A data-driven method for fault prediction and diagnosis of unmanned mining trucks

Abstract

This invention belongs to the field of fault early warning technology for unmanned mining trucks, and particularly relates to a data-driven fault pre-diagnosis method for unmanned mining trucks. This method considers the daily stable operation requirements of unmanned mining trucks and takes into account their complex structure. After detecting anomalies, it performs fault diagnosis and provides maintenance suggestions to maintenance personnel. This overcomes the shortcomings of traditional unmanned mining truck health management in terms of "pre-emptive" early warning and diagnosis, thereby improving the timeliness and effectiveness of fault pre-diagnosis and enhancing the stability and reliability of unmanned mining trucks in daily operation.

Description

Technical Field

[0001] This invention belongs to the field of fault early warning technology for unmanned mining trucks, and particularly relates to a data-driven method for fault pre-diagnosis of unmanned mining trucks. Background Technology

[0002] With the rapid development of autonomous driving technology, autonomous driving in closed environments, such as mining areas, has received increasing attention. Compared to manned driving, autonomous driving in mining areas offers advantages such as higher safety and more scientific transportation management. Furthermore, the rapid development of autonomous driving technology in the mining sector can significantly improve equipment production efficiency and reduce losses caused by improper human operation.

[0003] However, the development of unmanned mining trucks still faces considerable obstacles. For example, the fault diagnosis of unmanned mining trucks still has significant room for improvement and breakthroughs. There are three main existing methods for fault diagnosis of unmanned mining trucks: one is post-fault diagnosis and repair, also known as "reactive maintenance," where maintenance personnel manually troubleshoot and locate the fault type and implement targeted repairs when a major fault occurs and affects normal operation; another is "planned maintenance," which requires regular daily or monthly inspections of the mining trucks; and the last is "proactive maintenance," which detects anomalies before major faults occur and allows maintenance personnel to troubleshoot and repair them.

[0004] "Reactive maintenance" requires repairs by maintenance personnel only when a major malfunction occurs. However, malfunctions directly impact the normal operation of unmanned mining trucks, increasing costs. "Planned maintenance," on the other hand, involves periodic inspections regardless of whether a malfunction exists, also resulting in a waste of manpower and resources. Neither of these methods can prevent malfunctions in advance and often leads to unnecessary downtime risks. Existing "proactive maintenance" can only detect anomalies before major malfunctions occur, but it cannot determine the specific type of malfunction, still requiring manual troubleshooting by maintenance personnel. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a data-driven method for the pre-diagnosis of faults in unmanned mining trucks. This method considers the daily stable operation requirements of unmanned mining trucks and takes into account their complex structure. After detecting anomalies, it performs fault diagnosis and provides maintenance suggestions to maintenance personnel. This method overcomes the shortcomings of traditional unmanned mining truck health management in terms of "pre-event" early warning and diagnosis, thereby improving the timeliness and effectiveness of pre-diagnosis of faults in unmanned mining trucks and enhancing the stability and reliability of their daily operation.

[0006] The specific technical solution of the present invention is as follows:

[0007] A data-driven method for fault pre-diagnosis of unmanned mining trucks includes the following steps:

[0008] Step S1: Collect historical driving data of unmanned mining trucks, preprocess the historical driving data to obtain standard data, and store it in the standard database after completion;

[0009] Step S2: Compare the current driving log data of the unmanned mining truck with the standard data in the standard database to determine if there are any abnormalities in the driving log data. If there are no abnormalities, the fault is diagnosed and the driving log data is merged into the standard database; if there are abnormalities, proceed to step S3.

[0010] Step S3: Establish a fault diagnosis model and determine the fault type of the abnormal driving log data in step S2.

[0011] Step S4: Output and display the fault type judgment result obtained in step S3.

[0012] Preferably, the preprocessing in step S1 includes data filtering, feature dimensionality reduction, and feature extraction.

