CN116186910A - Method for establishing drilling tool wear prediction model and drilling tool wear prediction system - Google Patents

Method for establishing drilling tool wear prediction model and drilling tool wear prediction system Download PDF

Info

Publication number
CN116186910A
CN116186910A CN202211533649.0A CN202211533649A CN116186910A CN 116186910 A CN116186910 A CN 116186910A CN 202211533649 A CN202211533649 A CN 202211533649A CN 116186910 A CN116186910 A CN 116186910A
Authority
CN
China
Prior art keywords
drilling
data
drilling tool
tool
wear
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.)
Pending
Application number
CN202211533649.0A
Other languages
Chinese (zh)
Inventor
刘飞香
廖金军
于钟博
蒋海华
龚敏
胡及雨
吴士兰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Railway Construction Heavy Industry Group Co Ltd
Original Assignee
China Railway Construction Heavy Industry Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Railway Construction Heavy Industry Group Co Ltd filed Critical China Railway Construction Heavy Industry Group Co Ltd
Priority to CN202211533649.0A priority Critical patent/CN116186910A/en
Publication of CN116186910A publication Critical patent/CN116186910A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Abstract

The invention discloses a method for establishing a drilling tool abrasion prediction model and a drilling tool abrasion prediction system, wherein the method for establishing the drilling tool abrasion prediction model is used for obtaining the drilling tool abrasion prediction model by training based on deep learning by taking drilling parameters as input and drilling health and abrasion loss as output through establishing a mapping relation between drilling parameters and drilling tool abrasion loss, can solve the problems that the working condition is complex and the drilling tool abrasion situation is difficult to measure in the prior art, can obtain the drilling tool health by collecting the drilling parameters during subsequent application, does not need to arrange parts such as sensors and the like, and has good practicability; according to the drilling tool wear prediction system, the drilling tool wear prediction model of the wear prediction unit is input by only collecting the while-drilling parameters, so that the drilling wear and the health degree can be displayed on the display unit, and the monitoring and the control of the drilling tool wear condition are realized.

