CN116974265B - Underground mine car fault diagnosis method and system under no-signal scene - Google Patents
Underground mine car fault diagnosis method and system under no-signal scene Download PDFInfo
- Publication number
- CN116974265B CN116974265B CN202310846556.1A CN202310846556A CN116974265B CN 116974265 B CN116974265 B CN 116974265B CN 202310846556 A CN202310846556 A CN 202310846556A CN 116974265 B CN116974265 B CN 116974265B
- Authority
- CN
- China
- Prior art keywords
- fault
- mine car
- dimensional code
- underground mine
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012544 monitoring process Methods 0.000 claims abstract description 52
- 238000007637 random forest analysis Methods 0.000 claims abstract description 29
- 239000013307 optical fiber Substances 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims description 14
- 238000005065 mining Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 2
- 238000012360 testing method Methods 0.000 description 6
- 238000003066 decision tree Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 206010022562 Intermittent claudication Diseases 0.000 description 1
- 208000002193 Pain Diseases 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 208000024980 claudication Diseases 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007115 recruitment Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000011343 solid material Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Vehicle Cleaning, Maintenance, Repair, Refitting, And Outriggers (AREA)
Abstract
The invention discloses a fault diagnosis method and system for an underground mine car under a no-signal scene, wherein the method comprises the following steps: acquiring monitoring data of an underground mine car; inputting the monitoring data into an optimized random forest model for fault diagnosis, and outputting fault grade and fault positioning information; inquiring a corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information, and displaying the two-dimensional code through the underground mine car; capturing a two-dimensional code image displayed on an underground mine car, and transmitting the two-dimensional code image to an underground monitoring platform through an optical fiber; analyzing the two-dimensional code image to obtain the running state of the underground mine car.
Description
Technical Field
The invention belongs to the technical field of automatic driving, and particularly relates to a fault diagnosis method for an underground mine car, in particular to a fault diagnosis method and system for the underground mine car under a no-signal scene.
Background
Mineral industry is a solid material guarantee and important support for economic development in China: mining economics contribute more than 30% of the specific gravity of the national GDP. Because of the pain points of difficulty in recruitment, high cost ratio of transportation, safety accident risk in transportation links and the like, unmanned transportation gradually becomes the rigid need of mining enterprises. The use of the intelligent mine car promotes the production efficiency of mining areas, but potential safety hazards exist at the same time, so that fault diagnosis is an important link for guaranteeing the safe operation of the intelligent mine car.
Along with the gradual popularization of intelligent vehicles, in order to ensure the safe operation of the intelligent vehicles, the fault diagnosis of the intelligent vehicles is rapidly developed, the operation state of the vehicles is monitored at any time based on the fault diagnosis of a mode of combining an artificial intelligent method with historical fault data of the vehicles, and whether the vehicles are in normal operation is judged, so that the fault diagnosis is rapidly and correctly carried out, and reasonable suggestions are given. The fault diagnosis method of the intelligent vehicle has been developed from the traditional simple experience to the mode of combining the artificial intelligence method with the artificial experience, so that the fault diagnosis is more informative and intelligent.
In the prior art, a study result rich for diagnosing a failure of a local system of a vehicle or a non-autonomous vehicle has been studied. However, for the special working condition of the underground mining area, the following defects exist: firstly, most underground mining area scenes exist in the condition of no network signal or difficult network signal laying, and the traditional network is not feasible for transmitting information, so that underground personnel and underground information in the scenes cannot interact in time; secondly, the intelligent mine car of operation adopts the autopilot mode to work, lacks personnel to monitor the running state of vehicle, and the abominable operational environment in pit leads to the vehicle to break down easily, in case intelligent mine car breaks down, can cause great incident. So far, few documents and researches are related to intelligent vehicle real-time fault diagnosis for underground mining area no-signal scenes.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a fault diagnosis method and system for an underground mine car under a no-signal scene, which are used for solving at least one technical problem.