[0013] Preferably, the standard database in step S1 includes a normal data standard database and a fault data standard database, and the normal data and fault data in the standard data are stored in the normal data standard database and the fault data standard database, respectively.

[0014] Preferably, step S2 specifically includes:

[0015] Step S2-1: Obtain the current driving log data of the unmanned mining truck and extract its features;

[0016] Step S2-2: Compare the feature distribution of the vehicle log data extracted in step S2-1 with the normal data standard library to determine whether there are any anomalies in the vehicle log data. If there are no anomalies, the fault is diagnosed and the vehicle log data is merged into the standard database; if there are anomalies, proceed to step S3.

[0017] Preferably, step S3 specifically includes:

[0018] Step S3-1: Establish a fault diagnosis model and use the fault data standard library in the standard database to perform accuracy testing and logic adjustment on the fault diagnosis model;

[0019] Step S3-2: Input the abnormal driving log data from step S2 into the fault diagnosis model, including time, total steering angle command, front wheel steering angle feedback, and lateral deviation;

[0020] Step S3-3: Find the maximum and minimum values ​​of the total steering angle command and the front wheel steering angle feedback, record the time value of the data frame corresponding to each maximum and minimum value, record the difference between the time value of the maximum value of the front wheel steering angle feedback and the time value of the maximum value of the total steering angle command as the first difference, record the difference between the time value of the minimum value of the front wheel steering angle feedback and the time value of the minimum value of the total steering angle command as the second difference, and calculate the average of the first difference and the second difference as the steering lag time;

[0021] Step S3-4: Use the steering hysteresis time to calculate the data difference between the corresponding data frames of all total steering angle commands and front wheel steering angle feedback in the driving log data, and then calculate the difference of the lateral deviation between each adjacent data frame;

[0022] Step S3-5: Determine whether the difference in lateral deviation between adjacent data frames is less than a preset threshold to check for lateral deviation jump point faults. If the difference in lateral deviation between adjacent data frames is not less than the preset threshold, it is determined that the map path is abnormal or the positioning is abnormal, and step 4 is executed. If the difference in lateral deviation between all adjacent data frames is less than the preset threshold, step S3-6 is executed.

[0023] Step S3-6: Determine whether the steering lag time is less than a preset threshold. If it is not less than the threshold, it is determined to be a steering response too slow fault, and step 4 is executed; if it is less than the threshold, step S3-7 is executed.

[0024] Step S3-7: Calculate the average of the data differences between all total steering angle commands and front wheel steering angle feedback data frame points in step S3-4, and determine whether the average is greater than the first preset threshold. If it is greater, it is determined to be a rightward steering mechanism fault. Determine whether the average is less than the second preset threshold. If it is less, it is determined to be a leftward steering mechanism fault. If it is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, it is determined to be a control algorithm performance problem.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0026] (1) The present invention adopts a logical comparison method based on the historical driving data of unmanned mining trucks, which can accurately predict and locate the fault type of unmanned mining trucks, avoiding the judgment error caused by insufficient human experience when maintenance personnel manually troubleshoot in other existing methods.

[0027] (2) The present invention proposes a data-driven method for the pre-diagnosis of unmanned mining truck faults, which can detect abnormalities of unmanned mining trucks in advance and output their fault types, and provide maintenance suggestions for mining truck faults in advance.

[0028] (3) The unmanned mining truck fault pre-diagnosis method based on the present invention can realize the full-cycle health status detection of unmanned mining trucks, and is easy to implement in engineering. It does not require additional sensors, which can effectively reduce costs and reduce the loss of manpower and material resources. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the embodiments will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the accompanying drawings. The accompanying drawings are schematic and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart of a data-driven method for the pre-diagnosis of faults in unmanned mining trucks proposed in this invention.

[0031] Figure 2 This is the judgment logic diagram in the fault diagnosis model in step S3.

[0032] Figure 3 This is a diagram showing the relationship between the total steering angle command and the front wheel steering angle feedback.