Description

Method for establishing drilling tool wear prediction model and drilling tool wear prediction system
Technical Field
The invention relates to the technical field of rock drilling construction, in particular to a method for establishing a drilling tool abrasion prediction model and a drilling tool abrasion prediction system.
Background
In the engineering using the drilling and blasting method construction as the main tunneling mode, the digitalized and intelligent rock drilling trolley replaces the traditional manual rough operation mode, and can achieve the aims of more accuracy, higher efficiency and safer. In the rock drilling construction process, a drill bit and a drill rod of the rock drilling trolley are components in direct contact with rock, and through setting drilling parameters such as propulsion pressure, rotation pressure, percussion pressure, drilling speed and the like in a control system of the rock drilling trolley, actions such as propulsion, rotation, percussion and the like are generated under the combined action of a hydraulic propulsion system and a hydraulic rock drill, so that the rock breaking function is realized. However, because the tunnel construction environment is complex and geological survey information is limited, rock mass cracks and holes, uneven rock hardness, poor soil mass and the like are often generated in the actual service process of the drill bit and the drill rod, drill rod clamping, drill tool abrasion, drill rod bending and the like are generated, the construction process is required to be interrupted by frequent replacement of the drill tool, if the drill rod is broken or serious equipment faults are caused by the drill tool abrasion, the machine halt and overhaul are carried out, the construction period is seriously influenced, therefore, the health degree of the drill tool and the loss condition of the drill tool are required to be monitored in real time in the construction process, and whether the drill tool needs to be replaced in advance or not is judged, and effective predictive maintenance is carried out.
According to different industries, the application scenes of the drilling tool are different, such as geological exploration, petroleum industry, coal industry, tunnel construction and the like, the adopted drilling tool is different in terms of materials, technology, shape, size, construction environment and the like, so that the health degree of the drilling tool is difficult to identify by a general and effective method due to the complexity of the environment, and the abrasion condition of the drilling tool is difficult to be measured by directly deploying a sensor on the drilling tool due to the fact that the service environment of the drilling tool is bad and the movement form is complex, and therefore, the health degree of the drilling tool is monitored or predicted in an indirect mode.
In view of the foregoing, there is a strong need for a method and a system for establishing a tool wear prediction model to solve the problem of tool wear prediction in the prior art.
Disclosure of Invention
The invention aims to provide a method for establishing a drilling tool wear prediction model and a drilling tool wear prediction system to solve the problem of drilling tool wear prediction in the prior art, and the specific technical scheme is as follows:
a method for establishing a drilling tool wear prediction model comprises the following steps:
step S1, acquiring all single drilling data of the current shift and drilling tool abrasion photos after the single drilling work is completed;
s2, preprocessing all single drilling data;
s3, extracting characteristics of the while-drilling parameters in the single drilling data to obtain data characteristics of the while-drilling parameters;
s4, calculating the abrasion loss of the drilling tool after single drilling according to the abrasion photo of the drilling tool in the step S1, and defining the health degree of the drilling tool after each drilling according to the abrasion loss;
and S5, establishing a mapping relation between the data characteristics of the while-drilling parameters and the health degree based on a deep learning method, and constructing a wear prediction model of the drilling tool.
In the above technical solution, in step S2, the while-drilling parameters in the single drilling data include one or more of a pushing pressure, a turning pressure, a striking pressure, and a drilling rate.
Preferably, the step S2 includes a step S2.1 and a step S2.2;
step S2.1: data cleaning is carried out on all single drilling data, and the cleaning rules are as follows: judging the data length of the single-time drilling data, if the data length of the single-time drilling data is within the range of [300,500], reserving the single-time drilling data, otherwise, eliminating the single-time drilling data;
step S2.2: judging whether to reject the single drilling data according to the drilling speed in the single drilling data, wherein the rule is as follows:
and if the drilling speed is larger than N m/min in the single drilling data and the subsequent drilling speeds are all kept larger than N m/min within T seconds, eliminating the single drilling data.
Preferably, step S2 further includes step S2.3, and the drilling state distinguishing process is performed on the retained single-time drilling data, which specifically includes the following steps:
the first step: dividing the single drilling data into a plurality of drilling states along a time domain according to the drilling speed in the single drilling data;
and a second step of: and taking the data set in the same drilling state in the single drilling data as a similar sample, and taking the single drilling data containing a plurality of similar samples as a processing sample of the next step.
In the above technical solution, in step S3, the data features of the parameter while drilling include statistical features of the parameter while drilling and time domain/frequency domain features of the parameter while drilling;
extracting statistical characteristics of parameters while drilling: extracting the maximum value, the minimum value, the average value, the variance, the origin moment and the center moment of the parameter while drilling to obtain the statistical characteristics of the parameter while drilling;
extracting time domain/frequency domain characteristics of the parameter while drilling: and extracting by adopting a wavelet transform-based mode maximum feature extraction method in a multi-scale space to obtain the time domain/frequency domain features of the parameter while drilling.
Preferably, the step S4 includes:
step S4.1: noise reduction and redundant component removal are carried out on the drilling tool abrasion photo after the current single drilling work is completed;
step S4.2: mapping all pixels of the drilling tool abrasion photo to be aligned onto a standard drilling tool photo based on a characteristic image alignment method so as to align the two photos;
step S4.3: measuring radial dimension D of abrasion photo of drilling tool based on standard drilling tool photo i And axial dimension H of the drilling tooth to the measurement reference i The method comprises the steps of carrying out a first treatment on the surface of the By radial dimension D i And an axial dimension H i Calculating the actual axial dimension h of the drilling tool after the current single drilling work is completed i ；
