Unmanned mine car customized point inspection method based on importance degree

Abstract

The invention discloses an unmanned mine car customized point inspection method based on importance. The method comprises the steps that S1, all hardware of an unmanned mine car is initialized into a selected hardware set F = {f1, f2,..., fn} to be subjected to point inspection; s2, based on a data fusion method, calculating the importance degree of each hardware in the selected hardware set F under the current production task; s3, deleting the current hardware with the lowest importance from the selected hardware set F; s4, judging whether the comprehensive importance of the residual hardware in the selected hardware set F after the step S3 is executed is improved compared with that before deletion; if yes, returning to execute the step S2, recalculating the importance degree of each hardware based on the updated selected hardware set F, and performing the next round of screening; if not, cancelling the deletion operation in the step S3, and adding the hardware into the selected hardware set F again; and S5, outputting the customized point inspection scheme. The efficiency and the pertinence of spot inspection work are improved, and the operation and maintenance cost is reduced at the same time.

Description

Technical Field

[0001] This invention relates to the field of hardware inspection technology for unmanned mining trucks, and in particular to a customized inspection method for unmanned mining trucks based on importance. Background Technology

[0002] Hardware inspection of unmanned mining vehicles is a fundamental system in mine production management. It aims to accurately grasp the technical condition of mining vehicles through regular and systematic inspections, maintain and improve vehicle performance, prevent accidents, reduce unplanned downtime, extend equipment lifespan, and reduce maintenance costs, thereby ensuring safe, continuous, and efficient mine production.

[0003] Currently, in the operation and maintenance management of traditional and unmanned mining trucks, hardware inspections generally adopt a standardized, periodic, and comprehensive manual inspection model. Inspectors must check each hardware component of each vehicle according to a fixed checklist and process, and record or upload the results to the management system. While this model possesses a certain degree of systematicity and traceability, the following problems have gradually emerged in practical applications: 1. Low inspection efficiency and high resource consumption: Due to the uniform and comprehensive inspection of all vehicles, regardless of the actual operating status of the hardware, the historical failure rate, and its criticality to the current production task, the same manpower and time must be invested in the inspection. This results in a large amount of inspection resources being consumed on parts that are in good condition or of minor importance, and there is a clear phenomenon of "over-inspection".

[0004] 2. Inflexible and untargeted inspection strategies: Fixed inspection frequencies and scope are difficult to adapt to the differences in production tasks undertaken by different vehicles, changes in operating environments, and the actual importance of the hardware itself. For vehicles performing critical tasks or operating under harsh conditions, the inspection of their core hardware may be neglected due to insufficient frequency or depth, resulting in only surface checks and failing to detect potential faults in a timely manner; conversely, for non-critical or redundant components, unnecessary and frequent inspections may be performed.

[0005] 3. One-sided condition assessment and insufficient decision support: Traditional inspection results often reflect the instantaneous state of individual hardware components in isolation, lacking comprehensive analysis based on the collaborative relationships of multiple hardware components, historical data, and specific task requirements. Therefore, it is difficult to accurately and in real-time assess the overall health of the vehicle system, and it cannot provide accurate data support for developing differentiated and optimal preventive maintenance and repair plans.

[0006] With the advancement of smart mine construction, technologies such as the Internet of Things (IoT) and cloud computing have been applied to the collection and storage of inspection data, for example, synchronizing inspection results to a cloud database through inspection terminals. This lays the foundation for the in-depth utilization of inspection data. However, how to dynamically and intelligently generate customized inspection strategies closely related to production tasks based on this data, especially how to quantify and integrate multi-source information to scientifically assess the importance of different hardware in specific task scenarios, thereby achieving precise allocation and efficient utilization of inspection resources, remains a technical challenge that has not yet been effectively solved by existing technologies.

