, changes in driving state, and changes in the physical state of the tire itself will all cause fluctuations in tire pressure. If vehicle driving data can be combined, by analyzing the mutation points of the time and air pressure change curve and the correlation with the mutation points of time and driving data, the dynamic low-pressure situation can be identified more comprehensively and accurately.

[0004] When the current technology judges the low-pressure state, it lacks in-depth analysis of the continuity and mutation of pressure data. Relying only on comparison with the reference pressure, it is difficult to detect the trend and characteristics of abnormal changes in pressure data. If a proximity ratio data group is constructed, the run test method and information entropy are used to calculate the continuous determination value, and it is judged whether the low-pressure proximity ratio appears continuously, and then the local outlier factor algorithm is combined to accurately determine the mutation point. Compared with the current technology, it can capture abnormal changes in tire pressure more timely, which is beneficial to timely discovering potential low-pressure risks.

[0005] Therefore, the present invention provides a dynamic low-pressure recognition method based on tire pressure data. Summary of the Invention

[0006] The purpose of the present invention is to provide a dynamic low-pressure recognition method based on tire pressure data to solve the problems in the above background.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A dynamic low-pressure recognition method based on tire pressure data includes the following steps:

[0009] Step 1: Obtain the air pressure value of the tire, perform numerical analysis on the air pressure value of the tire, and obtain the low-pressure proximity ratio;

[0010] Step 2: Obtain the low-pressure proximity ratios at all monitoring times within the monitoring period, construct a proximity ratio data group, perform continuity analysis on the low-pressure proximity ratios in the proximity ratio data group, and determine whether the low-pressure proximity ratios appear continuously;

[0011] Step 3: If the low-pressure proximity ratios in the proximity ratio data group appear continuously, plot the change curve of time - low-pressure proximity ratio, and determine the mutation points of the low-pressure proximity ratio on the change curve;

[0012] Step 4: Based on the mutation points of the low-pressure proximity ratio on the change curve, obtain the mutation points of the vehicle driving data, and construct a low-pressure time group and a driving time group;

[0013] Step 5: Perform time distance analysis on the low-pressure time group and the driving time group, construct a normalized time distance matrix, and use machine learning methods to analyze the normalized time distance matrix to identify the low-pressure state related to the vehicle driving data.

[0014] As a further solution of the present invention: The obtaining method of the low-pressure proximity ratio is as follows:

[0015] During the monitoring period, obtain the air pressure value of the tire in real time at each monitoring time;

[0016] Obtain the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value, and use the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value as the low-pressure range value;

[0017] If the air pressure value is less than the minimum endpoint value of the low-pressure range value, generate a tire low-pressure warning to the vehicle control system;

[0018] If the air pressure value is within the low-pressure range value, perform difference processing on the air pressure value and the minimum endpoint value of the low-pressure range value to obtain the air pressure proximity difference;

[0019] Perform ratio processing on the air pressure proximity difference and the length of the low-pressure range value to obtain the low-pressure proximity ratio.

[0020] As a further solution of the present invention: The determination method of whether the low-pressure proximity ratio appears continuously is as follows:

[0021] If a low-pressure proximity ratio is generated at the monitoring time within the monitoring period, store the low-pressure proximity ratio in the proximity ratio data group;

[0022] If a low-pressure proximity ratio is not generated at the monitoring time within the monitoring period, store the value "0" in the proximity ratio data group and mark it as a null value proximity ratio;

[0023] Perform numerical analysis on the proximity ratio data group to obtain the probability of the low-pressure proximity ratio and the continuous null value ratio in each sub-section of the proximity ratio data group;

[0024] Based on the probability of the low - voltage approach ratio and the continuous null - value ratio within each sub - section, through the information - entropy formula, obtain the continuous determination value of the approach - ratio data group;

[0025] If the continuous determination value is lower than the continuous determination threshold, it is considered that the low - voltage approach ratios in the approach - ratio data group appear continuously.

[0026] As a further solution of the present invention: The method for obtaining the probability of the low - voltage approach ratio within each sub - section of the approach - ratio data group is as follows:

[0027] Based on the approach - ratio data group, divide the approach - ratio data group into multiple sections according to the length of the fixed monitoring time to obtain sub - sections;

[0028] Obtain the number of low - voltage approach ratios and the number of null - value approach ratios within the sub - section, sum up the number of low - voltage approach ratios and the number of null - value approach ratios to obtain the section - number sum;

[0029] Perform a ratio process on the number of low - voltage approach ratios within the sub - section and the section - number sum to obtain the probability of the low - voltage approach ratio within each sub - section.

[0030] As a further solution of the present invention: The method for obtaining the continuous null - value ratio is as follows:

[0031] Use the run - length detection method to calculate the maximum value of the number of consecutive null - value approach ratios within each sub - section to obtain the section - continuous null value;

[0032] Perform a ratio process on the section - continuous null value and the section - number sum to obtain the continuous null - value ratio.

[0033] As a further solution of the present invention: The method for obtaining the mutation point of the low - voltage approach ratio on the change curve is as follows:

[0034] If the low - voltage approach ratios in the approach - ratio data group appear continuously, with the monitoring time as the X - axis and the low - voltage approach ratio as the Y - axis, plot the time - low - voltage approach - ratio change curve in the two - dimensional rectangular coordinate system;

[0035] Use the local outlier factor algorithm to establish a mutation analysis model to determine the mutation point of the low - voltage approach ratio of the time - low - voltage approach - ratio change curve.

[0036] As a further solution of the present invention: The method for establishing the mutation analysis model is as follows:

[0037] Obtain the low - voltage approach ratio corresponding to each monitoring time within the time - low - voltage approach - ratio curve and mark it as a data detection point;

[0038] Obtain the reachable distance d of the data detection point to m data detection points around it in the two - dimensional rectangular coordinate system k , where k is the number of the surrounding data detection points, and the value range of k is [1,m];

[0039] By the formula: , obtain the reachable distance from the data detection point to the surrounding data detection points, where d k1 is the Chebyshev distance from the data detection point to the surrounding data detection points, and d k2 is the Euclidean distance from the data detection point to the surrounding data detection points;

[0040] By the formula: Obtain the anomaly factor Lo of the data detection point;

[0041] When the anomaly factor Lo is higher than the preset factor determination value, it is considered that the data detection point is a mutation point on the time - low - pressure approach ratio change curve.

[0042] As a further solution of the present invention: the construction methods of the low - pressure time group and the driving time group are as follows:

[0043] Based on the monitoring time corresponding to the mutation point of the low - pressure approach ratio on the change curve, obtain the vehicle driving data from the vehicle driving log;

[0044] Where the vehicle driving data includes: vehicle speed, vehicle bump value, and tire temperature;

[0045] Based on the vehicle driving data, with time as the X - axis and the vehicle driving data as the Y - axis, draw three change curves of time - vehicle driving data composed of vehicle speed, vehicle bump value, and tire temperature in a two - dimensional rectangular coordinate system;

[0046] Based on the local anomaly factor algorithm, obtain the mutation points of the three change curves of time - vehicle driving data, and mark them as driving data mutation points;

[0047] Obtain the monitoring time corresponding to the mutation point of the low - pressure approach ratio within the monitoring period, and construct the low - pressure time group;

[0048] Obtain the monitoring time corresponding to the driving data mutation point within the monitoring period, and construct the driving time group.

[0049] As a further solution of the present invention: the identification method of the low - pressure state related to the vehicle driving data is as follows:

[0050] Based on the low - pressure time group and the driving time group, calculate the time distance between the low - pressure time group and the driving time group and use it as the data element of the time - distance matrix, and construct the time - distance matrix;

[0051] Perform normalization processing on each data element in the time - distance matrix to construct a normalized time - distance matrix;

[0052] Obtain the historical normalized time - distance matrix of the vehicle from the vehicle driving log to obtain the historical time - distance matrix;

[0053] Use the historical moment distance matrix in combination with the normalized moment distance matrix and the support vector machine algorithm to identify, in the normalized moment distance matrix, the monitoring moments at which the mutation points in the low-pressure moment group are caused by the mutation points in the driving moment group;

[0054] Obtain the low-pressure approach ratio corresponding to the monitoring moments at which the mutation points in the low-pressure moment group are caused by the mutation points in the driving moment group, and obtain the low-pressure state related to the vehicle driving data.

