Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a drilling parameter standardized cleaning method and device based on dynamic linear piecewise representation, which reduces the influence of drilling parameter data fluctuation and randomness characteristics caused by non-geological factors such as working modes, machine vibration and the like on surrounding rock classification precision and solves the problems in the background art.
In order to achieve the above purpose, the present invention provides the following technical solutions: a drilling parameter standardized cleaning method based on dynamic linear piecewise representation, which is characterized by comprising the following steps:
s1, removing discrete values of drilling parameters by combining Bayesian confidence interval inspection;
s2, determining a drilling parameter segmentation depth interval by a dynamic linear expression method to obtain segmentation time points and segmentation depth points;
s3, processing the data set with the discrete values removed by using a filtering method, and reducing the fluctuation degree of the data;
and S4, calculating the drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value based on the filtered data.
Preferably, in the step S1, removing discrete values of drilling parameters in combination with bayesian confidence interval test specifically includes:
s11, for any drilling parameter, the drilling parameter is processed at any drilling time t i Y for time series data of (2) ti ,t i A representation;
s12, order[y min ,y max ]As drilling parameter Y t Mean characteristic parameter->Is uniformly distributed a priori;
S13、and Y is equal to t The joint distribution of (2) is expressed as:
in the method, in the process of the invention,is a variance characteristic parameter;
S14、the probability density function of the posterior distribution is expressed as:
S15、the posterior expectation estimate of (2) is expressed as:
s16, the confidence range of drilling parameter data generated based on a posterior expectation estimation method is expressed as follows:
wherein, alpha is the confidence level and takes the value of 0.005;
s17, then t 0 Time of day drilling parametersThe 1-alpha confidence range of (2) is:
if it isStatistical decision y ti Is a discrete value.
Preferably, in the step S2, the section depth interval of the drilling parameter is determined by a dynamic linear expression method, so as to obtain a section time point and a section depth point, which specifically includes:
s21, connecting the first data and the last data of the time sequence data as a first linear segmentation, and solving a linear equation as a fitting curve;
s22, obtaining the maximum error (vertical distance) d of the original data and the fitting curve max :
d max =max(|D l,i -D 0,i |)=max(d i )
Wherein D is l,i 、D 0,i 、d i The depth value, the original depth value and the vertical distance of the linear piecewise fitting curve of the ith drilling time are respectively represented;
s23, d max Performing second linear segmentation by taking the point as a new linear segmentation point, and solving a new linear equation (segmentation) as a new fitting curve;
s24, repeating the step S22 until the segmentation stopping condition is met: d, d max Less than or equal to 0.1m or the total number of linear segments n l =5, resulting in a segmentation time point and a segmentation depth point.
Preferably, the filtering method in the step S3 is kalman filtering.
Preferably, in the step S4, based on the filtered data, a drilling parameter mean value in each segmented depth interval range is calculated as a drilling parameter surrounding rock classification characteristic value, and specifically includes:
s41, directly solving the average value of all the linear piecewise drilling parameters as the representative value of the piecewise drilling parameters of the blasthole after filtering the drilling parameters except the drilling speed, wherein the calculation formula is as follows:
in the method, in the process of the invention,the drilling parameter mean value except the drilling speed is the ith linear segment; p (P) i,j Jth data for an ith linear segmented drilling parameter; n is n i The number of the linear piecewise drilling parameter data is the ith;
s42, regarding the drilling speed, taking the slope of each linear piecewise drilling time-depth curve secant as a representative value of the drilling speed of the blast hole, and calculating the following formula:
in the method, in the process of the invention,the mean value of the ith linear piecewise drilling speed; d (D) i+1 、T i+1 The depth and time of the ith linear piecewise endpoint; d (D) i 、T i Is the i-th linear starting point depth and time.
In addition, in order to achieve the above purpose, the present invention also provides the following technical solutions: a drilling parameter normalization cleaning device based on dynamic linear piecewise representation, the device comprising:
a discrete value removal module: removing discrete values of drilling parameters by combining Bayes confidence interval test;
dynamic linear segment representation module: determining a drilling parameter piecewise depth interval by a dynamic linear expression method to obtain piecewise time points and piecewise depth points;
and a filtering module: processing the data set with discrete values removed by using a filtering method, and reducing the fluctuation degree of the data;
drilling parameter mean value calculation module: and based on the filtered data, calculating a drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value.
In addition, in order to achieve the above purpose, the present invention also provides the following technical solutions: an electronic device, the electronic device comprising: a processor; and a memory for storing one or more programs;
the one or more programs, when executed by the processor, cause the processor to perform the drilling parameter normalization cleaning method.
In addition, in order to achieve the above purpose, the present invention also provides the following technical solutions: a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the drilling parameter normalization cleaning method.
The beneficial effects of the invention are as follows: according to the method, the influence of drilling parameter data fluctuation and randomness characteristics caused by non-geological factors such as working modes, machine vibration and the like on surrounding rock classification precision is reduced by the method for acquiring the surrounding rock classification drilling parameter characteristic value through dynamic linear piecewise representation.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some of the embodiments of the present application, but not all of the embodiments. All other embodiments, based on the embodiments herein, which would be apparent to one of ordinary skill in the art without undue burden are within the scope of the present application.