According to one aspect of the specification, the fault diagnosis method for the underground mine car in the no-signal scene comprises the following steps:
Acquiring monitoring data of an underground mine car;
inputting the monitoring data into an optimized random forest model for fault diagnosis, and outputting fault grade and fault positioning information;
inquiring a corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information, and displaying the two-dimensional code through the underground mine car;
capturing a two-dimensional code image displayed on an underground mine car, and transmitting the two-dimensional code image to an underground monitoring platform through an optical fiber;
analyzing the two-dimensional code image to obtain the running state of the underground mine car.
As a further technical solution, the method further includes:
establishing a fault library of the underground mine car, wherein the fault library comprises a fault name, a monitoring variable, a variable value, a fault grade and fault positioning;
And establishing a corresponding relation among the fault grade, the fault positioning information and the two-dimensional code format in the fault library to form a fault lookup table.
According to the technical scheme, aiming at the situation that most underground mining area scenes are remote in geographic position and far away from a signal base station, the underground environment is not suitable for laying a network, vehicle state data are acquired through a vehicle-mounted sensor and the like, a data set is acquired through data processing operation, a training model predicts and classifies the acquired data, and whether the vehicle is faulty or not and fault positioning information are judged, so that real-time fault diagnosis of an intelligent mine car is achieved, and the intelligent mine car adopts corresponding fault corresponding measures according to diagnosis results.
Further, the corresponding two-dimensional code is found through information such as fault diagnosis results and fault positioning information in a table look-up mode, the two-dimensional code on the electronic display of the intelligent mine car is captured by the optical fiber cameras on the two walls of the mining area and transmitted to the on-mine monitoring platform, and the platform analyzes the two-dimensional code to obtain diagnosis data, so that the operation state of the intelligent mine car is monitored remotely on the mine.
As a further technical solution, the establishing of the fault library further includes:
Acquiring historical monitoring data of an underground mine car;
Screening the stored monitoring data according to the monitoring variable;
Correspondingly storing the screened monitoring data according to the names of the monitoring variables to form a plurality of characteristic classes;
And introducing specific faults aiming at each feature class, uniformly storing fault data and normal data, and adding labels of normal, primary fault, secondary fault, tertiary fault or quaternary fault to the stored data to form a fault library.
Specifically, the fault library comprises fault names, monitoring variables, variable values, fault grading, fault positioning and the like. Aiming at the severe environment of the operation of the underground mine car, signals (such as positioning sensing, planning control, a car body system and the like) which are easy to make mistakes are selected as monitoring variables for data acquisition.
According to the influence of faults on the mine car, the faults are classified into the following four grades: the first-level fault relates to the safety of vehicle running, needs emergency stop and cannot be started; the secondary fault is needed to be maintained immediately, but the mine car can claudication; three-level faults, which do not affect the task, can be overhauled in the next routine maintenance; four-level faults, which have an impact only for a part of the users. In order to better prompt fault information of monitoring personnel, a corresponding display lamp is arranged on the on-well monitoring platform to alarm.
Further, in order to collect specific fault data, the fault is introduced at a specific time and the fault grade is marked, so that the fault data is formed.
As a further technical solution, the optimizing of the random forest model further includes:
selecting model parameters to be optimized, and constructing an objective function and constraint conditions according to the model parameters to be optimized;
Solving the objective function under a given constraint condition to obtain an optimal parameter solution;
Constructing an optimized random forest model according to the optimal parameter solution;
and training the optimized random forest model by utilizing the data set in the fault library to obtain a trained optimized random forest model.
As a further technical solution, the model parameters to be optimized include: the number of features, the number of trees, the number of nodes in the tree, and the depth of the tree.
As a further technical scheme, the objective function is min S (f, k, d, p), and the constraint condition is that
Where S (f, k, d, p) =a×n×log (n) ×f×k+b×d×k+c×p×k, a, b, c denote coefficients between model training time complexity, prediction time complexity, and run time complexity, n denotes a number of training samples, f denotes a number of features, k denotes a number of trees, d denotes a depth of a tree, p denotes a node in a tree, te1 denotes a lower limit value of prediction accuracy, and Te2 denotes a lower limit value of prediction accuracy.
As a further technical solution, the method further includes: and capturing a two-dimensional code image displayed on the underground mine car by using the underground mine wall camera.