Step S4.4: according to the actual axial dimension h i Calculating the abrasion loss after the current single drilling work is completed;
step S4.5: and defining the health degree of the drilling tool after the current single drilling work is completed according to the abrasion loss.
In the above technical solution, preferably, in the step S4.3, the actual axial dimension h i As shown in formula 1):
Figure BDA0003975378330000031
where d represents the radial dimension of the standard drill.
In the above technical solution, preferably, in the step S4.4, the abrasion loss weather i As shown in formula 2);
Figure BDA0003975378330000032
/>
wherein h is i Representing the current single drillThe actual axial dimension of the drilling tool after the hole work is completed; h is a i-1 Representing the actual axial dimension of the drilling tool after the last single drilling operation is completed; h represents the actual axial dimension of the brand new drilling tool; s=1, the drilling tool after the current drilling is completed is a new replaced drilling tool; s=0 indicates that the drill after the current drilling is completed is an old drill that is not replaced;
in step S4.5, the rules defining the health degree are as follows:
when 0 is equal to or less than Wear i When the diameter is less than or equal to 0.1mm, the health degree of the drilling tool is abrasion-free;
when 0.1mm < Wear i When the thickness is less than or equal to 1mm, the health degree of the drilling tool is slightly worn;
when 1mm < Wear i When the length is less than or equal to 2mm, the health degree of the drilling tool is moderate abrasion;
when 2mm < Wear i When the thickness is less than or equal to 4mm, the health degree of the drilling tool is severely worn;
when the weather is i At > 4mm, the health of the drill is compromised.
Preferably, in the above technical solution, in the step S5,
based on a deep learning method, taking the data characteristics of the while-drilling parameters of the single drilling data in the step S3 as input parameters of a prediction model, taking the health degree of the drilling tool in the step S4 as output of the prediction model, and constructing a training set of deep learning so as to obtain a drilling tool abrasion prediction model.
A drilling tool wear prediction system comprises a data acquisition unit, a wear prediction unit and a display unit;
the data acquisition unit is used for acquiring parameters while drilling in drilling work; the abrasion prediction unit is provided with a prediction model obtained according to a method for establishing the abrasion prediction model of the drilling tool; the prediction model is connected with the data acquisition unit; the prediction model is connected with a display unit, and the display unit is used for displaying the abrasion loss and the health degree of the drilling tool.
The technical scheme of the invention has the following beneficial effects:
(1) According to the drilling tool abrasion prediction model establishing method, the mapping relation between the drilling parameters and the drilling tool abrasion loss is established, the drilling tool health and the abrasion loss are used as the input and the output based on deep learning, the drilling tool prediction model is obtained through training, the problem that the drilling tool abrasion loss is difficult to measure due to complex working conditions in the prior art can be solved, and in the follow-up application, the health condition of the drilling tool can be obtained through collecting the drilling parameters, the arrangement of parts such as sensors is not needed, and the practicability is good.
(2) In the invention, the parameters while drilling can be identified, classified and labeled, and the method specifically comprises the following steps: identifying starting and ending time points of finishing single drilling, and corresponding to the abrasion loss of the drilling tool; classifying various working conditions in the drilling process, classifying the parameters while drilling, and performing better subsequent abnormal data processing and model training robustness; in the labeling drilling process, the data set in the same drilling state is used as a similar sample for analysis, so that the accuracy of analysis is improved, and the unpredictable error caused by different drilling modes is avoided.
(3) In the invention, the image parameters are used as the judgment basis of the abrasion loss of the drilling tool, the abrasion loss of the drilling tool after each drilling can be obtained through an algorithm in sequence, the abrasion loss of the drilling tool corresponds to the drilling parameters during drilling one by one, the accuracy of model training is improved, and the abrasion loss value and the health degree of the drilling tool can be predicted and identified in various modes such as drilling tool by tool through different training models.
(4) According to the drilling tool wear prediction system, the drilling tool wear prediction model of the wear prediction unit is input by only collecting the while-drilling parameters, so that the drilling wear and the health degree can be displayed on the display unit, and the monitoring and the control of the drilling tool wear condition are realized.
In addition to the objects, features and advantages described above, the present invention has other objects, features and advantages. The present invention will be described in further detail with reference to the drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
In the drawings:
FIG. 1 is a flowchart of a method for establishing a model for predicting the wear of a drilling tool in the present embodiment;
FIG. 2 is a schematic diagram of the measurement of the dimensions of the drill bit in this embodiment;
fig. 3 is a schematic diagram showing the state of drilling in this embodiment.
Detailed Description
Embodiments of the invention are described in detail below with reference to the attached drawings, but the invention can be implemented in a number of different ways, which are defined and covered by the claims.
Examples:
the method for establishing the drilling tool wear prediction model comprises the following steps S1 to S5, wherein the steps are as shown in FIG. 1, and specifically comprise the following steps:
step S1, collecting drilling data of a current shift, dividing the drilling data of the current shift into a plurality of groups of single drilling data to obtain all the single drilling data, photographing a drilling tool after each single drilling is completed through a camera device to obtain a drilling tool abrasion photo after the single drilling work is completed, wherein the method comprises the following steps of: acquiring drilling parameter data including whole vehicle current, whole vehicle voltage, propulsion speed, percussion pressure and rotation pressure through a drilling trolley arm support data system, photographing the side abrasion condition of a drilling tool after each single drilling of a drilling trolley drilling tool corresponding to the data, namely photographing once every time the drilling is completed, and establishing a mapping relation one by one conveniently;
the detailed description is as follows:
collecting drilling data: the drilling trolley arm support data system can record drilling trolley hydraulic data and electric data in real time in the construction process, wherein the drilling rate, the propelling pressure, the striking pressure, the rotation speed, the water pressure, the water flow and the like are included, the sample data are normalized according to different parameter units and orders of magnitude, the sample data are converted into a dimensionless form, and the data converted into the dimensionless form have the characteristics of no units and similar orders of magnitude, so that the subsequent mathematical modeling analysis is facilitated.
Collecting a drilling tool abrasion photo: the photographs of the drill loss can be taken on a construction site, a prototype photograph before the drill is installed is taken as a reference for the subsequent loss, the drill is photographed after each single drilling operation is completed, the drill and the drill rod can be detached, the side photographs are taken along the radial direction, and preferably, the photographing machine position and the drill placement position are fixed, so that the error of the subsequent drill wear analysis can be reduced.
In step S1, the acquired current shift drilling data needs to be divided into multiple sets of single drilling data, specifically: in this embodiment, the collected drilling data is uninterrupted data flow during debugging and drilling of the drill jumbo, and in the single drilling process, if drilling data (drilling speed) is greater than N m/min (N m/min is 5m/min in this embodiment) and new normal drilling data (here, normal drilling data such as drilling speed 2-4 m/min) is no longer available in 2 or 3 seconds, the data flow (i.e. drilling data) is divided into multiple groups of single drilling data according to this mode. Of course, other existing means of dividing the overall borehole data into single-pass borehole data may be employed in addition to this.
Step S2, preprocessing all single drilling data acquired in the current shift, namely cleaning the data (removing abnormal values, invalid values, redundant values and the like), which specifically includes the following steps:
step S2.1: data cleaning is carried out on all the separated single drilling data, and the cleaning rules are as follows: judging the data length of the single-time drilling data, if the data length of the single-time drilling data is within the range of [300,500], reserving the single-time drilling data, otherwise, eliminating; the principle explanation of this step S2.1 is: in the step S2.1, based on the specificity of the drilling jumbo data, the traditional outlier analysis method is improved, because the traditional method only eliminates the data which is too high or too low by setting a threshold value, and the data anomalies possibly caused by the work such as overexcavation, undermining, trial drilling and the like in the actual drilling process are not fully considered, so the embodiment adopts the global outlier data cleaning method to judge the length of single-set drilling data (comprising the length of the electric hydraulic data characteristics when judging the whole running of the jumbo), if the obvious data size is reduced, the data is eliminated, the effectiveness of training data is ensured, and in the embodiment, the data size reduction reason can be determined by combining the on-site recording condition, and the database is remarked as "undermining and drilling", "trial drilling", "drill bit fracture", "drill rod fracture", "other anomalies", and the like.
Step S2.2: judging whether the single drilling data is rejected or not according to the drilling speed in the single drilling data, wherein the judging rule is as follows: if the drilling speed is higher than N m/min (namely N m/min is also preferably 5 m/min) in the single drilling data, and the subsequent drilling speed is kept higher than 5m/min in T seconds (T seconds are preferably 2 or 3 seconds), the single drilling data are eliminated;
it should be noted that: if the drilling speed is greater than 5m/min in the single drilling data, but the subsequent drilling speed is recovered to the normal drilling speed after T seconds pass (the normal drilling speed is 2.5m/min-3.5m/min, for example), the single drilling data is reserved, namely the situation that holes in rock are met at the moment is considered, and the data should be reserved;
step S2.3: and carrying out drilling state distinguishing processing on the reserved single-time drilling data, wherein the method comprises the following steps of:
the first step: dividing the single drilling data into a plurality of drilling states along a time domain by the drilling speed in the single drilling data, namely dividing the drilling states of each group of single drilling data;
and a second step of: taking the data set in the same drilling state in the single drilling data as a similar sample, and taking the single drilling data containing a plurality of similar samples as a processing sample of the next step (namely step S3), wherein the explanation is that: the single drilling data can be divided into a plurality of drilling states, and the plurality of drilling states divided by the single drilling data are classified to obtain a sample of the single drilling data after classification;
a further explanation of this step S2.3 is:
the single drilling data is divided into a plurality of drilling states along the time domain through the drilling speed, the data is divided into common drilling states of high-impact, low-impact, anti-jamming, drill withdrawal and other field construction along the time domain direction, the data set in the same drilling state in the single drilling data is used as a similar sample for analysis, the accuracy of analysis is improved, the phenomenon that unpredictable errors are caused by different drilling modes is avoided, the different drilling states are mainly represented by the drilling speed, as shown in fig. 3, according to the system setting of the rock drilling trolley, the average value of the feeding speed in the high-impact state is usually 4m/min, the average value of the feeding speed in the low-impact state is usually 2.5m/min, rapid decline occurs in the drilling process to form anti-jamming actions, the specific means for dividing the drilling states according to the drilling speed are known to those skilled in the art, and the embodiment is not repeated.
Step S3, extracting characteristics of all single drilling data classified in the step S2.3, and extracting characteristics of the drilling parameters in the single drilling data to obtain data characteristics of the drilling parameters, wherein the data characteristics of the drilling parameters comprise statistical characteristics of the drilling parameters and time domain/frequency domain characteristics of the drilling parameters, and the data characteristics of the drilling parameters comprise the following steps:
extracting statistical characteristics of parameters while drilling: the method comprises the steps of extracting the maximum value, the minimum value, the average value, the variance, the origin moment and the center moment of the while-drilling parameters in single drilling data, for example, the maximum value, the minimum value, the average value, the variance and the like of the propulsion pressure aiming at the while-drilling parameters such as the propulsion pressure;