[0007] Therefore, there is an urgent need for a method that can overcome the above-mentioned defects and dynamically customize the inspection strategy of unmanned mining trucks based on the importance of hardware, so as to improve the intelligence level, efficiency and economic benefits of inspection work. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a customized inspection method for unmanned mining trucks based on importance, comprising the following steps: S1. Initialize the hardware set for inspection: Initialize all hardware of the driverless mining truck to the selected hardware set F = {f1, f2, ..., f...} n}; where, according to the production task to be performed by the unmanned mining truck, the hardware directly associated with the production task is set to the highest importance, and the importance of other indirectly associated hardware is initially set to 0. S2. Calculate hardware importance: Based on the data fusion method, calculate the importance of each piece of hardware in the selected hardware set F under the current production task; S3. Hardware filtering: Remove the hardware with the lowest current importance from the selected hardware set F; S4. Judgment and Iteration: Determine whether the overall importance of the remaining hardware in the selected hardware set F after executing S3 has increased compared to before deletion; If so, return to execute S2, recalculate the importance of each hardware based on the updated selected hardware set F, and perform the next round of screening; If not, then cancel the deletion operation in S3 and add the hardware back to the selected hardware set F; S5. Output customized inspection plan: Output the final selected hardware set F as the best inspection hardware combination for the unmanned mining truck to perform the current production task, so as to guide customized inspection operations.

[0009] Furthermore, in step S2, the calculation of hardware importance based on the data fusion method specifically includes: S21. Construct a mutual support matrix: based on the initial importance of each hardware component. (i=1,2,…,n), calculate any two hardware... and Mutual support between ,form cross support matrix ,in, ; S22. Calculate the support distance: based on the mutual support matrix. Calculate any two hardware and Support distance between The support distance This indicates the difference in consistency between the two hardware components in their support for production tasks. S23. Basic Trust Allocation: Determine the set of weight coefficients for each hardware component based on historical data. (p=1,2,…,n), and based on the mutual support and the weighting coefficients Calculation targeting hardware Basic trust allocation value ; S24. Final Importance Calculation: Based on the basic trust allocation value, data fusion is performed using evidence combination rules to obtain the final importance value of each hardware component.

[0010] Furthermore, the support distance The calculation formula is:

[0011] In the formula, , The mutual support matrix represents the following respectively. No. row and number Row vectors.

[0012] Furthermore, the basic trust assignment value The calculation formula is:

[0013] In the formula, Indicates from the first Based on the evidence provided by the hardware, it is believed that the first The importance assessment of each hardware component is as follows: The level of trust; , These represent the first [number] based on historical data. , Each hardware component has a pre-set weighting coefficient; , They represent the first The hardware pair The level of hardware support, the first The hardware pair The level of hardware support; This indicates the total number of hard components.

[0014] Furthermore, the data fusion using evidence combination rules specifically includes: Calculate the trust level allocation value after fusion It satisfies the formula:

[0015] Then through the formula Calculate the hardware ultimate importance .

[0016] Furthermore, the determination of whether the overall importance of the remaining hardware in the selected hardware set F after executing S3 has increased compared to before deletion specifically involves: After performing the deletion operation, determine whether the sum of the importance values ​​of all remaining hardware in the selected hardware set F is greater than the sum of the importance values ​​of all hardware in the selected hardware set F before deletion.

[0017] Furthermore, in step S4, after determining that the deletion operation is not successful and canceling it, the following steps are also included: Determine whether the selected hardware set F has reached a stable state. If it has, proceed directly to step S5.

[0018] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention initializes the hardware importance associated with tasks and iteratively calculates and dynamically filters hardware based on data fusion methods, ultimately outputting a customized inspection hardware combination. This achieves a shift in inspection strategy from "fixed full inspection" to "dynamic fine inspection," effectively overcoming the problems of over-inspection, resource waste, and rigid strategies in traditional inspection methods. It improves the efficiency and targeting of inspection work, while reducing operation and maintenance costs, providing reliable decision support for the intelligent and precise maintenance of unmanned mining trucks. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1This is the overall flowchart disclosed in this invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] This invention aims to provide a priority-based customized inspection method for unmanned mining trucks, addressing the problems of resource waste and lack of specificity in traditional inspection methods. The core of this method lies in dynamically assessing the importance of each piece of hardware on the vehicle based on specific production tasks, and automatically determining an optimal, customized subset of hardware for inspection through an iterative selection mechanism, thereby achieving precise inspection and improving efficiency.