As a further technical solution, the method further includes: and transmitting the monitoring data of the underground mine car to the vehicle-mounted controller in a CAN line or ROS topic subscription and release mode, wherein the vehicle-mounted controller processes the monitoring data, outputs a diagnosis result and displays the diagnosis result on the vehicle-mounted display.
According to an aspect of the present disclosure, there is provided a fault diagnosis system for a downhole mine car in a no-signal scenario, including:
The acquisition module is used for acquiring monitoring data of the underground mine car;
The fault diagnosis module is used for inputting the monitoring data into the optimized random forest model for fault diagnosis and outputting fault grade and fault positioning information;
The fault inquiry display module is used for inquiring the corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information and displaying the two-dimensional code through the underground mine car;
The image capturing and transmitting module is used for capturing a two-dimensional code image displayed on the underground mine car and transmitting the two-dimensional code image to the underground monitoring platform through optical fibers;
And the analysis module is used for analyzing the two-dimensional code image and acquiring the running state of the underground mine car.
Compared with the prior art, the invention has the beneficial effects that:
(1) Aiming at the problem of fault diagnosis of the intelligent mine car without a signal scene in the pit, the invention adopts a method of a multi-classification model-random forest model. Firstly, selecting parameters of a model, solving optimal parameters of the model under constraint by constructing constraint conditions, and reducing the time of model prediction diagnosis on the premise of ensuring the prediction effect; secondly, a fault lookup table is established, a one-to-one correspondence relation between model fault diagnosis results and the like and the two-dimensional codes is established, and the lookup operation is a simpler mode and consumes less time; third, the optical fiber is adopted to transmit information, the operation speed of the optical fiber reaches 2.5GB per second, and the transmission time to the uphole monitoring platform can be greatly reduced under the condition that the transmission information is ensured.
(2) Aiming at the problem that a conventional network remote monitoring means is not enabled to timely monitor the running state of an underground intelligent mine car due to no network signal in an underground mine area, the invention provides a method for establishing a one-to-one correspondence between a model diagnosis result, fault positioning information and the like and a two-dimension code format, establishing a fault lookup table (model fault diagnosis result, two-dimension code), reading the fault diagnosis result of the intelligent mine car through a two-dimension code on an electronic display on a mine wall camera reading car, and transmitting the fault diagnosis result to an underground monitoring platform through an optical fiber form to realize real-time monitoring of the running state of the underground intelligent mine car under a no-signal scene.
Drawings
FIG. 1 is a schematic diagram of a model training optimization and application flow according to an embodiment of the present invention.
FIG. 2 is a diagram of a partial fault query representation of an embodiment of the present invention.
Fig. 3 is a schematic diagram of a QR Code two-dimensional Code format according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made more apparent and fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
A fault diagnosis method of an underground mine car aiming at an underground no-signal scene mainly comprises the following steps: information such as a sensing and positioning module, a planning control module, a car body system and the like is acquired and transmitted to a car-mounted controller arranged on an intelligent mine car in a CAN line or ROS topic subscription and release mode, cleaning and dimension reduction operation is carried out through data, and the data are stored to form a fault base; and (3) carrying out data training prediction by adopting a multi-classification random forest model, constructing constraint conditions of the model, optimizing parameter selection of the model under the constraint conditions, calling the optimization model to carry out real-time fault diagnosis, acquiring corresponding two-dimensional codes by searching a fault query table according to model fault diagnosis results and the like, capturing two-dimensional code format images by utilizing cameras on a mine wall, transmitting the two-dimensional code format images to an on-well monitoring platform through optical fibers, and analyzing the two-dimensional codes by the platform to acquire diagnosis data so as to realize reliable monitoring on the running state of the underground mine car.
Specifically, the fault diagnosis method for the underground mine car under the no-signal scene comprises the following steps:
Acquiring monitoring data of an underground mine car;
inputting the monitoring data into an optimized random forest model for fault diagnosis, and outputting fault grade and fault positioning information;
inquiring a corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information, and displaying the two-dimensional code through the underground mine car;
capturing a two-dimensional code image displayed on an underground mine car, and transmitting the two-dimensional code image to an underground monitoring platform through an optical fiber;
analyzing the two-dimensional code image to obtain the running state of the underground mine car.