extracting time domain/frequency domain characteristics of the parameter while drilling: the method for extracting the mode maximum feature of the multi-scale space based on wavelet transformation is adopted to extract the time domain/frequency domain feature of the parameter while drilling, the time domain/frequency domain feature can be extracted by referring to the prior art, and the embodiment provides a specific step for extracting the time domain/frequency domain feature, which comprises the following steps:
the first step: the wavelet basis function performs scaling and translation transformation:
Figure BDA0003975378330000071
wherein ψ (·) is the wavelet basis function, t is the original data abscissa, a is the scale factor, and b is the translation factor.
And a second step of: the continuous wavelet transform for any raw data f (t) is:
Figure BDA0003975378330000081
in which W is f (a, b) is a wavelet transformed data function comprising two independent variables of a scaling factor a and a shifting factor b; it can be seen that the continuous wavelet transform is made from f (t) →W f The mapping of (a, b) is inverse transformed to:
Figure BDA0003975378330000082
in the method, in the process of the invention,
Figure BDA0003975378330000083
to meet the constraint of inverse transformation->
Figure BDA0003975378330000084
Fourier transform of ψ (t);
and a third step of: performing wavelet transformation on the single-time drilling data based on a Mallat algorithm, and then decomposing the single-time drilling data into high-frequency and low-frequency components to obtain time domain/frequency domain characteristics (namely, the single-time drilling data can be decomposed into frequency domain components after being subjected to wavelet transformation through the steps);
step S4, calculating the abrasion loss of the drilling tool after single drilling according to the abrasion photo of the drilling tool in the step S1, and defining the health degree of the drilling tool after each drilling according to the abrasion loss, wherein the general thought of the step S4 is as follows: performing image processing on the shot drilling tool abrasion picture, removing redundant components and noise points of the picture, calibrating a reference calculation position of the drilling tool (drill rod) through an alignment algorithm, and measuring to obtain the radial abrasion loss of the drilling tool (drill bit); meanwhile, the actual condition of the drill bit is combined, the health degree of the loss of the drilling tool after single drilling is established, namely, health degree indexes are defined, and the loss condition is recorded, and specifically, the method comprises the following steps:
step S4.1: after the single drilling is completed, the drilling tool abrasion photo is shot to reduce noise and remove redundant components, and the noise reduction and the redundant component removal can be performed by referring to the prior art;
step S4.2: the feature-based image alignment method maps all pixels of the drill wear photograph to be aligned onto a standard drill photograph to align the two photographs as follows:
first, in this step S4.2, the general idea of the feature-based image alignment method is:
acquiring a brand new drill bit picture (namely a picture of a standard drilling tool), and converting the picture into a standard side view by using a Photoshop tool to serve as a reference picture for alignment of all drilling tool abrasion pictures;
the ORB is used for extracting the characteristic points, the ORB is a characteristic point detector and consists of two parts, namely a positioner, a point with rotation invariance, scaling invariance and affine invariance on a picture is found, and a descriptor is obtained to distinguish the characteristic points by obtaining the appearance codes of the characteristic points, so that the characteristic points can be represented by using the descriptor, and ideally, the same physical point corresponding to different pictures should have the same descriptor;
the method comprises the following specific steps:
1): respectively reading a reference picture (namely a picture of a standard drilling tool) and a picture to be aligned (namely a wear picture of the drilling tool) into a memory;
2) Detecting ORB feature points for the two graphs, controlling the number of the detected feature points by using a parameter MAX_FEATURES, detecting the feature points by using a detectANdCommutte function and calculating descriptors;
3) And finding out matched characteristic points in the two graphs, arranging according to the matching degree, reserving the least matched part, and measuring the similarity of the descriptors of the two characteristic points by using Hamming distance.
4) The matched characteristic points generated in the previous step may have certain errors, and a homography matrix is calculated under the condition that certain matching errors possibly occur by using a random sampling coincidence algorithm (Random Sample Consensus);
5) Mapping all pixels of the picture to be aligned to the alignment picture on the reference picture by using a warp Perselected function, so that the alignment of the reference picture and the picture to be aligned can be realized;
step S4.3: radial dimension D of abrasion photo of drilling tool is measured respectively i And axial dimension H of the drilling tooth to the measurement reference i The method comprises the steps of carrying out a first treatment on the surface of the By radial dimension D i And an axial dimension H i Calculating the actual axial dimension h of the drilling tool after the current single drilling work is completed i The method specifically comprises the following steps: opening the registered picture P using Photoshop tool i Respectively measure the radial dimension D i And axial dimension H of the drilling tooth to the measurement reference i As shown in fig. 2;
wherein, the radial dimension d of the known standard drill bit (namely the brand new drill bit) is 45mm, then the picture P i Is greater than the actual axial dimension h of i As shown in formula 1):
Figure BDA0003975378330000091
step S4.4: according to the actual axial dimension h i The abrasion loss after the current single drilling operation is completed is calculated, in the embodiment, the abrasion loss of the drill bit is defined as the difference of the axial dimensions of the drill bit in two continuous drilling operations, and the abrasion loss Wear is defined i As shown in formula 2):
Figure BDA0003975378330000092
wherein h is i Representing the actual axial dimension of the drilling tool after the current single drilling operation is completed; h is a i-1 Representing the actual axial dimension of the drilling tool after the last single drilling operation is completed; h represents the actual axial dimension (i.e., standard dimension) of the new drilling tool;
s=1 indicates that the drilling tool after the single drilling is completed is a new replacement drilling tool; s=0 indicates that the drilling tool after the single drilling is completed is an old drilling tool that is not replaced;
further explanation of formula 2) follows:
if the drilling tool is a new drilling tool (i.e. s=1), the actual axial dimension h of the drilling tool after the current single drilling operation is completed i With the actual axial direction of standard drilling toolsDimension h i The difference value of the two is the abrasion loss after the current single drilling work is completed;
if the drilling tool is an old drilling tool (i.e. s=0), the actual axial dimension h of the drilling tool after the current single drilling operation is completed i And the actual axial dimension h of the drilling tool after the last single drilling operation is finished i-1 The difference value of the two is the abrasion loss after the current single drilling work is completed;
the distinguishing standard of the new drilling tool and the old drilling tool is as follows: if the drilling times of the drilling tool are smaller than or equal to one time, the drilling tool is a new drilling tool, and otherwise, the drilling tool is an old drilling tool.
Step S4.5: and defining the health degree of the drilling tool after the current single drilling work is completed according to the abrasion loss, and recording the abrasion loss, wherein the abrasion loss is as follows:
when 0 is equal to or less than Wear i When the diameter is less than or equal to 0.1mm, the health degree of the drilling tool is abrasion-free;
when 0.1mm < Wear i When the thickness is less than or equal to 1mm, the health degree of the drilling tool is slightly worn;
when 1mm < Wear i When the length is less than or equal to 2mm, the health degree of the drilling tool is moderate abrasion;
when 2mm < Wear i When the thickness is less than or equal to 4mm, the health degree of the drilling tool is severely worn;
when the weather is i When the diameter is more than 4mm, the health degree of the drilling tool is damaged;
preferably, the health of the drilling tool is defined, and the actual use time of the drilling tool can be evaluated.
Step S5, based on a deep learning method, establishing a mapping relation between data characteristics and health of parameters while drilling, and training to obtain a wear prediction model of the drilling tool, wherein the method comprises the following steps of:
in the step S3, a data feature matrix (i.e. time domain/frequency domain feature) and statistical features of the current drilling state of the drilling tool are obtained according to the while-drilling parameters, the data feature matrix and the statistical features are used as input parameters of a mathematical model (i.e. a convolutional neural network model), meanwhile, the abrasion loss after each drilling is solved based on the image alignment algorithm in the step S4, the health degree is defined as output of the mathematical model, a training set for training the deep learning mathematical model is constructed, and training and parameter adjustment optimization are carried out on the mathematical model based on the training set, so that the loss health degree prediction model suitable for the drilling tool of the rock drilling trolley is obtained.
The embodiment also discloses a drilling tool abrasion prediction system, which comprises a data acquisition unit, an abrasion prediction unit and a display unit; the data acquisition unit is used for acquiring parameters while drilling in drilling work as input; the abrasion prediction unit is provided with a drilling tool abrasion prediction system, and the drilling tool abrasion prediction system is obtained by establishing according to the drilling tool abrasion prediction model establishing method; the drilling tool abrasion prediction model is connected with the data acquisition unit, the abrasion prediction model is input along with drilling parameters, the abrasion prediction model can output the abrasion loss and the health degree of the drilling tool, the prediction model is connected with the display unit, and the display unit is used for displaying the abrasion loss and the health degree of the drilling tool.
In this embodiment, a specific example is provided for the method for establishing the drilling tool wear prediction model, as follows:
as shown in table 1, table 1 illustrates the parameter while drilling data collected in the above step S1;
table 1 partial while drilling parameters for single borehole data
Figure BDA0003975378330000111
Step S2: preprocessing the acquired while-drilling parameter data, cleaning the acquired while-drilling parameter data, dividing the cleaned data according to drilling tool construction actions, and obtaining corresponding data blocks of construction actions (drilling states) such as preparation actions, high flushing, low flushing, drill clamping prevention, drill withdrawal and the like, wherein 1 line and 2 lines of data are prepared actions in table 1, 1000-1002 lines of data represent a high flushing stage, and 4204 lines of data represent drill withdrawal states.
Step S3: based on the data obtained by preprocessing in the previous step, constructing a characteristic matrix of the working loss of the drill jumbo by using a mode maximum value characteristic extraction method of a multi-scale space based on wavelet transformation, wherein table 2 shows partial data of the characteristic matrix X of the construction parameters of the drill jumbo in a certain stage;
table 2 drill jumbo construction parameter feature matrix X partial data
Figure BDA0003975378330000121
The construction parameter feature matrix X has a matrix dimension of (29, 400, 120), and table 2 shows X (0, 400, 120) partial data.
Step S4: performing image processing on drilling tool damage pictures corresponding to 30 groups of single drilling data, removing redundant components and noise points of the pictures, calibrating drill rod reference calculation positions through an alignment algorithm, and measuring to obtain the abrasion loss of the drill bit, wherein the abrasion loss is shown in tables 3-1 and 3-2:
TABLE 3-1 front 15 drilling tool wear
Figure BDA0003975378330000122
Table 3-2 post 15 drill wear level
Figure BDA0003975378330000131
Meanwhile, the actual service time of the drill bit is combined, a health degree label of the loss of the drilling tool after single drilling is established, namely, health degree indexes are defined, loss conditions are recorded, and the health degree is defined as shown in the following table 4:
table 4 health of drilling tools
Figure BDA0003975378330000132
Step S5: training a convolutional neural network model by using sample data and health, wherein the accuracy of a sample set is 95%; and selecting the other set of data sets containing 30 sets of samples while drilling parameters as a test set, and verifying the test set by using a convolutional neural network model obtained through training, wherein the accuracy of the verification result reaches 80%.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. The method for establishing the drilling tool wear prediction model is characterized by comprising the following steps of:
step S1, acquiring all single drilling data of the current shift and drilling tool abrasion photos after the single drilling work is completed;
s2, preprocessing all single drilling data;
s3, extracting characteristics of the while-drilling parameters in the single drilling data to obtain data characteristics of the while-drilling parameters;