[0023] The following is combined Figure 1 The flowchart shown illustrates each step of the method of the present invention in detail.

[0024] S1. Initialize the hardware set for inspection: Initialize all hardware of the driverless mining truck to the selected hardware set F = {f1, f2, ..., f...} n}. Among them, based on the production tasks to be performed by the unmanned mining truck, the hardware directly related to the production tasks is set to the highest importance, and the importance of other indirectly related hardware is initially set to 0.

[0025] Those skilled in the art will understand that when an unmanned mining truck receives a specific production task, such as transporting ore from mine A to crushing station B, the system initiates a customized inspection strategy process. First, all inspectable hardware components of the mining truck (such as lidar, millimeter-wave radar, cameras, computing units, battery systems, drive motors, braking systems, etc.) are included in an initial set of selected hardware, denoted as F = {f1, f2, ..., f...} n}, where n is the total number of hardware components.

[0026] Subsequently, based on the nature of the production task, initial importance values ​​are assigned to the hardware in set F. The assignment rule is as follows: hardware directly related to the current production task and having a decisive impact on its completion is assigned the highest importance. For example, for a long-distance transportation task, the battery system and drive motor might be set to the highest importance; for a complex environment navigation task, the perception kit (LiDAR, camera) might be set to the highest importance. Other hardware indirectly related to the current task or of secondary importance is set to an initial importance of 0. For example, the vehicle interior lighting system might be set to 0 for an ore transportation task. The purpose of this step is to establish an initial evaluation benchmark that is strongly related to the task.

[0027] S2. Calculate hardware importance: Based on the data fusion method, calculate the importance of each hardware in the selected hardware set F under the current production task.

[0028] Those skilled in the art will understand that this step aims to utilize data fusion technology to calculate a more dynamic hardware importance that better reflects the actual task requirements, by integrating initial importance, inter-hardware relationships, and historical data. Specifically, this includes the following details: S21. Construct a mutual support matrix: based on the initial importance of each hardware component. (i=1,2,…,n), calculate any two hardware... and Mutual support between ,form cross support matrix .

[0029] in, , and Hardware and The initial importance. When and When they are equal, A value of 1 indicates the highest level of support; the greater the difference between the two, the lower the support level. The closer a value is to 0, the lower the support level.

[0030] Based on the pairwise relationships between all hardware Construct an n×n mutual support matrix. :

[0031] Mutual support matrix It is a symmetric matrix ( = ), its first row vector = [ , , ..., Characterizes the hardware The system is supported by all hardware components.

[0032] S22. Calculate support distance: based on the mutual support matrix. Calculate any two hardware and Support distance between Among them, the supported distance This indicates a difference in the consistency of how well two hardware components support production tasks.

[0033] hardware With hardware Support distance between The calculation formula is:

[0034] In the formula, , Representing the mutual support matrix No. row and number Row vectors. The smaller the value, the better the hardware performance. and The more similar their task support patterns are, the more consistent their roles in the system.

[0035] S23. Basic Trust Allocation: This step incorporates historical experience data to weight and adjust the mutual support levels, forming a preliminary evidence body. First, the set of weight coefficients for each hardware component is determined based on historical data. (p=1,2,…,n), where It is for hardware The weighting coefficients are derived from a comprehensive evaluation of long-term operating data of the hardware, such as historical failure rate, mean time between failures (MTBF), and maintenance costs. A higher value indicates that historical data suggests the hardware is more critical or more vulnerable. Then, calculations are performed on the hardware from the "source of evidence." From the perspective of "proposition" (i.e., hardware) Importance is Basic trust allocation value Basic trust assignment value The calculation formula is:

[0036] In the formula, Indicates from the first Based on the evidence provided by the hardware, it is believed that the first The importance assessment of each hardware component is as follows: The level of trust; , These represent the first [number] based on historical data. , Each hardware component has a pre-set weighting coefficient; , They represent the first The hardware pair The level of hardware support, the first The hardware pair The level of hardware support; This represents the total number of hardware components. The formula is normalized. It means that the hardware... The trust gained depends on its relationship with the source of evidence. Real-time mutual support It also depends on its own historical weight. .