Optionally, in the step of data acquisition and transmission to the vehicle-mounted controller, the sensing and positioning module mainly acquires data such as four radars respectively arranged at the front and rear ends of the vehicle, an IMU installed on the underground mine car, and a camera installed in front of the mine car; the planning control module mainly acquires data such as heading angle, speed, acceleration and path generation of the vehicle, and the acquired data is transmitted to a vehicle-mounted controller (in the example, MDC 600F) on the intelligent mine car in a manner of subscribing and publishing ROS topics; the vehicle body system mainly acquires ECU data information in a driving system, a braking system, a steering system and the like, and the acquired information is transmitted to the controller through a CAN line.
Optionally, in the data processing step, the collected data is first subjected to a conventional data cleansing. Before processing, a normal threshold value of the monitored quantity is required to be obtained, the data abnormality is judged according to the threshold value, then the sklearn library in python is called to process the data in the fault library, and the data is cleaned to ensure the quality of the data in the fault library. In order to better store the data in the vehicle-mounted controller to generate a fault library, performing PCA dimension reduction operation on the cleaned data, reducing the storage space on the premise of keeping the characteristic information as much as possible, standardizing and normalizing the data, dividing the data into a training set and a testing set according to a certain proportion by utilizing train_test_split, wherein 90% of the data set can be used as the training set, and the rest 10% can be used as the testing set. (for monitoring quantity, firstly, a normal value of the monitoring value is obtained by a supplier or a product instruction book, then, the normal threshold value of the monitoring value is determined on a rack (the normal value fluctuates up and down or has reasonable deviation), the data is cleaned according to the data label, and the misclassified data is removed)
Optionally, in the step of generating the fault library, the fault library is established according to the data processed in the foregoing, and the corresponding storage is performed according to the name of the monitored variable to form a feature class, so that the variable in the running state of the whole vehicle is monitored, the fault data and the data in the normal running state are uniformly stored, and the fault class labels of "normal", "primary fault", "secondary fault", "tertiary fault", "quaternary fault" and fault location (mainly divided into location sensing, planning, communication and specific executing mechanisms (steering wheel, brake, accelerator and the like)) are added to the stored data to form the fault library. The fault locating information can help maintenance personnel to quickly locate a fault module and find a fault reason so as to quickly find out a problem.
Optionally, in the model selection step, since the present invention is a multi-classification problem, a random forest model in machine learning is selected. The Random Forest (RF) is an integrated algorithm, and the decision tree is integrated to improve the classification accuracy, so that the method has the advantages of good robustness, high prediction accuracy and difficult occurrence of the phenomenon of overfitting. The randomness of the random forest model is mainly reflected in two aspects, the first aspect is random selection of data, the random forest model generally adopts a bootstrap (resampling) method to repeatedly and randomly extract samples from original data (the number of the samples is consistent with that of the original data set) in a put-back way to generate a sub-data set, so that elements in the same sub-data set can be repeated, and elements in different data sets can also be repeated; and secondly, randomly selecting features, wherein the random forest forms a random forest model by constructing a sub-decision tree, all features are not used in each splitting process (namely, each branch node) of the sub-decision tree, a certain number of features are randomly selected from all the features, and then the optimal features are selected from the randomly selected features. The random forest model is divided into a classification tree-based random forest model and a regression tree-based random forest model, wherein the classification decision tree-based random forest model is mainly adopted, monitoring data is input, classification labels are output, and whether the vehicle is faulty or not is judged according to the labels, so that the aim of fault diagnosis is achieved.
Optionally, in the model parameter optimization step, constraint conditions of the model are built according to the prediction effect and timeliness, and in the model training process, model parameters have great influence on model prediction precision and diagnosis instantaneity. The real-time performance of the random forest model is mainly represented by the following:
Training time complexity: o (n) log (n) f k)
Prediction time complexity: o (d.times.k)
Runtime complexity: o (p.times.k)
The real-time performance of the whole model is set as follows: s (f, k, d, p) =a×n×log (n) ×f×k+b×d×k+c×p×k
Wherein n=training sample number is a coefficient among the three, and is obtained through multiple experimental calibration, f=feature number, k=number of trees, p=number of nodes in the tree, and d=depth of the tree is a model parameter, and the method belongs to a range of parameter optimization.