s4, calculating the abrasion loss of the drilling tool after single drilling according to the abrasion photo of the drilling tool in the step S1, and defining the health degree of the drilling tool after each drilling according to the abrasion loss;
and S5, establishing a mapping relation between the data characteristics of the while-drilling parameters and the health degree based on a deep learning method, and constructing a wear prediction model of the drilling tool.
2. The method according to claim 1, wherein in the step S2, the while-drilling parameters in the single-pass drilling data include one or more of a push pressure, a swing pressure, a percussion pressure, and a drilling rate.
3. The method of creating a predicted model of drill wear according to claim 2, wherein step S2 includes steps S2.1 and S2.2;
step S2.1: data cleaning is carried out on all single drilling data, and the cleaning rules are as follows: judging the data length of the single-time drilling data, if the data length of the single-time drilling data is within the range of [300,500], reserving the single-time drilling data, otherwise, eliminating the single-time drilling data;
step S2.2: judging whether to reject the single drilling data according to the drilling speed in the single drilling data, wherein the rule is as follows:
and if the drilling speed is larger than N m/min in the single drilling data and the subsequent drilling speeds are all kept larger than N m/min within T seconds, eliminating the single drilling data.
4. The method for building a predicted model of drill wear according to claim 3, wherein step S2 further comprises step S2.3 of performing a drilling state discrimination process on the retained single-pass drilling data, specifically as follows:
the first step: dividing the single drilling data into a plurality of drilling states along a time domain according to the drilling speed in the single drilling data;
and a second step of: and taking the data set in the same drilling state in the single drilling data as a similar sample, and taking the single drilling data containing a plurality of similar samples as a processing sample of the next step.
5. The method according to claim 3 or 4, wherein in step S3, the data features of the parameter while drilling include statistical features of the parameter while drilling and time/frequency domain features of the parameter while drilling;
extracting statistical characteristics of parameters while drilling: extracting the maximum value, the minimum value, the average value, the variance, the origin moment and the center moment of the parameter while drilling to obtain the statistical characteristics of the parameter while drilling;
extracting time domain/frequency domain characteristics of the parameter while drilling: and extracting by adopting a wavelet transform-based mode maximum feature extraction method in a multi-scale space to obtain the time domain/frequency domain features of the parameter while drilling.
6. The method for building a predicted model of drill wear according to claim 1, wherein the step S4 comprises:
step S4.1: noise reduction and redundant component removal are carried out on the drilling tool abrasion photo after the current single drilling work is completed;
step S4.2: mapping all pixels of the drilling tool abrasion photo to be aligned onto a standard drilling tool photo based on a characteristic image alignment method so as to align the two photos;
step S4.3: measuring radial dimension D of abrasion photo of drilling tool based on standard drilling tool photo i And axial dimension H of the drilling tooth to the measurement reference i The method comprises the steps of carrying out a first treatment on the surface of the By radial dimension D i And an axial dimension H i Calculating the actual axial dimension h of the drilling tool after the current single drilling work is completed i ；
Step S4.4: according to the actual axial dimension h i Calculating the abrasion loss after the current single drilling work is completed;
step S4.5: and defining the health degree of the drilling tool after the current single drilling work is completed according to the abrasion loss.
7. The method according to claim 6, wherein in the step S4.3, the actual axial dimension h i As shown in formula 1):
Figure FDA0003975378320000021
where d represents the radial dimension of the standard drill.
8. The method for building a predicted model of drill Wear according to claim 7, wherein in the step S4.4, the Wear amount spar i As shown in formula 2);
Figure FDA0003975378320000022
wherein h is i Representing the actual axial dimension of the drilling tool after the current single drilling operation is completed; h is a i-1 Representing the actual axial dimension of the drilling tool after the last single drilling operation is completed; h represents the actual axial dimension of the brand new drilling tool; s=1, the drilling tool after the current drilling is completed is a new replaced drilling tool; s=0 represents the current drilling after completion of the drillingThe drilling tool is an old drilling tool which is not replaced;
in step S4.5, the rules defining the health degree are as follows:
when 0 is equal to or less than Wear i When the diameter is less than or equal to 0.1mm, the health degree of the drilling tool is abrasion-free;
when 0.1mm < Wear i When the thickness is less than or equal to 1mm, the health degree of the drilling tool is slightly worn;
when 1mm < Wear i When the length is less than or equal to 2mm, the health degree of the drilling tool is moderate abrasion;
when 2mm < Wear i When the thickness is less than or equal to 4mm, the health degree of the drilling tool is severely worn;
when the weather is i At > 4mm, the health of the drill is compromised.
9. The method for creating a model for predicting wear of a drilling tool according to any one of claims 6 to 8, wherein in the step S5,
based on a deep learning method, taking the data characteristics of the while-drilling parameters of the single drilling data in the step S3 as input parameters of a prediction model, taking the health degree of the drilling tool in the step S4 as output of the prediction model, and constructing a training set of deep learning so as to obtain a drilling tool abrasion prediction model.
10. The drilling tool wear prediction system is characterized by comprising a data acquisition unit, a wear prediction unit and a display unit;
the data acquisition unit is used for acquiring parameters while drilling in drilling work; the wear prediction unit is provided with a prediction model obtained by the method for establishing a drilling tool wear prediction model according to claim 9; the prediction model is connected with the data acquisition unit; the prediction model is connected with a display unit, and the display unit is used for displaying the abrasion loss and the health degree of the drilling tool.
CN202211533649.0A 2022-12-01 2022-12-01 Method for establishing drilling tool wear prediction model and drilling tool wear prediction system Pending CN116186910A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211533649.0A CN116186910A (en) 2022-12-01 2022-12-01 Method for establishing drilling tool wear prediction model and drilling tool wear prediction system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211533649.0A CN116186910A (en) 2022-12-01 2022-12-01 Method for establishing drilling tool wear prediction model and drilling tool wear prediction system