[0037] S24. Final Importance of Fusion Computation: To fuse the viewpoints of all hardware (i.e., all sources of evidence) and resolve potential conflicts of evidence (such as low mutual support among certain hardware components), the following combined evidence rules are adopted for fusion: First, calculate the trust level allocation value after merging. It satisfies the formula:

[0038] The first term of the formula is the product of the trust values ​​of each piece of evidence, representing the fusion of the consistent parts; the second term redistributes the conflict probability (1 minus the sum of the consistent parts) according to the average trust level of each piece of evidence.

[0039] Then through the formula Calculate the hardware ultimate importance .

[0040] Those skilled in the art will understand that this calculation uses the initial importance as a possible "proposition" value, and the fused confidence level... A weighted average is then applied, using these weights to obtain a revised and more reliable importance estimate. All subsequent references to hardware importance in this step refer to this final importance estimate. .

[0041] S3. Hardware Filtering: Remove the hardware with the lowest current importance from the selected hardware set F.

[0042] Specifically, within the currently selected hardware set F, find the one with the lowest final importance value. hardware f min Temporarily remove (delete) the hardware f from set F. minThe purpose of this step is to tentatively streamline the inspection set, eliminating hardware that contributes the least to the current task.

[0043] S4. Judgment and Iteration: This is a crucial feedback and decision-making step. System Judgment: After removing hardware f... min After that, the remaining hardware set (F\{f min Is the sum of the final importance of all hardware in set F greater than the sum of the final importance of all hardware in the original set F before removal? In other words, is the overall importance of the remaining hardware in the selected hardware set F improved after execution of S3 compared to before deletion? If the value increases, it means that removing the low-importance hardware actually improved the "task focus" or "health status attention" of the remaining hardware as a whole. This may be because removing distracting or redundant items allows resources to be more concentrated on critical hardware. At this point, the system confirms the deletion operation and uses a new, smaller hardware set F (which no longer contains f) as the new set. min If the selected hardware set F is the basis, then return to execute S2, recalculate the importance of each hardware based on the updated selected hardware set F, and perform the next round of screening.

[0044] If no improvement is achieved, it indicates that the deletion may have disrupted hardware combinations that positively impact production tasks, removing seemingly unimportant but actually synergistic hardware. In this case, the system reverts the deletion operation in S3 and restores the hardware f. min Re-add to the selected hardware set F. Then, it can be determined that the current set F has reached a local optimum, or other hardware can be removed. As a preferred implementation, when several consecutive attempts (e.g., 3 times) to remove different hardware fail to increase the total importance, the screening process can be considered converged, and the process proceeds to step S5.

[0045] S5. Output customized inspection plan: Output the final selected hardware set F as the best inspection hardware combination for the unmanned mining truck to perform the current production task, so as to guide customized inspection operations.

[0046] Specifically, when the iterative screening process ends (e.g., the convergence condition in step S4 is met), the system outputs the current selected hardware set F as the optimal customized hardware combination for performing the current specific production task for the unmanned mining truck. This combination is the result of dynamic customization; it may only contain a portion of all hardware, but it is the set of core components that most need to be inspected, derived from task importance analysis. Mine maintenance personnel or automated inspection systems can then perform efficient and accurate inspections based on this customized list, rather than performing indiscriminate inspections of all hardware.

[0047] This invention achieves intelligent customization of inspection strategies through a mechanism of task-related initialization, dynamic importance assessment through data fusion, and iterative backward deletion filtering. It can automatically identify and focus on the hardware most critical to the current production task, effectively avoiding over-inspection, improving the efficiency and economic benefits of inspection work, and providing accurate data support for preventative maintenance.

[0048] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A customized inspection method for unmanned mining trucks based on importance, characterized in that, Includes the following steps: S1. Initialize the hardware set for inspection: Initialize all hardware of the driverless mining truck to the selected hardware set F = {f1, f2, ..., f...} n }; where, according to the production task to be performed by the unmanned mining truck, the hardware directly associated with the production task is set to the highest importance, and the importance of other indirectly associated hardware is initially set to 0. S2. Calculate hardware importance: Based on the data fusion method, calculate the importance of each piece of hardware in the selected hardware set F under the current production task; S3. Hardware filtering: Remove the hardware with the lowest current importance from the selected hardware set F; S4. Judgment and Iteration: Determine whether the overall importance of the remaining hardware in the selected hardware set F after executing S3 has increased compared to before deletion; If so, return to execute S2, recalculate the importance of each hardware based on the updated selected hardware set F, and perform the next round of screening; If not, then cancel the deletion operation in S3 and add the hardware back to the selected hardware set F; S5. Output customized inspection plan: Output the final selected hardware set F as the best inspection hardware combination for the unmanned mining truck to perform the current production task, so as to guide customized inspection operations.

2. The method for customized inspection of unmanned mining trucks based on importance according to claim 1, characterized in that, In step S2, the calculation of hardware importance based on the data fusion method specifically includes: S21. Construct a mutual support matrix: based on the initial importance of each hardware component. (i=1,2,…,n), calculate any two hardware... and Mutual support between ,form cross support matrix ,in, ; S22. Calculate the support distance: based on the mutual support matrix. Calculate any two hardware and Support distance between The support distance This indicates the difference in consistency between the two hardware components in their support for production tasks. S23. Basic Trust Allocation: Determine the set of weight coefficients for each hardware component based on historical data. (p=1,2,…,n), and based on the mutual support and the weighting coefficients Calculation targeting hardware Basic trust allocation value ; S24. Final Importance Calculation: Based on the basic trust allocation value, data fusion is performed using evidence combination rules to obtain the final importance value of each hardware component.

3. The method for customized inspection of unmanned mining trucks based on importance according to claim 2, characterized in that, In S22, the support distance The calculation formula is: In the formula, , The mutual support matrix represents the following respectively. No. row and number Row vectors.

4. The method for customized inspection of unmanned mining trucks based on importance according to claim 2, characterized in that, In S23, the basic trust allocation value The calculation formula is: In the formula, Indicates from the first Based on the evidence provided by the hardware, it is believed that the first The importance assessment of each hardware component is as follows: The level of trust; , These represent the first [number] based on historical data. , Each hardware component has a pre-set weighting coefficient; , They represent the first The hardware pair The level of hardware support, the first The hardware pair The level of hardware support; This indicates the total number of hard components.

5. The importance-based customized inspection method for unmanned mining trucks according to claim 4, characterized in that, In step S24, the data fusion using evidence combination rules specifically includes: Calculate the trust level allocation value after fusion It satisfies the formula: Then through the formula Calculate the hardware ultimate importance .

6. The method for customized inspection of unmanned mining trucks based on importance according to claim 1, characterized in that, The determination of whether the overall importance of the remaining hardware in the selected hardware set F after executing S3 has increased compared to before deletion is specifically as follows: After performing the deletion operation, determine whether the sum of the importance values ​​of all remaining hardware in the selected hardware set F is greater than the sum of the importance values ​​of all hardware in the selected hardware set F before deletion.

7. The method for customized inspection of unmanned mining trucks based on importance according to claim 1, characterized in that, In step S4, after determining that the deletion operation is not successful and canceling the current deletion operation, the following steps are also included: Determine whether the selected hardware set F has reached a stable state. If it has, proceed directly to step S5.

Images