The prediction effect of the random forest model is mainly represented by the following:
prediction accuracy:
Prediction accuracy:
wherein true example (TP): the model predicts the samples of the True type in the test samples as the number of samples of True
False negative example (FN): the model predicts the True type samples in the test samples as the False number of samples
False Positive (FP): the model predicts the False type samples in the test samples as True sample numbers
True negative example (TN): the model predicts the False type samples in the test samples as the False number of samples
And (5) constructing constraint conditions by the model, and obtaining optimal parameters under the constraint conditions.
Te1 and Te2 are lower limits of values (if the precision is required to be 0.95, the lower limit can be 0.95), namely, the real-time performance is minimum on the premise of guaranteeing the prediction effect, and the optimal parameters of the model are solved.
Optionally, in the steps of storing the model and applying the model, constructing a fault diagnosis model under the parameter by the optimal parameter under the solving constraint condition, performing data processing on the acquired latest real-time data, inputting the latest real-time data into the constructed fault diagnosis model, and rapidly outputting the diagnosis result by the model, thereby judging whether the data is normal or not according to the result, and achieving the purpose of real-time online fault diagnosis.
Optionally, in the step of searching the fault lookup table, the fault lookup table (fault level, fault positioning information, two-dimensional Code) is established, as shown in fig. 2, the fault lookup table is searched according to the foregoing real-time fault diagnosis result, the QR Code two-dimensional Code format under the diagnosis result is found, the two-dimensional Code is displayed by the vehicle-mounted electronic display, and the adoption of red can be beneficial to capturing the two-dimensional Code by the downhole camera, as shown in fig. 3.
Optionally, in the step of uploading data, a camera on the mine wall is used for capturing a two-dimensional code on an on-vehicle display, and transmitting the two-dimensional code to an on-well monitoring platform through an optical fiber, and the platform is used for analyzing the two-dimensional code to obtain diagnostic data, so that the running state of the intelligent mine car at the moment is obtained, and the purpose of real-time monitoring is realized.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced with equivalents; these modifications or substitutions do not depart from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.
Claims (8)
1. The fault diagnosis method for the underground mine car under the signal-free scene is characterized by comprising the following steps of:
Acquiring monitoring data of an underground mine car;
inputting the monitoring data into an optimized random forest model for fault diagnosis, and outputting fault grade and fault positioning information;
Inquiring a corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information, and displaying the two-dimensional code through the underground mine car; the presetting of the fault lookup table comprises the following steps: establishing a fault library of the underground mine car, wherein the fault library comprises a fault name, a monitoring variable, a variable value, a fault grade and fault positioning; establishing a corresponding relation among the fault level, the fault positioning information and the two-dimensional code format in the fault library to form a fault lookup table;
Capturing two-dimensional code images displayed on an underground mine car by using optical fiber cameras on two walls of a mining area, and transmitting the two-dimensional code images to an uphole monitoring platform through optical fibers;
analyzing the two-dimensional code image to obtain the running state of the underground mine car.
2. A method for fault diagnosis of a downhole mine car in a no signal scenario according to claim 1, wherein the establishing of the fault library further comprises:
Acquiring historical monitoring data of an underground mine car;
Screening the stored monitoring data according to the monitoring variable;
Correspondingly storing the screened monitoring data according to the names of the monitoring variables to form a plurality of characteristic classes;
And introducing specific faults aiming at each feature class, uniformly storing fault data and normal data, and adding labels of normal, primary fault, secondary fault, tertiary fault or quaternary fault to the stored data to form a fault library.
3. A method of diagnosing a fault in a downhole mine car in a no signal scenario as in claim 2, wherein the optimization of the random forest model further comprises:
selecting model parameters to be optimized, and constructing an objective function and constraint conditions according to the model parameters to be optimized;
Solving the objective function under a given constraint condition to obtain an optimal parameter solution;
Constructing an optimized random forest model according to the optimal parameter solution;
and training the optimized random forest model by utilizing the data set in the fault library to obtain a trained optimized random forest model.
