A health assessment and prediction method for the steering system of an autonomous mine car

Abstract

The present invention relates to the technical field of equipment health management, and in particular to a health assessment and prediction method for an autonomous driving mine car steering system, comprising: accumulating multiple groups of vehicle operating condition vectors and health index vectors during the historical operation of the autonomous driving mine car to form a historical vehicle driving data set; performing weighted clustering processing on the historical vehicle driving data set to obtain multiple clustering clusters and a clustering model; determining health index evaluation criteria for multiple clustering clusters based on the historical vehicle driving data set; analyzing current operating condition data and health index data based on the clustering model and the health index evaluation criteria of each clustering cluster to obtain a real-time comprehensive health score; accumulating the real-time comprehensive health score to obtain a historical health status evaluation result; performing nonlinear prediction based on the historical health status evaluation result to obtain a comprehensive health score for a future time period; the present invention can accurately assess the health status of the mine car.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment health management, and in particular to a health assessment and prediction method for a steering system of an autonomous driving mine car. Background Art

[0002] The transportation environment in open-pit mines is harsh and complex, with high safety risks and huge losses caused by accidents. Unmanned driving in mines can ensure the "intrinsic safety" of transportation. However, unmanned transportation equipment in mines is prone to failure in harsh environments, posing safety hazards. The steering system is responsible for controlling the direction of travel of the mine car and is one of the most critical systems of the mine car. The steering performance directly affects the safety and stability of the mine car. Therefore, the safety of the mine car steering system is particularly important. Vehicle failures are often not unforeseen and sudden, but are caused by the decline of the health status of various vehicle systems to a certain extent.

[0003] At present, the operation and maintenance mode of autonomous mining vehicles still adopts the post-maintenance method, and the operation and maintenance costs of mining areas remain high. The binary evaluation of good or bad cannot accurately grasp the health status of the vehicle, lacks the ability to warn of transportation vehicle failures, and cannot reasonably arrange maintenance plans before failures occur. In addition, in a highly dynamic operating environment, it is more difficult to accurately evaluate the true health status of the vehicle.

[0004] Therefore, a full life cycle health assessment and prediction method for the steering system of an autonomous mining car is needed, which can realize autonomous maintenance decisions of the equipment and ensure the safe and reliable operation of the vehicle. Summary of the invention

[0005] In view of the above problems, the present invention provides a health assessment and prediction method for the steering system of an autonomous driving mine car, which solves the technical problems of the shortcomings of the operation and maintenance mode of autonomous driving mine cars in the prior art and the difficulty in evaluating the health status in a dynamic driving environment.

[0006] The present invention provides a health assessment and prediction method for an autonomous driving mine car steering system, comprising the following steps:

[0007] Step S1, determining the health index of the steering system of the autonomous mine car, accumulating multiple sets of vehicle operating condition vectors and health index vectors during the historical operation of the autonomous mine car, and forming a historical vehicle driving data set;

[0008] Step S2, performing weighted clustering processing on the historical vehicle travel data set to obtain multiple clusters and a clustering model;

[0009] Step S3, determining health indicator evaluation criteria of the plurality of clusters based on the historical vehicle driving data set;

[0010] Step S4: Based on the clustering model and the health indicator evaluation criteria of each clustering cluster, the operating condition data and health indicator data of the current autonomous driving mine car are analyzed to obtain a real-time comprehensive health score;

[0011] Step S5: accumulate the real-time comprehensive health scores to obtain historical health status assessment results; perform nonlinear prediction based on the historical health status assessment results to obtain the comprehensive health score for a future time period.

[0012] Preferably, the vehicle operating condition vector and health index vector in step S1 are specifically:

[0013] The health index vector of the automatic mine car steering system is h = h1,...,h n , specifically including: steering response time, steering angle deviation, lateral deviation, heading deviation, etc., a total of n;

[0014] The operating condition vector that affects the performance of the steering system is c=c1,...,c m , specifically including: path curvature, vehicle speed, road slope, road transverse slope, vehicle load, etc., a total of m.