Publications (1)

Publication Number Publication Date
CN116186910A true CN116186910A (en) 2023-05-30

Family

ID=86435275

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211533649.0A Pending CN116186910A (en) 2022-12-01 2022-12-01 Method for establishing drilling tool wear prediction model and drilling tool wear prediction system

Country Status (1)

Country Link
CN (1) CN116186910A (en)

Similar Documents

Publication Publication Date Title
KR102008973B1 (en) Apparatus and Method for Detection defect of sewer pipe based on Deep Learning
US11227375B2 (en) Machine vision and machine intelligence aided electronic and computer system for assisting on formation drilling, boring, tunneling on-line guidance, assisted decision and dull grading system for drilling tool and associated drill string components
US11030735B2 (en) Subterranean drill bit management system
CN115638833B (en) Monitoring data processing method and system
WO2015200567A1 (en) Anomaly recognition system and methodology
CN103927553B (en) Coal and rock recognition method based on multi-scale micro-lamination and contrast ratio joint distribution
CN114387328A (en) RQD calculation method based on deep learning model and core image
CN111582135A (en) Excavator hand operation proficiency evaluation method and device based on working stage recognition
CN113962951B (en) Training method and device for detecting segmentation model, and target detection method and device
CN109543268B (en) TBM propulsion main influencing factor identification method based on kriging model
CN114429563A (en) Surrounding rock integrity identification method and system based on while-drilling test and TBM rock slag image
CN116186910A (en) Method for establishing drilling tool wear prediction model and drilling tool wear prediction system
Hamdi et al. Fully automated approach for patterned fabric defect detection
CN117522149A (en) Tunnel security risk identification method and device and security management platform
KR102311558B1 (en) System and method for detecting structure damage using artificial intelligence, and a recording medium having computer readable program for executing the method
Lin et al. A novel domain adversarial time-varying conditions intervened neural network for drill bit wear monitoring of the jumbo drill under variable working conditions
CN110107274B (en) Real-time online monitoring system and monitoring method for TBM (Tunnel boring machine) jumbolter based on hydraulic system
CN116597193A (en) Tunnel intelligent construction-oriented tunnel face unmanned geological logging method and system
Myrans et al. Using Automatic Anomaly Detection to Identify Faults in Sewers:(027)
CN116296956A (en) Hobbing cutter abrasion detection method and device
CN116229354A (en) Face image surrounding rock grade identification method based on characteristic parameter automatic extraction
Lemy et al. Image analysis of drill core
CN110458824B (en) Drill bit wear detection method
Arnaout et al. Drilling events detection using hybrid intelligent segmentation algorithm
US20230167732A1 (en) Detecting Defects in Tubular Structures

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