Accordingly, the following detailed description of the embodiments of the present application, provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, based on the embodiments herein, which would be apparent to one of ordinary skill in the art without undue burden are within the scope of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
The inventor provides a drilling parameter cleaning standardization method based on dynamic linear piecewise representation through long-term research, and aims to reduce influences of drilling parameter data fluctuation and randomness characteristics caused by non-geological factors such as working modes, machine vibration and the like on surrounding rock classification precision. The data processing method is used for processing the raw data of the blast hole, and can be still applied when different drilling parameter grading characteristics (such as drilling specific energy and the like) are adopted.
Referring to fig. 1 to 7, the present invention provides a technical solution: a drilling parameter standardized cleaning method based on dynamic linear piecewise representation, as shown in fig. 1, comprises the following steps:
s1, testing and removing discrete values of drilling parameters by combining Bayesian confidence intervals.
The method comprises the following specific steps:
s11, for any drilling parameter, the drilling parameter is processed at any drilling time t i Y for time series data of (2) ti ,t i A representation;
s12, order[y min ,y max ]As drilling parameter Y t Mean characteristic parameter->Is uniformly distributed a priori;
S13、and Y is equal to t The joint distribution of (2) is expressed as:
in the method, in the process of the invention,is a variance characteristic parameter;
S14、the probability density function of the posterior distribution is expressed as:
S15、the posterior expectation estimate of (2) is expressed as:
s16, the confidence range of drilling parameter data generated based on a posterior expectation estimation method is expressed as follows:
wherein, alpha is the confidence level and takes the value of 0.005;
s17, then t 0 Time of day drilling parametersThe 1-alpha confidence range of (2) is:
if it isStatistical decision y ti Is a discrete value.
S2, determining a drilling parameter segmentation depth interval by a dynamic linear expression method to obtain segmentation time points and segmentation depth points.
The method comprises the following specific steps:
s21, connecting the first data and the last data of the time sequence data as a first linear segmentation, and solving a linear equation as a fitting curve;
s22, obtaining the maximum error (vertical distance) d of the original data and the fitting curve max :
d max =max(|D l,i -D 0,i |)=max(d i )
Wherein D is l,i 、D 0,i 、d i The depth value, the original depth value and the vertical distance of the linear piecewise fitting curve of the ith drilling time are respectively represented;
s23, d max Performing second linear segmentation by taking the point as a new linear segmentation point, and solving a new linear equation (segmentation) as a new fitting curve;
s24, repeating the step S22 until the segmentation stopping condition is met: d, d max Less than or equal to 0.1m or the total number of linear segments n l =5, resulting in a segmentation time point and a segmentation depth point.
S3, processing the data set with the discrete values removed by using a filtering method, and reducing the fluctuation degree of the data.
In a specific embodiment, the data set after the discrete points are removed using a Kalman filter process.
And S4, calculating the drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value based on the filtered data.
The method comprises the following specific steps:
s41, directly solving the average value of all the linear piecewise drilling parameters as the representative value of the piecewise drilling parameters of the blasthole after filtering the drilling parameters except the drilling speed, wherein the calculation formula is as follows:
in the method, in the process of the invention,the drilling parameter mean value except the drilling speed is the ith linear segment; p (P) i,j Jth data for an ith linear segmented drilling parameter; n is n i The number of the linear piecewise drilling parameter data is the ith;
s42, regarding the drilling speed, taking the slope of each linear piecewise drilling time-depth curve secant as a representative value of the drilling speed of the blast hole, and calculating the following formula:
in the method, in the process of the invention,the mean value of the ith linear piecewise drilling speed; d (D) i+1 、T i+1 Is the firsti linear piecewise endpoint depths, times; d (D) i 、T i Is the i-th linear starting point depth and time.
In a particular embodiment, the drilling parameters include feed rate V p Percussion pressure P h Swing pressure P r Propulsion pressure P f The drilling parameter processing flow chart is shown in fig. 3.
In a specific embodiment, a typical borehole drilling parameter timing curve is shown in FIG. 4. As can be seen from fig. 4, the drilling parameters of the individual blastholes have distinct longitudinal segmentation, waviness and discrete characteristics. The longitudinal segmentation is mainly caused by personnel operation, a trolley control system and surrounding rock quality change; the fluctuations are mainly caused by machine vibrations or abrupt changes in the surrounding rock; the discrete features are mainly caused by sensor recording anomalies.
The time sequence curve of the standardized drilling parameters formed after the blast hole is cleaned by the steps is shown in fig. 5.
As can be seen by comparing fig. 4 and fig. 5, each drilling parameter after cleaning exhibits obvious sectional uniformity characteristics, and is more uniform than the original drilling parameters and data, and more suitable for geological characteristics that the surrounding rock level is relatively uniform within a certain range and frequent and excessive changes do not occur. Therefore, the influence of non-geological factors such as personnel operation, a control system, machine vibration, recording abnormality and the like on intelligent classification of the surrounding rock is fully considered.