4. A method of diagnosing a fault in a downhole mine car in a signalless scenario according to claim 3, wherein the model parameters to be optimized comprise: the number of features, the number of trees, the number of nodes in the tree, and the depth of the tree.
5. A method for fault diagnosis of a downhole mine car in a no-signal scenario according to claim 3, wherein the objective function is min S (f, k, d, p), and the constraint condition isWhere S (f, k, d, p) =a×n×log (n) ×f×k+b×d×k+c×p×ka, b, c denote coefficients between model training time complexity, prediction time complexity, and run time space complexity, n denotes a number of training samples, f denotes a number of feature, k denotes a number of trees, d denotes a depth of a tree, p denotes a node in a tree, te1 denotes a lower limit value of prediction accuracy, and Te2 denotes a lower limit value of prediction accuracy.
6. A method of diagnosing a fault in a downhole mine car in a no signal scenario as in claim 1, further comprising: and capturing a two-dimensional code image displayed on the underground mine car by using the underground mine wall camera.
7. A method of diagnosing a fault in a downhole mine car in a no signal scenario as in claim 1, further comprising: and transmitting the monitoring data of the underground mine car to the vehicle-mounted controller in a CAN line or ROS topic subscription and release mode, wherein the vehicle-mounted controller processes the monitoring data, outputs a diagnosis result and displays the diagnosis result on the vehicle-mounted display.
8. The utility model provides a down-hole mine car fault diagnosis system under no signal scene which characterized in that includes:
The acquisition module is used for acquiring monitoring data of the underground mine car;
The fault diagnosis module is used for inputting the monitoring data into the optimized random forest model for fault diagnosis and outputting fault grade and fault positioning information;
The fault inquiry display module is used for inquiring the corresponding two-dimensional code in a preset fault inquiry table according to the fault grade and the fault positioning information and displaying the two-dimensional code through the underground mine car; the presetting of the fault lookup table comprises the following steps: establishing a fault library of the underground mine car, wherein the fault library comprises a fault name, a monitoring variable, a variable value, a fault grade and fault positioning; establishing a corresponding relation among the fault level, the fault positioning information and the two-dimensional code format in the fault library to form a fault lookup table;
The image capturing and transmitting module is used for capturing two-dimensional code images displayed on the underground mine car by utilizing optical fiber cameras on two walls of a mining area and transmitting the two-dimensional code images to an underground monitoring platform through optical fibers;
And the analysis module is used for analyzing the two-dimensional code image and acquiring the running state of the underground mine car.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310846556.1A CN116974265B (en) | 2023-07-11 | 2023-07-11 | Underground mine car fault diagnosis method and system under no-signal scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310846556.1A CN116974265B (en) | 2023-07-11 | 2023-07-11 | Underground mine car fault diagnosis method and system under no-signal scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116974265A CN116974265A (en) | 2023-10-31 |
| CN116974265B true CN116974265B (en) | 2024-04-26 |
Family
ID=88472236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310846556.1A Active CN116974265B (en) | 2023-07-11 | 2023-07-11 | Underground mine car fault diagnosis method and system under no-signal scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116974265B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010094220A1 (en) * | 2009-02-18 | 2010-08-26 | 华为终端有限公司 | Method and device for acquiring two-dimensional code from video signal |
| CN102841581A (en) * | 2012-08-07 | 2012-12-26 | 邯郸供电公司 | Remote monitoring and scheduling system for unattended factory station microcomputer-type equipment |
| CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
| CN207366979U (en) * | 2017-11-17 | 2018-05-15 | 烟台维度机器人有限公司 | A kind of gold mine well pittop scheduling system of view-based access control model |