[0015] Preferably, step S2 specifically includes:

[0016] Step S2-1, calculate the correlation r between the health index vector h and the operating condition vector c ij , according to r ij Calculate the weight w of each operating condition i ; where i = 1, ..., m; j = 1, ..., n, r ij represents the i-th health indicator h i and the jth operating condition c j The correlation between i is the weight of the i-th operating condition;

[0017] Step S2-2: weighted clustering is performed on all operating condition vectors in the historical vehicle driving data set to obtain K clusters and the cluster center of each cluster, and a cluster model is obtained.

[0018] Preferably, step S2-2 specifically includes:

[0019] Step S2-2-1, pre-set the number of clusters K;

[0020] Step S2-2-2, taking all the operating condition vectors in the historical vehicle driving data set as the operating condition sample point set to be clustered, and randomly selecting an operating condition sample point for each preset cluster from the operating condition sample point set to be clustered as the initial value of the cluster center of each cluster;

[0021] Step S2-2-3: Calculate the weighted distance from the operating condition sample point to the cluster center, where the weight is the weight w of each operating condition. i , the operating condition sample point set is allocated to the closest cluster according to the minimum distance principle;

[0022] Step S2-2-4, using the mean of the operating condition sample points of the cluster to update the cluster center of each cluster;

[0023] Step S2-2-5, return to step S2-2-3, until the cluster center no longer changes, output the final K clusters and the cluster center of each cluster, and obtain the clustering model.

[0024] Preferably, step S3 specifically includes:

[0025] Based on the historical vehicle driving data set, the operating condition sample points and the corresponding health indicators in each cluster are obtained. For each cluster obtained by clustering, the health indicator evaluation criteria of each cluster are determined according to the following requirements:

[0026] (1) Arrange the health indicators from the best to the worst, with the top 50% of the health indicators having a health score of 100;

[0027] (2) The top 80% of the best health indicators, with a health score of 80 points;

[0028] (3) 0.3% of the people with the worst health indicators, with a critical health score of 60 points;

[0029] (4) health scores decreased linearly from 60 to 80 and from 80 to 100 points with respect to health indicators;

[0030] (5) When the health score is below 60 points, the health index decreases linearly, and the slope of the decline is twice that of 60 to 80, until it drops to 0 points.

[0031] Preferably, step S4 specifically includes:

[0032] Step S4-1, inputting the operating condition of the mine car in real time, and obtaining the cluster to which the current operating condition belongs according to the clustering model;

[0033] Step S4-2: Analyze and judge the health index evaluation criteria of the cluster to which the current operating condition belongs to obtain the scores of various health indicators of the current operating condition.

[0034] Step S4-3: Calculate the average value of the health scores of the various health evaluation indicators as the real-time comprehensive health score.

[0035] Preferably, step S5 specifically includes:

[0036] Step S5-1, accumulating the real-time comprehensive health score to obtain historical health status assessment results, and establishing a polynomial nonlinear regression model based on the historical health status assessment results for a period of time before the current moment;

[0037] Step S5-2: predicting the comprehensive health score in the next half hour based on the polynomial nonlinear regression model.

[0038] Preferably, it also includes:

[0039] Step S6: Maintaining the autonomous driving mine car based on the real-time comprehensive health score and the evaluation result maintenance strategy, specifically including:

[0040] Step S6-1, according to the real-time comprehensive health score, obtain the average comprehensive health score sequence value within 1 minute; determine the comprehensive health score corresponding to the health level: level 1 is 90-100 points, level 2 is 80-90 points, level 3 is 60-80 points, and level 4 is 0-60 points;

[0041] Step S6-2: Set a corresponding maintenance strategy according to the health level, and classify the 60 average comprehensive health score sequence values ​​within 1 hour.

[0042] (1) If more than 66% of the sequence values ​​in one hour are classified as Level 3 or more than 33% of the sequence values ​​are classified as Level 4, the strategy is to immediately stop the operation and conduct targeted maintenance based on the scores of specific health indicators;

[0043] (2) If the sequence value is more than 33% within 1 hour, it belongs to Level 3. The strategy is that the vehicle can continue to operate, but it needs to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators;

[0044] (3) Sequence values ​​above 33% within 1 hour belong to Level 2, and the strategy is normal vehicle transportation, with maintenance scheduled every half month;

[0045] (4) More than 66% of the sequence values ​​within 1 hour belong to level 1, and the strategy is normal operation, and no maintenance is required.

[0046] Preferably, it also includes:

[0047] Step S7: maintaining the autonomous mining vehicle based on the predicted maintenance strategy according to the comprehensive health score in the future time period, specifically including:

[0048] Classify the predicted comprehensive health scores for the next period of time:

[0049] (1) If the prediction result for more than 8 minutes within 10 minutes is that the health score drops to less than 60 points after half an hour, the strategy is to immediately stop the operation and conduct targeted maintenance based on the scores of specific health indicators;

[0050] (2) If the vehicle is delayed for more than 3 minutes within 10 minutes, the prediction result is that the health score will drop to less than 60 points after half an hour. The strategy is that the vehicle can continue to operate, but it needs to be inspected as soon as possible after the transportation is completed, and targeted inspections should be carried out according to the scores of specific health indicators;

[0051] (3) If the vehicle is delayed for more than 8 minutes within 10 minutes, the prediction result is that the health score will drop to less than 80 points after half an hour. The strategy is that the vehicle can continue to operate, but it needs to be inspected as soon as possible after the transportation is completed, and targeted inspections should be carried out according to the scores of specific health indicators;

[0052] (4) If the prediction result exceeds 8 minutes within 10 minutes, the health score will be less than 90 points after half an hour. The strategy is normal transportation of the vehicle and maintenance is set once every half a month.

[0053] (5) If the prediction result exceeds 2 minutes within 10 minutes, the health score will be > 90 points half an hour later. The strategy is normal operation and no maintenance is required.

[0054] Compared with the prior art, the present invention has at least the following beneficial effects:

[0055] (1) Based on determining the health evaluation index of the steering system and screening the operating conditions that affect the system performance, the present invention performs weighted clustering on the operating conditions of the autonomous driving mine car, thereby fully considering the impact of the operating conditions on the health status, making the evaluation result more accurate and comprehensive.

[0056] (2) Based on normal driving data and combined with expert experience, the present invention designs evaluation criteria for health indicators for each cluster obtained by clustering, thereby realizing real-time quantitative evaluation of the health status of the mine car steering system and improving the monitoring capability of the system.

[0057] (3) The present invention is based on the historical health status assessment results of the vehicle steering system and predicts the system health change trend through a nonlinear prediction algorithm, which can help arrange maintenance strategies in advance, reduce the occurrence of failures, and improve operation and maintenance efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The drawings are only for the purpose of illustrating particular embodiments and are not to be construed as limiting the invention.

[0059] Figure 1 The present invention is a flowchart of an embodiment of a method for health assessment and prediction of an autonomous driving mine car steering system disclosed in the present invention.

[0060] Figure 2The present invention discloses a flow chart for setting operating condition weights based on the correlation between operating conditions and evaluation indicators.

[0061] Figure 3 The present invention is a flowchart of weighted clustering of operating conditions disclosed in the present invention.

[0062] Figure 4 It is a schematic diagram of the health status assessment criteria disclosed in the present invention.

[0063] Figure 5 The present invention discloses a flow chart for implementing health status assessment of a steering system based on operating condition clustering.

[0064] Figure 6 It is a schematic diagram of the health status assessment score results of the steering system of the autonomous driving mine car disclosed in the present invention.

[0065] Figure 7 A flow chart of making maintenance decisions based on the health status assessment results disclosed in the present invention.

[0066] Figure 8 This is a schematic diagram of the health status trend prediction disclosed in the present invention.

[0067] Fig. 9 The present invention discloses a flow chart for making maintenance decisions based on health prediction results.

[0068] Fig.10 This is a schematic diagram of the health prediction results of the steering system of the autonomous driving mine car disclosed in the present invention. DETAILED DESCRIPTION

[0069] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0070] In order to illustrate the effectiveness of the method proposed by the present invention, the above technical solution of the present invention is described in detail below through a specific embodiment. Figure 1 As shown, a health assessment and prediction method for an autonomous driving mine car steering system is disclosed, comprising the following steps:

[0071] Step S1, determine the health index of the steering system of the autonomous driving mine car, accumulate multiple groups of vehicle operating condition vectors and health index vectors during the historical operation of the autonomous driving mine car, and form a historical vehicle driving data set.

[0072] The present invention combines expert experience and actual engineering needs to determine the health indicators of the automatic mine car steering system and screen the operating conditions that affect the steering system performance. Specifically, the steps of determining and screening the health indicators and operating conditions are as follows:

[0073] Determine the health index vector h=h1,...,h of the automatic mine car steering system n , specifically including: steering response time, steering angle deviation, lateral deviation, heading deviation, etc., a total of n.

[0074] Screening the operating condition vector c=c1,...,c that affects the performance of the steering system m , specifically including: path curvature, vehicle speed, road slope, road transverse slope, vehicle load, etc., a total of m.

[0075] In some embodiments, the self-driving mining vehicle is equipped with onboard sensors that can collect real-time vehicle operating condition data and health indicator data. These data can be transmitted to a centralized data storage system, and the data can be cleaned and sorted regularly to form a structured historical vehicle driving data set. The collection of these data can provide a basis for subsequent equipment health assessment and prediction.

[0076] Step S2: performing weighted clustering processing on the historical vehicle travel data set to obtain multiple clusters and a clustering model.

[0077] like Figure 2 As shown, the process of weighted clustering processing of the historical vehicle driving data set by the present invention is as follows:

[0078] Calculate the correlation r between the health index vector h and the operating condition vector c ij , i=1,...,m; j=1,...,n,r ij represents the i-th health indicator h i and the jth operating condition c j The correlation between them. The correlation is calculated using the Pearson coefficient, expressed as:

[0079]

[0080] Among them, X and Y are the historical data of a health indicator to be calculated and a certain operating condition, cov(X,Y) represents the covariance of X and Y, σ X ,σ Y is the variance of X and Y, μ X ,μ Y is the mean of X and Y, and E(·) represents the expected value.

[0081] Get the correlation r ij Afterwards, according to rij Calculate the weight w of each operating condition i :

[0082]

[0083] Among them, w i is the weight of the ith operating condition.

[0084] All operating condition vectors in the historical vehicle driving data set are weighted clustered. Specifically, Figure 3 As shown in Figure 2, the steps for weighted clustering of operating condition vectors are as follows:

[0085] (1) Predetermine the number of clusters K.

[0086] (2) All operating condition vectors in the historical vehicle driving data set are used as a set of operating condition sample points to be clustered, and an operating condition sample point is randomly selected for each preset cluster from the set of operating condition sample points to be clustered as the initial value of the cluster center of each cluster.

[0087] (3) Calculate the weighted distance from the operating condition sample point to the cluster center, where the weight is the weight w of each operating condition. i , the operating condition sample point set is allocated to the nearest cluster according to the minimum distance principle.

[0088] (4) Use the mean of the operating condition sample points of the cluster to update the cluster center of each cluster.

[0089] (5) Return to step (3) until the cluster center no longer changes, output the final K clusters and the cluster center of each cluster, and obtain the clustering model.

[0090] The clustering model can output the cluster cluster to which the corresponding operating condition vector belongs after inputting the operating condition vector of the mine car.

[0091] Step S3: Determine the health indicator evaluation criteria of each cluster based on the historical vehicle driving data set.

[0092] Specifically, based on the historical vehicle driving data set, the operating condition sample points and the corresponding health indicators in each cluster are obtained. The step of determining the health indicator evaluation criterion for each cluster obtained by clustering is as follows: Figure 4 As shown, including:

[0093] ① Arrange the health indicators from the best to the worst, with the top 50% of the health indicators having a health score of 100 points;

[0094] ② The top 80% with the best health indicators, with a health score of 80 points;

[0095] ③ The 0.3% with the worst health indicators, with a critical health score of 60 points;

[0096] ④ The health score decreases linearly from 60 to 80 and from 80 to 100 points with respect to health indicators;

[0097] ⑤ When the health score is below 60 points, the health index decreases linearly, and the downward slope is twice that of 60 to 80, until it drops to 0 points.

[0098] Finally, the health indicator evaluation criteria for each cluster are obtained.

[0099] Step S4: Based on the clustering model and the health indicator evaluation criteria of each cluster cluster, the operating condition data and health indicator data of the current autonomous driving mine car are analyzed to obtain a real-time comprehensive health score.

[0100] Specifically, Figure 5 As shown, the steps to obtain a real-time comprehensive health score are as follows:

[0101] (1) Inputting the operating conditions of the mine car in real time, and obtaining the cluster to which the current operating conditions belong according to the clustering model;

[0102] (2) Analyze and judge the health indicator evaluation criteria of the cluster to which the current operating condition belongs, and obtain the scores of various health indicators of the current operating condition.

[0103] (3) Calculate the average of the health scores of the various health evaluation indicators as the real-time comprehensive health score. Figure 6 As shown in the figure, the health status assessment result of an autonomous mining vehicle in a certain period of time is shown, including the quantitative score of each health indicator and the comprehensive health score.

[0104] Step S5: accumulate the real-time comprehensive health scores to obtain historical health status assessment results; perform nonlinear prediction based on the historical health status assessment results to obtain the comprehensive health score for a future time period.

[0105] Specifically, Figure 8 As shown, the specific steps of health change trend prediction in this step are as follows:

[0106] (1) A polynomial nonlinear regression model is established based on the real-time comprehensive health score 4 hours before the current moment.

[0107] (2) Based on the polynomial nonlinear regression model, predict the comprehensive health score in the next half hour.

[0108] The embodiment of the present invention further includes the step of maintaining the autonomous mine car according to the health assessment and prediction results of the steering system of the autonomous mine car, which is specifically as follows:

[0109] Step S6: Maintain the autonomous mining vehicle based on the real-time comprehensive health score and the maintenance strategy based on the evaluation result.

[0110] like Figure 7 As shown, the specific steps for developing an assessment result maintenance strategy are as follows:

[0111] (1) According to the real-time comprehensive health score, the average comprehensive health score sequence value within 1 minute is obtained; the comprehensive health score corresponding to the health level is determined: level 1 is 90-100, level 2 is 80-90, level 3 is 60-80, and level 4 is 0-60.

[0112] (2) According to the health level setting corresponding maintenance strategy, for 60 average comprehensive health score sequence values ​​within 1 hour:

[0113] ① Poor: More than 66% of the sequence values ​​within 1 hour belong to level 3 or more than 33% of the sequence values ​​belong to level 4. The strategy is to immediately stop the operation and conduct targeted maintenance according to the scores of specific health indicators;

[0114] ② Poor: More than 33% of the sequence values ​​within 1 hour belong to the third level. The strategy is that the vehicle can continue to operate, but it is necessary to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators;

[0115] ③Good: More than 33% of the sequence values ​​within 1 hour belong to Level 2, the strategy is normal vehicle transportation, and maintenance is set once every half a month;

[0116] ④Excellent: More than 66% of the sequence values ​​within 1 hour are classified as Level 1, the strategy is normal operation, and no maintenance is required.

[0117] In an embodiment of the present invention, Figure 6 As shown in the figure, at about 0:26, the vehicle has met the condition of level 3 for more than 33% of the time within 1 hour. The strategy is that the vehicle can continue to operate, but it needs to be inspected as soon as possible after completing this transportation; at about 1:54, the vehicle has met the condition of level 3 for more than 66% of the time within 1 hour. The strategy is to stop operation immediately and conduct targeted maintenance based on the scores of specific health indicators. However, since this method is not equipped in practice, the above strategy is not adopted in time, and a steering system failure occurs at about 3:00.

[0118] Step S7: maintaining the autonomous driving mine car based on the comprehensive health score in the future time period and the maintenance strategy based on the prediction result.

[0119] Specifically, Fig. 9 As shown in Figure 2, the specific steps for developing a prediction result maintenance strategy are as follows:

[0120] For the predicted comprehensive health score for the next half hour:

[0121] ① Poor: If the health score exceeds 8 minutes within 10 minutes, it is predicted that the health score will drop to less than 60 points in half an hour. The strategy is to immediately stop the operation and conduct targeted maintenance according to the scores of specific health indicators;

[0122] ② Poor: If the delay exceeds 3 minutes within 10 minutes, the health score is predicted to drop to less than 60 points in half an hour. The strategy is that the vehicle can continue to operate, but it is necessary to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators;

[0123] ③ Poor: If the delay exceeds 8 minutes within 10 minutes, the health score is predicted to drop to less than 80 points in half an hour. The strategy is that the vehicle can continue to operate, but it is necessary to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators;

[0124] ④Good: If the predicted health score is less than 90 points in half an hour for more than 8 minutes within 10 minutes, the vehicle is transported normally and maintenance is set once every half a month;

[0125] ⑤Excellent: The service life exceeds 2 minutes within 10 minutes, and the predicted health score is >90 points after half an hour. The strategy is normal operation and no maintenance is required.

[0126] like Fig.10 As shown, from about 6:10 to 6:24, the predicted health score dropped to <80 points in half an hour, which met the condition that the health score dropped to <80 points in half an hour for more than 8 minutes within 10 minutes. The strategy is that the vehicle can continue to operate, but it is necessary to arrange an inspection as soon as possible after completing this transportation, and conduct targeted inspections based on the scores of specific health indicators.

[0127] Although the specific embodiments of the present invention depict various actions or steps in a specific order, this should be understood as requiring such actions or steps to be performed in the specific order shown or in a sequential order, or requiring all illustrated actions or steps to be performed to obtain the desired results. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of a separate embodiment can also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation can also be implemented in multiple implementations individually or in any suitable sub-combination. The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

[0128] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A health assessment and prediction method for an autonomous mine car steering system, characterized in that: The following steps are involved: Step S1, determining the health index of the steering system of the autonomous mine car, accumulating multiple sets of vehicle operating condition vectors and health index vectors during the historical operation of the autonomous mine car, and forming a historical vehicle driving data set; Step S2, performing weighted clustering processing on the historical vehicle travel data set to obtain multiple clusters and a clustering model; Step S3, determining health indicator evaluation criteria of the plurality of clusters based on the historical vehicle driving data set; Step S4: Based on the clustering model and the health indicator evaluation criteria of each clustering cluster, the operating condition data and health indicator data of the current autonomous driving mine car are analyzed to obtain a real-time comprehensive health score; Step S5, accumulating the real-time comprehensive health score to obtain a historical health status assessment result; A nonlinear prediction is performed based on the historical health status assessment results to obtain a comprehensive health score for a future period of time.

2. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 1 is characterized in that: The vehicle operating condition vector and health index vector in step S1 are specifically: The health index vector of the automatic mine car steering system is h = h1,...,h n , specifically including: steering response time, steering angle deviation, lateral deviation, and heading deviation, a total of n; The operating condition vector that affects the performance of the steering system is c=c1,...,c m , specifically including: path curvature, vehicle speed, road slope, road transverse slope, and vehicle load, a total of m.

3. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 2 is characterized in that: Step S2 specifically includes: Step S2-1, calculate the correlation r between the health index vector h and the operating condition vector c ij , according to r ij Calculate the weight w of each operating condition i ; where i = 1, ..., m; j = 1, ..., n, r ij represents the i-th health indicator h i and the jth operating condition c j The correlation between i is the weight of the i-th operating condition; Step S2-2: weighted clustering is performed on all operating condition vectors in the historical vehicle driving data set to obtain K clusters and the cluster center of each cluster, and a cluster model is obtained.

4. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 3 is characterized in that: Step S2-2 specifically includes: Step S2-2-1, pre-set the number of clusters K; Step S2-2-2, taking all the operating condition vectors in the historical vehicle driving data set as the operating condition sample point set to be clustered, and randomly selecting an operating condition sample point for each preset cluster from the operating condition sample point set to be clustered as the initial value of the cluster center of each cluster; Step S2-2-3: Calculate the weighted distance from the operating condition sample point to the cluster center, where the weight is the weight w of each operating condition. i , the operating condition sample point set is allocated to the closest cluster according to the minimum distance principle; Step S2-2-4, using the mean of the operating condition sample points of the cluster to update the cluster center of each cluster; Step S2-2-5, return to step S2-2-3, until the cluster center no longer changes, output the final K clusters and the cluster center of each cluster, and obtain the clustering model.

5. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 4 is characterized in that: Step S3 specifically includes: Based on the historical vehicle driving data set, the operating condition sample points and the corresponding health indicators in each cluster are obtained. For each cluster obtained by clustering, the health indicator evaluation criteria of each cluster are determined according to the following requirements: (1) Arrange the health indicators from the best to the worst, with the top 50% of the health indicators having a health score of 100; (2) The top 80% of the best health indicators, with a health score of 80 points; (3) 0.3% of the people with the worst health indicators, with a critical health score of 60 points; (4) health scores decreased linearly from 60 to 80 and from 80 to 100 points with respect to health indicators; (5) When the health score is below 60 points, the health index decreases linearly, and the slope of the decline is twice that of 60 to 80, until it drops to 0 points.

6. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 5 is characterized in that: Step S4 specifically includes: Step S4-1, inputting the operating condition of the mine car in real time, and obtaining the cluster to which the current operating condition belongs according to the clustering model; Step S4-2: Analyze and judge the health index evaluation criteria of the cluster to which the current operating condition belongs to obtain the scores of various health indicators of the current operating condition. Step S4-3: Calculate the average value of the health scores of the various health evaluation indicators as the real-time comprehensive health score.

7. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 6, characterized in that: Step S5 specifically includes: Step S5-1, accumulating the real-time comprehensive health score to obtain historical health status assessment results, and establishing a polynomial nonlinear regression model based on the historical health status assessment results for a period of time before the current moment; Step S5-2: predicting the comprehensive health score in the next half hour based on the polynomial nonlinear regression model.

8. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 7 is characterized in that: Also includes: Step S6: Maintaining the autonomous driving mine car based on the real-time comprehensive health score and the maintenance strategy based on the evaluation result, specifically including: Step S6-1, according to the real-time comprehensive health score, obtain the average comprehensive health score sequence value within 1 minute; determine the comprehensive health score corresponding to the health level: level 1 is 90-100 points, level 2 is 80-90 points, level 3 is 60-80 points, and level 4 is 0-60 points; Step S6-2: Set the corresponding maintenance strategy according to the health level, and classify the 60 average comprehensive health score sequence values ​​within 1 hour: (1) If more than 66% of the sequence values ​​in one hour are classified as Level 3 or more than 33% of the sequence values ​​are classified as Level 4, the strategy is to immediately stop the operation and conduct targeted maintenance based on the scores of specific health indicators; (2) Sequence values ​​above 33% within 1 hour belong to Level 3. The strategy is that the vehicle continues to operate, but it is necessary to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators; (3) Sequence values ​​above 33% within 1 hour belong to Level 2, and the strategy is normal vehicle transportation, with maintenance scheduled every half month; (4) More than 66% of the sequence values ​​within 1 hour belong to level 1, and the strategy is normal operation, and no maintenance is required.

9. The health assessment and prediction method of the steering system of an autonomous driving mine car according to claim 8, characterized in that: Also includes: Step S7: maintaining the autonomous driving mine car based on the comprehensive health score in the future time period and the maintenance strategy based on the prediction result, specifically including: The predicted comprehensive health scores for the next period of time are classified: (1) If the prediction result for more than 8 minutes within 10 minutes is that the health score drops to less than 60 points after half an hour, the strategy is to immediately stop the operation and conduct targeted maintenance based on the scores of specific health indicators; (2) If the vehicle is delayed for more than 3 minutes within 10 minutes, the prediction result is that the health score will drop to less than 60 points after half an hour. The strategy is to continue the operation of the vehicle, but it is necessary to arrange a spot check as soon as possible after completing the transportation, and conduct targeted inspections based on the scores of specific health indicators; (3) If the vehicle is delayed for more than 8 minutes within 10 minutes, the prediction result is that the health score will drop to less than 80 points after half an hour. The strategy is to continue the operation of the vehicle, but it is necessary to arrange a spot check as soon as possible after the completion of this transportation, and conduct targeted inspections based on the scores of specific health indicators; (4) If the prediction result exceeds 8 minutes within 10 minutes, the health score will be less than 90 points after half an hour. The strategy is normal transportation of the vehicle and maintenance is set once every half a month. (5) If the prediction result exceeds 2 minutes within 10 minutes, the health score will be > 90 points half an hour later. The strategy is normal operation and no maintenance is required.

Images