[0033] Figure 4 This is a schematic diagram of the maximum and minimum values ​​of the characteristic curve. Detailed Implementation

[0034] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other.

[0035] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0036] Because the transportation routes in enclosed mining areas are singular and daily operational data is similar, this invention constructs a standard database based on historical normal and fault data, and builds a fault diagnosis model. The technical solution is as follows: Figure 1 As shown, the details are as follows:

[0037] Step S1: Collect historical driving data of different unmanned mining trucks in different mining areas. This historical driving data includes normal data and fault data. Fault data is categorized for subsequent accuracy testing and logic adjustments. The historical driving data undergoes data filtering, feature reduction, and feature extraction to obtain standard data, which is then stored in a standard database. This standard database includes a normal data standard library and a fault data standard library, with normal data and fault data stored in the respective libraries.

[0038] Step S2: Obtain the current driving log data of the unmanned mining truck and extract its features. Use the KL divergence method to compare the feature distribution of the extracted driving log data with the normal data standard library to determine whether there are any anomalies in the driving log data. If there are no anomalies, it indicates that there is no fault trend, and the driving log data is diagnosed as fault-free and merged into the standard database; if there are anomalies, proceed to step S3.

[0039] Step S3: Establish a fault diagnosis model and determine the fault type of the abnormal driving log data from Step S2. Specifically, as follows... Figure 2 As shown, the judgment logic is as follows:

[0040] Step S3-1: Establish a fault diagnosis model and use the fault data standard library in the standard database to perform accuracy testing and logic adjustment on the fault diagnosis model;

[0041] Step S3-2: Input the abnormal driving log data from Step S2 into the fault diagnosis model. This data is time-series data, including time, total steering angle command, front wheel steering angle feedback, and lateral deviation. The front wheel steering angle feedback is the response feedback to the total steering angle command, such as... Figure 3 As shown, the solid line represents the total steering angle command, and the dashed line represents the front wheel steering angle feedback;

[0042] Step S3-3: Locate the maximum and minimum values ​​of the total steering angle command and the front wheel steering angle feedback, and record the time value of the data frame corresponding to each maximum and minimum value, such as... Figure 4 As shown, the difference between the time value of the maximum value of the front wheel steering angle feedback and the time value of the maximum value of the total steering angle command is recorded as the first difference, and the difference between the time value of the minimum value of the front wheel steering angle feedback and the time value of the minimum value of the total steering angle command is recorded as the second difference. The average of the first difference and the second difference is the steering lag time.

[0043] Step S3-4: Use the steering hysteresis time to calculate the data difference between the corresponding data frames of all total steering angle commands and front wheel steering angle feedback in the driving log data, and then calculate the difference of the lateral deviation between each adjacent data frame;

[0044] Step S3-5: Determine whether the difference in lateral deviation between adjacent data frames is less than a preset threshold to check for lateral deviation jump point faults. If the difference in lateral deviation between adjacent data frames is not less than the preset threshold, it is determined that the map path is abnormal or the positioning is abnormal, and step 4 is executed. If the difference in lateral deviation between all adjacent data frames is less than the preset threshold, step S3-6 is executed.

[0045] Step S3-6: Determine whether the steering lag time is less than a preset threshold. If it is not less than the threshold, it is determined to be a steering response too slow fault for maintenance personnel to refer to, and step 4 is executed; if it is less than the threshold, step S3-7 is executed.

[0046] Step S3-7: Calculate the average of the data differences between all total steering angle commands and front wheel steering angle feedback data frame points in step S3-4, and determine whether the average is greater than the first preset threshold. If it is greater, it is determined to be a rightward steering mechanism fault. Determine whether the average is less than the second preset threshold. If it is less, it is determined to be a leftward steering mechanism fault. If it is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, it is determined to be a control algorithm performance problem.

[0047] Step S4: Output and display the fault type judgment result obtained in step S3, and notify the maintenance personnel to carry out targeted maintenance so that the maintenance personnel can pay attention to the health problems of the unmanned mining truck in advance before a major fault occurs.

[0048] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0049] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0050] In this invention, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The term "multiple" refers to two or more unless otherwise expressly defined.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data-driven method for pre-diagnosing faults in unmanned mining trucks, characterized in that, Includes the following steps: Step S1: Collect historical driving data of unmanned mining trucks, preprocess the historical driving data to obtain standard data, and store it in the standard database after completion; Step S2: Compare the current driving log data of the unmanned mining truck with the standard data in the standard database to determine if there are any abnormalities in the driving log data. If there are no abnormalities, the fault is diagnosed and the driving log data is merged into the standard database; if there are abnormalities, proceed to step S3. Step S3: Establish a fault diagnosis model and determine the fault type of the abnormal driving log data in step S2. Step S4: Output and display the fault type judgment result obtained in step S3; Step S3 specifically includes: Step S3-1: Establish a fault diagnosis model and use the fault data standard library in the standard database to perform accuracy testing and logic adjustment on the fault diagnosis model; Step S3-2: Input the abnormal driving log data from step S2 into the fault diagnosis model, including time, total steering angle command, front wheel steering angle feedback, and lateral deviation; Step S3-3: Find the maximum and minimum values ​​of the total steering angle command and the front wheel steering angle feedback, record the time value of the data frame corresponding to each maximum and minimum value, record the difference between the time value of the maximum value of the front wheel steering angle feedback and the time value of the maximum value of the total steering angle command as the first difference, record the difference between the time value of the minimum value of the front wheel steering angle feedback and the time value of the minimum value of the total steering angle command as the second difference, and calculate the average of the first difference and the second difference as the steering lag time; Step S3-4: Use the steering hysteresis time to calculate the data difference between the corresponding data frames of all total steering angle commands and front wheel steering angle feedback in the driving log data, and then calculate the difference of the lateral deviation between each adjacent data frame; Step S3-5: Determine whether the difference in lateral deviation between adjacent data frames is less than a preset threshold to check for lateral deviation jump point faults. If the difference in lateral deviation between adjacent data frames is not less than the preset threshold, it is determined that the map path is abnormal or the positioning is abnormal, and step 4 is executed. If the difference in lateral deviation between all adjacent data frames is less than the preset threshold, step S3-6 is executed. Step S3-6: Determine whether the steering lag time is less than a preset threshold. If it is not less than the threshold, it is determined to be a steering response too slow fault, and step 4 is executed; if it is less than the threshold, step S3-7 is executed. Step S3-7: Calculate the average of the data differences between all total steering angle commands and front wheel steering angle feedback data frame points in step S3-4, and determine whether the average is greater than the first preset threshold. If it is greater, it is determined to be a rightward steering mechanism fault. Determine whether the average is less than the second preset threshold. If it is less, it is determined to be a leftward steering mechanism fault. If it is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, it is determined to be a control algorithm performance problem.

2. The method for pre-diagnosing faults in unmanned mining trucks according to claim 1, characterized in that, The preprocessing in step S1 includes data filtering, feature dimensionality reduction, and feature extraction.

3. The method for pre-diagnosing faults in unmanned mining trucks according to claim 1, characterized in that, The standard database in step S1 includes a normal data standard database and a fault data standard database. The normal data and fault data in the standard data are stored in the normal data standard database and the fault data standard database, respectively.

4. The method for pre-diagnosing faults in unmanned mining trucks according to claim 3, characterized in that, Step S2 specifically includes: Step S2-1: Obtain the current driving log data of the unmanned mining truck and extract its features; Step S2-2: Compare the feature distribution of the vehicle log data extracted in step S2-1 with the normal data standard library to determine whether there are any anomalies in the vehicle log data. If there are no anomalies, the fault is diagnosed and the vehicle log data is merged into the standard database; if there are anomalies, proceed to step S3.

Images