In a specific embodiment, probability density distribution histograms of standard deviations of drilling parameters of the face before and after processing of the sample data of the face samples of the class II, class III and class IV are calculated and plotted respectively, and standard deviation mean statistics are shown in table 1.
TABLE 1 Standard deviation statistics of drilling parameters for face before and after data processing
As can be seen from fig. 6 and table 1, the standard deviation of the drilling parameters of the face under the same surrounding rock level after data processing is reduced (the histogram moves downwards), the average reduction is 28.72% -82.68%, which indicates that the dispersion of the drilling parameters of each face after data processing is reduced, the representative enhancement of the drilling parameter mean value of the face on all the drilling parameters of the face is improved, and the use of the drilling parameter mean value of the face as the input of the intelligent surrounding rock classification model is more scientific and reasonable, thereby being beneficial to the improvement of the intelligent surrounding rock classification effect based on the drilling parameters.
In a specific embodiment, 6 kinds of machine learning algorithms such as a support vector machine (Support Vector Machine, abbreviated as SVM), a K nearest neighbor algorithm (K-Nearest Neighbour, abbreviated as KNN), a Random Forest (RF), an Extreme Tree (ET), a bagging method (Bagging Classifier, abbreviated as Bag), a gradient elevator (Gradient Boosting, abbreviated as GB) and the like are used as inputs to train 6 kinds of intelligent surrounding rock classification models, so as to compare model performances before and after data cleaning. The comparison chart of the model accuracy before and after data processing is shown in fig. 7.
As can be seen from fig. 7, the accuracy of the intelligent classification model of each surrounding rock after data cleaning is improved from 85.3% -88.8% to 88.1% -89.9%, and the accuracy is improved by 1.0% -3.6%, so that the effectiveness of the drilling parameter standardized cleaning method provided by the invention is verified.
The drilling parameter standardized cleaning method based on dynamic linear piecewise representation is a full-flow data cleaning method established aiming at the characteristic that the surrounding rock level does not change too much in a small scale range under the background of wide application of the surrounding rock quality identification related technology based on the drilling parameters, so that the drilling parameters used for grading are relatively stable. Compared with original data, the standardized data characteristics formed after cleaning can effectively reduce the influence of drilling parameter data fluctuation and randomness characteristics caused by non-geological factors such as working modes, machine vibration and the like on the precision of surrounding rock classification tasks, improve the application level of the surrounding rock intelligent classification technology based on drilling parameters, and provide references for surrounding rock quality evaluation based on the drilling parameters and time sequence data processing tasks with similar characteristics.
Based on the same inventive concept as the above method embodiment, the present application further provides a drilling parameter standardization cleaning device based on dynamic linear piecewise representation, which can implement the functions provided by the above method embodiment, as shown in fig. 8, and the device includes:
discrete value removal module 110: removing discrete values of drilling parameters by combining Bayes confidence interval test;
dynamic linear segment representation module 120: determining a drilling parameter piecewise depth interval by a dynamic linear expression method to obtain piecewise time points and piecewise depth points;
the filtering module 130: processing the data set with discrete values removed by using a filtering method, and reducing the fluctuation degree of the data;
drilling parameter mean calculation module 140: and based on the filtered data, calculating a drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value.
Based on the same inventive concept as the method embodiment described above, an embodiment of the present application further provides an electronic device, as shown in fig. 9, including: a processor 210; and a memory 220 for storing one or more programs;
the one or more programs, when executed by the processor 210, cause the processor to perform the drilling parameter normalization cleaning method.
The drilling parameter standardized cleaning method specifically comprises the following steps:
removing discrete values of drilling parameters by combining Bayes confidence interval test;
determining a drilling parameter piecewise depth interval by a dynamic linear expression method to obtain piecewise time points and piecewise depth points;
processing the data set with discrete values removed by using a filtering method, and reducing the fluctuation degree of the data;
and based on the filtered data, calculating a drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value.
Based on the same inventive concept as the above-described method embodiments, the present application further provides a computer-readable storage medium, characterized in that: on which a computer program is stored which, when being executed by the processor 210, implements the described drilling parameter normalization cleaning method.
The drilling parameter standardized cleaning method specifically comprises the following steps:
removing discrete values of drilling parameters by combining Bayes confidence interval test;
determining a drilling parameter piecewise depth interval by a dynamic linear expression method to obtain piecewise time points and piecewise depth points;
processing the data set with discrete values removed by using a filtering method, and reducing the fluctuation degree of the data;
and based on the filtered data, calculating a drilling parameter mean value in the range of each segmented depth interval as a drilling parameter surrounding rock grading characteristic value.
According to the method, through cleaning drilling parameters, standardized data characteristics are formed, and influences of drilling parameter data fluctuation and randomness characteristics caused by non-geological factors such as working modes, machine vibration and the like on precision of surrounding rock classification tasks can be reduced.
Although the present invention has been described with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described, or equivalents may be substituted for elements thereof, and any modifications, equivalents, improvements and changes may be made without departing from the spirit and principles of the present invention.