[0055] As a further solution of the present invention: The construction method of DT in the moment distance matrix is as follows:

[0056] Through the formula: Obtain the time distance D(t j1 , s j2 ), where represents each monitoring moment in the low-pressure moment group, represents each monitoring moment in the driving moment group;

[0057] Through the recurrence formula: Fill the data elements of DT in the moment distance matrix,

[0058] where respectively represent the moment distance of the cell above DT in the moment distance matrix, the moment distance of the cell on the left, and the moment distance of the cell in the upper left.

[0059] Advantages of the present invention:

[0060] 1. By obtaining the tire pressure value in real time and calculating the low-pressure approach ratio, and using the run test method and information entropy to analyze the continuity of the low-pressure approach ratio, it is beneficial to timely monitor the abnormal fluctuations of the low-pressure approach ratio. For example, during long-term vehicle driving, the tire pressure may slowly decrease due to factors such as temperature change and slight air leakage. Traditional methods may not be able to detect this gradual change in time. However, through continuous monitoring and analysis of the low-pressure approach ratio in the present invention, these abnormalities can be captured in advance, and a warning can be issued before the tire pressure problem deteriorates.

[0061] 2. By using the local outlier factor algorithm to determine the mutation points of the low-pressure approach ratio, and at the same time combining the vehicle driving data (vehicle speed, vehicle bump value, tire temperature) to construct the low-pressure moment group and the driving moment group for correlation analysis, it can be judged that the mutation of the low-pressure approach ratio is caused by the change of the vehicle driving state; for example, when the vehicle makes a sharp turn, the low-pressure approach ratio of the tire changes suddenly, and the mutation moment highly coincides with the mutation point of the driving data, which is beneficial to accurately judge that this low-pressure abnormality is caused by the vehicle driving operation, providing a basis for taking targeted measures subsequently. Brief Description of the Drawings

[0062] The present invention will be further described below in conjunction with the accompanying drawings.

[0063] Figure 1 is a flowchart of the air pressure acquisition and continuity analysis of the present invention;

[0064] Figure 2 is a flowchart of the correlation analysis between air pressure and vehicle driving state in the present invention. Specific Embodiments

[0065] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0066] Embodiment 1

[0067] Please refer to Figure 1 As shown, the present invention is a dynamic low-pressure identification method based on tire pressure data, including the following steps:

[0068] Step 1: Obtain the air pressure value of the tire, perform numerical analysis on the air pressure value of the tire, and obtain the low-pressure proximity ratio;

[0069] In some embodiments, a pressure sensor is installed inside the tire to collect the air pressure value inside the tire in real time;

[0070] During the monitoring period, the air pressure value of the tire is obtained in real time at each monitoring moment;

[0071] It should be noted that there are multiple monitoring moments during the monitoring period;

[0072] Obtain the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value, and use the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value as the low-pressure range value;

[0073] It should be noted that the air pressure range value and the low-pressure warning value are set by those skilled in the art, and the low-pressure warning value is lower than the minimum endpoint value of the air pressure range value;

[0074] If the air pressure value is less than the minimum endpoint value of the low-pressure range value, a tire low-pressure warning is generated to the vehicle control system;

[0075] If the air pressure value is within the low-pressure range value, perform a difference process on the air pressure value and the minimum endpoint value of the low-pressure range value to obtain an air pressure proximity difference;

[0076] Perform a ratio process on the air pressure proximity difference and the length of the air pressure range value to obtain the low-pressure proximity ratio;

[0077] Step 2: Obtain the low - pressure approach ratios at all monitoring times within the monitoring period, construct an approach - ratio data group, perform a continuity analysis on the low - pressure approach ratios in the approach - ratio data group, and determine whether the low - pressure approach ratios appear continuously;

[0078] Within the monitoring period, judge the low - pressure approach ratios generated at each monitoring time and construct an approach - ratio data group;

[0079] Specifically, if a low - pressure approach ratio is generated at a monitoring time within the monitoring period, store the low - pressure approach ratio in the approach - ratio data group;

[0080] If no low - pressure approach ratio is generated at a monitoring time within the monitoring period, store the value "0" in the approach - ratio data group and mark it as a null - value approach ratio;

[0081] Based on the approach - ratio data group, divide the approach - ratio data group into multiple segments according to the length of the fixed monitoring time to obtain sub - segments;

[0082] Obtain the number of low - pressure approach ratios and the number of null - value approach ratios within the sub - segment, sum up the number of low - pressure approach ratios and the number of null - value approach ratios to obtain the segment quantity sum;

[0083] Perform a ratio process on the number of low - pressure approach ratios within the sub - segment and the segment quantity sum to obtain the probability of the low - pressure approach ratio within each sub - segment, and mark the probability of the low - pressure approach ratio within each sub - segment as q i , where i is the number of the sub - segment, the value range of i is [1, n], and n is the total number of sub - segments of the approach - ratio data group;

[0084] Use the run - test method to calculate the maximum value of the number of consecutive null - value approach ratios within each sub - segment to obtain the segment consecutive null value;

[0085] It should be noted that the run - test method scans the low - pressure approach ratios within each sub - segment. If a null - value approach ratio is scanned, continue to scan whether the next data in the sub - segment is a null - value approach ratio. If it is a null - value approach ratio, continue to scan until the data of the low - pressure approach ratio is scanned to obtain the consecutive null - value approach ratios. When the run - test method identifies the last data within the sub - segment, compare all the consecutive null - value approach ratios within the sub - segment to obtain the maximum value of the number of consecutive null - value approach ratios within the sub - segment, that is, the segment consecutive null value;

[0086] Exemplarily, if the sub - segment is [0.25, 0, 0, 0.75, 0, 1.1], then the segment consecutive null value of the sub - segment is 2;

[0087] Perform a ratio process on the segment consecutive null value and the segment quantity sum to obtain the consecutive null - value ratio, and mark the consecutive null - value ratio as l i ;

[0088] Through the formula: Obtain the continuous determination value H of the approach ratio data group, where a1 and a2 are 1.25 and 0.86 respectively;

[0089] It should be noted that 、 respectively represent the information entropy of the low-pressure approach ratio and the continuous null value ratio in the approach ratio data group, which is used to determine the continuity of the low-pressure approach ratio in the approach ratio data group. When the information entropy of the low-pressure approach ratio is smaller, it indicates that the distribution of the low-pressure approach ratio in the approach ratio data group is more concentrated. When the information entropy of the continuous null value ratio is smaller, it indicates that the continuous null value ratio appears more concentrated;

[0090] Compare the continuous determination value H with the preset continuous determination threshold to determine whether the low-pressure approach ratio in the approach ratio data group appears continuously;

[0091] If the continuous determination value H is lower than the continuous determination threshold, it indicates that the distribution of the low-pressure approach ratio is more concentrated and continuous, then it is considered that the low-pressure approach ratio in the approach ratio data group appears continuously;

[0092] If the continuous determination value H is higher than the continuous determination threshold, the distribution of the low-pressure approach ratio is more dispersed and discrete, then it is considered that the low-pressure approach ratio in the approach ratio data group does not appear continuously;

[0093] The technical solution of this embodiment is: obtain the air pressure value of the tire, perform numerical analysis on the air pressure value of the tire to obtain the low-pressure approach ratio, obtain the low-pressure approach ratios at all monitoring times within the monitoring period, construct an approach ratio data group, perform continuity analysis on the low-pressure approach ratios in the approach ratio data group, and determine whether the low-pressure approach ratio appears continuously, which is beneficial to timely monitoring of the abnormal fluctuations of the low-pressure approach ratio of the tire, and at the same time judge the change trend of the tire low pressure in the time dimension.

[0094] Embodiment 2

[0095] Step 3: If the low-pressure approach ratio in the approach ratio data group appears continuously, draw a time-low-pressure approach ratio change curve and determine the mutation point of the low-pressure approach ratio on the change curve;

[0096] If the low-pressure approach ratio in the approach ratio data group appears continuously, with the monitoring time as the X-axis and the low-pressure approach ratio as the Y-axis, draw a time-low-pressure approach ratio change curve in the two-dimensional rectangular coordinate system;

[0097] Use the local outlier factor algorithm to construct a mutation analysis model and determine the mutation point of the low-pressure approach ratio of the time-low-pressure approach ratio change curve;

[0098] It should be noted that in the time - low pressure approach ratio change curve, the mutation point refers to the point where the low pressure approach ratio value changes sharply at a certain monitoring moment. When the low pressure approach ratio of the tire is at the mutation point, it may be due to a change in the physical state of the tire. For example, the tire is punctured by a sharp object, causing the gas inside the tire to leak to the outside, resulting in a change in the tire pressure;

[0099] If the pressure sensor malfunctions at a certain monitoring moment, such as momentary signal interference or hardware failure, resulting in a change in the pressure value of the pressure sensor inside the tire, or if the vehicle driving state changes, such as the vehicle passing through a bumpy road section, sudden braking, sharp turning, etc., resulting in a change in the tire pressure, which causes a mutation in the low pressure approach ratio;

[0100] Specifically, the establishment method of the mutation analysis model is as follows:

[0101] Obtain the low pressure approach ratio corresponding to each monitoring moment in the time - low pressure approach ratio curve and mark it as a data detection point;

[0102] Obtain the reachable distance d of the data detection point to m data detection points around it in the two - dimensional rectangular coordinate system k , where k is the number of the surrounding data detection points, and the value range of k is [1, m];

[0103] Preferably, m = 5;

[0104] Through the formula: , obtain the reachable distance from the data detection point to the surrounding data detection points, where d k1 is the Chebyshev distance from the data detection point to the surrounding data detection points, and d k2 is the Euclidean distance from the data detection point to the surrounding data detection points;

[0105] Exemplarily, if the coordinates of the data detection point are (x0, y0) and the coordinates of one of the surrounding data detection points are (x1, y1), through the formula: Obtain the Chebyshev distance, and through the formula: Obtain the Euclidean distance;

[0106] It should be noted that the Chebyshev distance can reflect the influence of the more significant factor in the monitoring moment or the low pressure approach ratio;

[0107] The Euclidean distance comprehensively considers the information of both the monitoring moment and the low pressure approach ratio, reflects the straight - line distance between two points in the two - dimensional space. By comparing and analyzing the Chebyshev distance and the Euclidean distance, the distance relationship between data points can be measured from different angles, more comprehensively capturing the differences between data points in different dimensions, which helps to more accurately identify the mutation points;

[0108] Through the formula: Obtain the anomaly factor Lo of the data detection point;

[0109] When the anomaly factor Lo is higher than the preset factor determination value, it is considered that the data detection point is a mutation point on the time - low - pressure approach ratio change curve;

[0110] Step 4: Based on the mutation points of the low - pressure approach ratio on the change curve, obtain the mutation points of the vehicle driving data, and construct a low - pressure time group and a driving time group;

[0111] Based on the monitoring time corresponding to the mutation points of the low - pressure approach ratio on the change curve, obtain the vehicle driving data from the vehicle's driving log;

[0112] Wherein the vehicle driving data includes: vehicle speed, vehicle bump value, and tire temperature;

[0113] Wherein the vehicle speed and tire temperature are directly measured by the vehicle's speed sensor and temperature sensor, and the vehicle bump value is obtained by the displacement sensor installed on the vehicle suspension by measuring the vertical displacement of the suspension;

[0114] Based on the vehicle driving data, with time as the X - axis and the vehicle driving data as the Y - axis, in a two - dimensional rectangular coordinate system, draw three change curves of time - vehicle driving data composed of vehicle speed, vehicle bump value, and tire temperature;

[0115] Based on the local anomaly factor algorithm, obtain the mutation points of the three change curves of time - vehicle driving data, and mark them as driving data mutation points;

[0116] Obtain the monitoring time corresponding to the mutation points of the low - pressure approach ratio within the monitoring period, and construct a low - pressure time group, where each monitoring time in the low - pressure time group is marked as t j1 , j1 represents the number of each monitoring time in the low - pressure time group, and the value range of j1 is [1, a1];

[0117] Obtain the monitoring time corresponding to the mutation points of the driving data within the monitoring period, and construct a driving time group, where each monitoring time in the driving time group is marked as s j2 , j2 represents the number of each monitoring time in the driving time group, and the value range of j2 is [1, a2];

[0118] Step 5: Conduct time - distance analysis on the low - pressure time group and the driving time group, construct a normalized time - distance matrix DTW, and use machine learning methods to analyze the normalized time - distance matrix DTW to identify the low - pressure state related to the vehicle driving data;

[0119] Based on the low - pressure time group and the driving time group, calculate the time distance D(t j1 , sj2 ), and as the data elements of the time-distance matrix, construct the time-distance matrix DT;

[0120] It should be noted that the low-voltage time group records the monitoring time when the low-voltage approach ratio approaches the mutation point, and the driving time group records the monitoring time when the driving data mutates. During the vehicle operation, there may be a time dislocation in the changes of the low-voltage time group and the driving time group;

[0121] The time-distance matrix DT can quantify their alignment degree in time under the condition of allowing the time axis to bend. For example, even if the mutations of the low-voltage approach ratio and the driving data do not occur strictly synchronously, DT can find the most reasonable corresponding relationship through dynamic programming and judge the tightness of the time correlation between the low-voltage time group and the driving time group;

[0122] Among them, the dimension of the time-distance matrix DT is a1*a2;

[0123] Specifically, through the formula: Obtain the time distance D(t j1 , s j2 ), where represents each monitoring time in the low-voltage time group, represents each monitoring time in the driving time group;

[0124] Through the recursive formula: Fill the data elements in the time-distance matrix DT,

[0125] where respectively represent the time distance of the cell above the time-distance matrix DT, the time distance of the cell on the left, and the time distance of the cell in the upper left;

[0126] It should be noted that if the time distances of the cell above the time-distance matrix DT, the cell on the left, and the cell in the upper left do not exist, the numerical values of the time distances of the cell above the time-distance matrix DT, the cell on the left, and the cell in the upper left are 0;

[0127] Perform normalization processing on each data element in the time-distance matrix DT, and map each data element in the time-distance matrix DT to the range of [0,1];

[0128] Through the formula: Obtain the value DTW of each normalized data element;

[0129] Based on the normalized data element value DTW, construct the normalized time-distance matrix DTW;

[0130] Obtain the historical driving data of the vehicle from the vehicle driving log, construct the normalized moment distance matrix DTW of the vehicle history, and obtain the historical moment distance matrix;

[0131] Use the support vector machine algorithm by combining the historical moment distance matrix with the normalized moment distance matrix DTW to identify, in the normalized moment distance matrix DTW, the monitoring moments at which the mutation points in the low-voltage moment group are caused by the mutation points in the driving moment group;

[0132] Obtain the low-voltage approach ratio corresponding to the monitoring moments at which the mutation points in the low-voltage moment group are caused by the mutation points in the driving moment group, obtain the low-voltage state related to the vehicle driving data, and feedback it to the vehicle control system;

[0133] It should be noted that the support vector machine algorithm performs model training on the normalized moment distance matrix of the vehicle history and the current normalized moment distance matrix, selects the SVM kernel function to identify the low-voltage approach ratio corresponding to the monitoring moments at which the mutation points in the low-voltage moment group are caused by the mutation points in the driving moment group; the vehicle control system is for the vehicle itself, a system that assists the driver in driving the vehicle or replaces the driver in autonomous driving, and is included in the vehicle itself.

[0134] The technical solution of this embodiment is as follows: If the low-voltage approach ratios in the approach ratio data group appear continuously, draw the change curve of time - low-voltage approach ratio, determine the mutation points of the low-voltage approach ratio on the change curve, obtain the mutation points of the vehicle driving data based on the mutation points of the low-voltage approach ratio on the change curve, construct the low-voltage moment group and the driving moment group, perform moment distance analysis on the low-voltage moment group and the driving moment group, construct the normalized moment distance matrix DTW, and use machine learning methods to analyze the normalized moment distance matrix DTW to identify the low-voltage state related to the vehicle driving data, which is beneficial to accurately judge whether this low-voltage anomaly is caused by vehicle driving operations and provide a basis for subsequent targeted measures.

[0135] The above has described a detailed description of an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as used to limit the scope of implementation of the present invention. All equal changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the present invention.

Claims

1. A dynamic low pressure identification method based on tire pressure data, characterized in that: It includes the following steps: Step 1: Obtain the air pressure value of the tire, conduct numerical analysis on the air pressure value of the tire, and obtain the low-pressure proximity ratio; The obtaining method of the low-pressure proximity ratio is as follows: During the monitoring period, obtain the air pressure value of the tire at each monitoring moment in real time; Obtain the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value, and use the minimum endpoint value of the preset air pressure range value and the preset low-pressure warning value as the low-pressure range value; If the air pressure value is within the low-pressure range value, perform a difference process on the air pressure value and the minimum endpoint value of the low-pressure range value to obtain the air pressure proximity difference; Perform a ratio process on the air pressure proximity difference and the length of the low-pressure range value to obtain the low-pressure proximity ratio; Step 2: Obtain the low-pressure proximity ratios at all monitoring moments during the monitoring period, construct a proximity ratio data group, conduct continuity analysis on the low-pressure proximity ratios in the proximity ratio data group, and determine whether the low-pressure proximity ratios appear continuously; Step 3: If the low-pressure proximity ratios in the proximity ratio data group appear continuously, draw a change curve of time - low-pressure proximity ratio, and determine the mutation points of the low-pressure proximity ratio on the change curve; Step 4: Based on the mutation points of the low-pressure proximity ratio on the change curve, obtain the mutation points of the vehicle driving data, and construct a low-pressure moment group and a driving moment group; Step 5: Conduct moment distance analysis on the low-pressure moment group and the driving moment group, construct a normalized moment distance matrix, and use machine learning methods to analyze the normalized moment distance matrix to identify the low-pressure state related to the vehicle driving data.

2. The dynamic low pressure identification method based on tire pressure data according to claim 1, wherein: The determination method of whether the low-pressure proximity ratios appear continuously is as follows: If a low-pressure proximity ratio is generated at a monitoring moment during the monitoring period, store the low-pressure proximity ratio in the proximity ratio data group; If no low-pressure proximity ratio is generated at a monitoring moment during the monitoring period, store the value "0" in the proximity ratio data group and mark it as a null proximity ratio; Conduct numerical analysis on the proximity ratio data group to obtain the probability of the low-pressure proximity ratio and the continuous null ratio in each sub-section of the proximity ratio data group; Based on the probability of the low-pressure proximity ratio and the continuous null ratio in each sub-section, obtain the continuous determination value of the proximity ratio data group through the information entropy formula; If the continuous determination value is lower than the continuous determination threshold, it is considered that the low-pressure proximity ratios in the proximity ratio data group appear continuously.

3. The dynamic low pressure identification method based on tire pressure data according to claim 2, characterized in that: The obtaining method of the probability of the low-pressure proximity ratio in each sub-section of the proximity ratio data group is as follows: Based on the proximity ratio data group, divide the proximity ratio data group into multiple sections according to the length of the fixed monitoring moment to obtain sub-sections; Obtain the number of low-pressure proximity ratios and the number of null proximity ratios in the sub-section, and perform a summation process on the number of low-pressure proximity ratios and the number of null proximity ratios to obtain the section number sum; Perform a ratio process on the number of low-pressure proximity ratios in the sub-section and the section number sum to obtain the probability of the low-pressure proximity ratio in each sub-section.

4. The dynamic low pressure identification method based on tire pressure data according to claim 3, characterized in that: The obtaining method of the continuous null ratio is as follows: Use the run test method to calculate the maximum value of the number of continuous null proximity ratios in each sub-section to obtain the section continuous null value; Perform a ratio process on the section continuous null value and the section number sum to obtain the continuous null ratio.

5. The dynamic low pressure identification method based on tire pressure data according to claim 1, characterized in that: The obtaining method of the mutation points of the low-pressure proximity ratio on the change curve is as follows: If the low-pressure approach ratio continuously appears in the approach ratio data group, with the monitoring time as the X-axis and the low-pressure approach ratio as the Y-axis, plot the time-low-pressure approach ratio change curve in a two-dimensional rectangular coordinate system; Use the local outlier factor algorithm to establish a mutation analysis model and determine the mutation points of the low-pressure approach ratio of the time-low-pressure approach ratio change curve.

6. The dynamic low pressure identification method based on tire pressure data according to claim 5, characterized in that: The establishment method of the mutation analysis model is as follows: Obtain the low-pressure approach ratio corresponding to each monitoring time in the time-low-pressure approach ratio curve and mark it as a data detection point; Obtain the reachable distance d of m data detection points around the data detection point in the two-dimensional rectangular coordinate system k , where k is the number of the surrounding data detection points, and the value range of k is [1, m]; Obtain the reachable distance from the data detection point to the surrounding data detection points through the formula: , where d k1 is the Chebyshev distance from the data detection point to the surrounding data detection points, and d k2 is the Euclidean distance from the data detection point to the surrounding data detection points; Through the formula: Obtain the anomaly factor Lo of the data detection point; When the outlier factor Lo is higher than the preset factor determination value, the data detection point is considered as the mutation point on the time-low-pressure approach ratio change curve.

7. The dynamic low pressure identification method based on tire pressure data according to claim 1, wherein: The construction methods of the low-pressure time group and the driving time group are as follows: Based on the monitoring time corresponding to the mutation point of the low-pressure approach ratio on the change curve, obtain the vehicle driving data from the vehicle driving log; Among them, the vehicle driving data includes: vehicle speed, vehicle bump value, and tire temperature; Based on the vehicle driving data, plot three change curves of time-vehicle driving data composed of vehicle speed, vehicle bump value, and tire temperature in a two-dimensional rectangular coordinate system; Based on the local outlier factor algorithm, obtain the mutation points of the three change curves of time-vehicle driving data and mark them as driving data mutation points; Obtain the monitoring time corresponding to the mutation point of the low-pressure approach ratio within the monitoring period and construct a low-pressure time group; Obtain the monitoring time corresponding to the mutation point of the driving data within the monitoring period and construct a driving time group.

8. The dynamic low pressure identification method based on tire pressure data according to claim 7, characterized in that: The identification method of the low-pressure state related to the vehicle driving data is as follows: Based on the low-pressure time group and the driving time group, calculate the time distance between the low-pressure time group and the driving time group and use it as the data element of the time distance matrix to construct a time distance matrix; Normalize each data element in the time distance matrix to construct a normalized time distance matrix; Obtain the historical normalized time distance matrix of the vehicle from the vehicle driving log to get the historical time distance matrix; Use the support vector machine algorithm to combine the historical time distance matrix with the normalized time distance matrix to identify the monitoring time in the normalized time distance matrix where the mutation point in the driving time group causes the mutation point in the low-pressure time group; Obtain the low-pressure approach ratio corresponding to the monitoring time when the mutation point in the driving time group causes the mutation point in the low-pressure time group to obtain the low-pressure state related to the vehicle driving data.

9. The dynamic low pressure identification method based on tire pressure data according to claim 8, characterized in that: The construction method of DT in the time distance matrix is as follows: Through the formula: Obtain the time distance D(t j1 , s j2 ), where represents each monitoring moment in the low-voltage moment group, represents each monitoring moment in the driving moment group; Through the recursive formula: Fill the data elements of DT in the time-distance matrix, where respectively represent the time-distance of the cell above the time-distance matrix DT, the time-distance of the cell to the left, and the time-distance of the cell to the upper left.

Images