| CN108958215A (en) * | 2018-06-01 | 2018-12-07 | 天泽信息产业股份有限公司 | A kind of engineering truck failure prediction system and its prediction technique based on data mining |
| CN111161445A (en) * | 2020-01-09 | 2020-05-15 | 安徽乾坤电力工程有限公司 | Electric power inspection system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108388860B (en) * | 2018-02-12 | 2020-04-28 | 大连理工大学 | Aero-engine rolling bearing fault diagnosis method based on power entropy spectrum-random forest |
| US20220277045A1 (en) * | 2021-02-26 | 2022-09-01 | Halliburton Energy Services, Inc. | Diagnostic Trouble Code Signature Classification For Downhole Tool Fault Identification |
-
2023
- 2023-07-11 CN CN202310846556.1A patent/CN116974265B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010094220A1 (en) * | 2009-02-18 | 2010-08-26 | 华为终端有限公司 | Method and device for acquiring two-dimensional code from video signal |
| CN102841581A (en) * | 2012-08-07 | 2012-12-26 | 邯郸供电公司 | Remote monitoring and scheduling system for unattended factory station microcomputer-type equipment |
| CN105045256A (en) * | 2015-07-08 | 2015-11-11 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on data comparative analysis |
| CN207366979U (en) * | 2017-11-17 | 2018-05-15 | 烟台维度机器人有限公司 | A kind of gold mine well pittop scheduling system of view-based access control model |
| CN108958215A (en) * | 2018-06-01 | 2018-12-07 | 天泽信息产业股份有限公司 | A kind of engineering truck failure prediction system and its prediction technique based on data mining |
| CN111161445A (en) * | 2020-01-09 | 2020-05-15 | 安徽乾坤电力工程有限公司 | Electric power inspection system |
Non-Patent Citations (1)
| Title |
|---|
| 五沟矿无极绳绞车无线视频监控系统设计与实现;夏后强;包敬华;;能源技术与管理;20161028(05);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116974265A (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gu et al. | A novel lane-changing decision model for autonomous vehicles based on deep autoencoder network and XGBoost | |
| JP6003810B2 (en) | Vehicle reference value generator | |
| CN111414477A (en) | Vehicle fault automatic diagnosis method, device and equipment | |
| CN102765643A (en) | Elevator fault diagnosis and early-warning method based on data drive | |
| EP3865963A1 (en) | Method and device for analyzing vehicle failure | |
| JP2018073363A (en) | Vehicle operation data acquisition apparatus, vehicle operation data acquisition system, and vehicle operation data acquisition method | |
| CN115659263B (en) | Ship control behavior risk assessment system and method based on big data | |
| CN110782670A (en) | Scene restoration method based on data fusion, vehicle cloud platform and storage medium | |
| CN111898502A (en) | Dangerous goods vehicle identification method and device, computer storage medium and electronic equipment | |
| CN116257565A (en) | Ship abnormal behavior detection method | |
| CN115291515A (en) | Automatic driving simulation test system and evaluation method based on digital twinning | |
| CN115865483A (en) | Abnormal behavior analysis method and device based on machine learning | |
| CN114408694B (en) | Elevator fault prediction system and prediction method thereof | |
| Elspas et al. | Leveraging regular expressions for flexible scenario detection in recorded driving data | |
| CN117523499B (en) | Forest fire prevention monitoring method and system based on Beidou positioning and sensing | |
| CN115035328A (en) | Converter image increment automatic machine learning system and establishment training method thereof | |
| CN116974265B (en) | Underground mine car fault diagnosis method and system under no-signal scene | |
| CN113807404A (en) | Intelligent monitoring and emergency system of intelligent traffic network for multi-source information fusion | |
| CA3189344A1 (en) | Explaining machine learning output in industrial applications | |
| CN112150443A (en) | Train-mounted air conditioner residual life prediction method based on air quality data map | |
| KR102105007B1 (en) | Edge-cloud system for collecting and providing data of connected car | |
| CN115083229B (en) | Intelligent recognition and warning system of flight training equipment based on AI visual recognition | |
| CN115758561A (en) | Method for generating flight simulation parameter data of airplane | |
| US11912289B2 (en) | Method and device for checking an AI-based information processing system used in the partially automated or fully automated control of a vehicle | |
| CN114861762A (en) | Information processing method and device, storage medium and vehicle-mounted electronic controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |