CN109443719B - Drill bit vibration signal online virtual test method and system thereof - Google Patents

Drill bit vibration signal online virtual test method and system thereof Download PDF

Info

Publication number
CN109443719B
CN109443719B CN201811295029.1A CN201811295029A CN109443719B CN 109443719 B CN109443719 B CN 109443719B CN 201811295029 A CN201811295029 A CN 201811295029A CN 109443719 B CN109443719 B CN 109443719B
Authority
CN
China
Prior art keywords
data
drill bit
signal
vibration
signals
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.)
Expired - Fee Related
Application number
CN201811295029.1A
Other languages
Chinese (zh)
Other versions
CN109443719A (en
Inventor
乔美英
闫书豪
兰建义
王波
陶慧
许城宽
汤夏夏
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.)
Henan University of Technology
Original Assignee
Henan University of Technology
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 Henan University of Technology filed Critical Henan University of Technology
Priority to CN201811295029.1A priority Critical patent/CN109443719B/en
Publication of CN109443719A publication Critical patent/CN109443719A/en
Application granted granted Critical
Publication of CN109443719B publication Critical patent/CN109443719B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention discloses a drill bit vibration signal online virtual test method and a system thereof, wherein the method comprises the following steps: s1, collecting vibration signals of the drill bit to be tested by a sensor and uploading the vibration signals to a signal conditioning module; s2, preprocessing the vibration signal by the signal conditioning module and the data acquisition card; s3, uploading the signals to a calculation processing center by a data acquisition card, and performing data format conversion on the vibration signals; s4: the calculation processing center utilizes a VMD-TEO energy entropy algorithm to perform feature extraction on input data, utilizes a GA-SVM algorithm to classify the input data, and judges the running state of the drill bit; s5: when the drill bit breaks down, the control system sends out an instruction to stop the drill bit. The system comprises a piezoelectric acceleration sensor, a signal conditioning module, a data acquisition card and a calculation processing center; the piezoelectric acceleration sensor collects vibration signals of a drill column of a tested drill bit and uploads the vibration signals to the signal conditioning module, and the signal conditioning module and the data acquisition card preprocess the signals and upload the signals to the calculation processing center through the serial port communication module.

Description

Drill bit vibration signal online virtual test method and system thereof
Technical Field
The invention relates to the technical field of drill bit abrasion detection in drilling in industries such as petroleum, mine, geological exploration and the like, in particular to a drill bit vibration signal online virtual test method and a system thereof.
Background
The reduction of the drilling cost is always an important problem in industries such as petroleum, mines, geological exploration and the like, most drilling bits are made of diamonds and are expensive, and therefore, how to detect and control the abrasion of the drilling bits to reduce the drilling cost is a key factor for reducing the drilling cost. The invention is mainly applied to detecting the abrasion degree of the drill bit in the drilling process. In past engineering application, empirical modes are mostly adopted for testing the drill bit, and the modes have low accuracy and low detection efficiency. The method detects the vibration signals of the drill bit by using a LabVIEW and MATLAB method, and trains the VMD-TEO energy entropy support vector machine algorithm by using data of different abrasion degrees in the running process of the drill bit so as to quickly judge the quality of the drill bit.
The drill bit fault detection device widely used at present is a traditional hardware testing instrument, and most of the devices adopt a hardware circuit to process vibration signals and then display the waveforms of the vibration signals. The waveform displayed needs to be judged by a professional inspector, the requirement of the drill bit inspection method on the inspector is high, and the inspection efficiency and accuracy can be influenced by the inspector. The traditional test instrument mostly adopts a closed structure, is difficult to change and expand, and has poor adaptability to different objects. The virtual test system is developed by combining LabVIEW and MATLAB, and can effectively combine an intelligent algorithm and a virtual instrument technology to realize fault diagnosis. The virtual test system adopts software design, and has better flexibility for processing signals; the test system integrates multiple functions of signal acquisition, conversion, data processing, intelligent judgment, result display and the like, only software needs to be modified when different test objects are faced, and the use flexibility, the test efficiency and the accuracy of the test system are greatly improved.
Currently, for the field of mechanical fault diagnosis, an Empirical Mode Decomposition (EMD) method is mostly used to extract feature quantities. EMD is a self-adaptive signal processing method, and can decompose a fault signal into a plurality of IMF modal components and then extract fault features by utilizing the IMF components obtained by decomposition. Because EMD is a recursive modal decomposition method, in the decomposition process, due to the propagation characteristic of the envelope estimation error, the error is continuously accumulated, and modal aliasing occurs in the decomposition result.
The Variational Mode Decomposition (VMD) is a new signal processing method compared to the EMD, and has a solid theoretical basis. The VMD adopts a non-recursive decomposition mode, and the method determines the center frequency and the bandwidth of each modal component in a mode of iteratively searching the optimal solution of the variation model, so that the signals can be adaptively split in the frequency domain and each component can be effectively separated, the modal aliasing phenomenon existing in the EMD is overcome, and the decomposition has better noise robustness and modal separation effect.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a drill bit vibration signal online virtual test method and a system thereof, and aims to solve the problems of complex acquisition and detection device, high requirement on an operator, poor use flexibility and the like of a drill bit fault detection device in the prior art. The invention decomposes the fault signal by using a VMD method, overcomes the modal aliasing phenomenon existing in EMD, and ensures that the decomposition has better noise robustness and modal separation effect.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
an online virtual test method for a drill vibration signal comprises the following steps:
s1, arranging a sensor on the drill bit to be measured, connecting the sensor with a signal conditioning module, and acquiring a vibration signal of the drill bit to be measured by the sensor and uploading the vibration signal to the signal conditioning module;
s2, the signal conditioning module is connected with the data acquisition card, and the signal conditioning module transmits the data to the data acquisition card after filtering and amplifying; the data acquisition card processes the signals and converts the acquired analog signals into digital signals;
s3, connecting a data acquisition card with a serial communication module and connecting the data acquisition card with a calculation processing center through the serial communication module, wherein a drill bit vibration signal software system developed based on LabVIEW and MATLAB is arranged in the calculation processing center, and the data acquisition card transmits a digital signal to the calculation processing center through the serial communication module and the digital signal is processed by the calculation processing center;
s4: the computing processing center preprocesses and stores the original data and converts the data format; generating input data; under a LabVIEW platform, an MATLABScript node is utilized to call a VMD-TEO support vector machine algorithm in MATLAB to classify input data;
firstly, decomposing input data by a Variational Modal Decomposition (VMD) method; screening out the optimal five signal components by utilizing a kurtosis criterion, calculating a Teager Energy Operator (TEO) of the five components, and then calculating the TEO energy entropy to form a TEO energy entropy matrix; taking the TEO energy entropy matrix as the input of a genetic algorithm optimized support vector machine (GA-SVM), and classifying the TEO energy entropy matrix by utilizing the GA-SVM; judging the running state of the drill bit;
s5: the calculation processing center is externally connected with a control system through a serial port communication module, the control system is connected with a drill bit to be tested, and when the drill bit fails, the control system sends an instruction to stop the drill bit from running; otherwise, the drill bit continues to run until the stop button is pressed.
As an improvement to the above technical solution, the VMD-TEO support vector machine algorithm of step S4 includes:
s41, reading input data, taking 2048 signal points to form an input signal x (t), and initializing parameters of the VMD, where k is 2, α is 2000, and τ is 0;
s42, carrying out variation modal decomposition on the input signal x (t) to obtain each IMF component ui(t) calculating the center frequency of each component, and judging whether the adjacent center frequencies are similar; if the similarity is found, determining the mode number to be k-1, and if the similarity is not found, repeating the step by using the mode number to be k + 1;
s43 based on kurtosis criterion
Figure GDA0002395724270000041
Calculating kurtosis values of IMF components after VMD decomposition, and screening five IMF components with the maximum kurtosis value;
s44, calculating the five IMF components screened in the step S43 by using a text operator formula of a discrete signal, wherein the text operator formula of each componentInstantaneous amplitude A and instantaneous amplitude omegac：
ψD[x(tq)]＝x2(tq)-x(tq-1)x(tq+1)
Figure GDA0002395724270000042
Figure GDA0002395724270000043
Each IMF component u is then calculatediInstantaneous energy spectrum T ofiAnd the energy entropy probability distribution pi：
Ti(t)＝ψ[ui(t)]＝(Ai(t)ωi(t))2，(i＝1,2,…,5)
Figure GDA0002395724270000044
Figure GDA0002395724270000045
The TEO energy entropy is therefore:
Figure GDA0002395724270000046
thus forming a five-dimensional feature matrix e1,e2,e3,e4,e5}；
And S45, constructing a GA-SVM classifier, and classifying the five-dimensional feature matrix output in the step S44.
As a further improvement to the above solution, the VMD algorithm of step S41 includes:
s411, a Variational Modal Decomposition (VMD), wherein the VMD algorithm decomposes the input signal into a plurality of Intrinsic Mode Functions (IMFs) under a variational framework, and different IMFs are concentrated on corresponding central frequencies (omega)k) Nearby, the center frequency and bandwidth of each IMF component are solvedContinuously updating in the process;
the variation problem is as follows:
Figure GDA0002395724270000051
Figure GDA0002395724270000052
in the above formula: f represents the input signal: x (t); { uk}＝{u1,u2,…,ukThe method comprises the steps of (1) decomposing an input signal to obtain k IMF components; { omega [ [ omega ] ]k}＝{ω1,ω2,…,ωk-is the center frequency of each IMF component;
the constraint condition can be kept strict by utilizing a Lagrange multiplication operator lambda (t), and the augmented Lagrange function is expressed as follows:
Figure GDA0002395724270000053
alternate update with alternate multiplier Algorithm (AMDD) for the above equation
Figure GDA0002395724270000054
And
Figure GDA0002395724270000055
the method is used for solving the optimal solution of the augmented Lagrange function, and comprises the following specific steps:
a initialization
Figure GDA0002395724270000056
And n;
b, executing a cycle: n is n +1
c updating according to
Figure GDA0002395724270000057
And ωk：
Figure GDA0002395724270000058
Figure GDA0002395724270000059
d, updating lambda by using the following formula:
Figure GDA0002395724270000061
e, judging whether the iteration customization condition is met:
Figure GDA0002395724270000062
if not, returning to the step b, repeating the steps b, c and d;
if the condition is satisfied, the output of the end iteration is
Figure GDA0002395724270000063
Then converting the frequency domain signal into a time domain signal u through Fourier transformk(t) to obtain k IMF components { u } respectively1(t),u2(t),…,uk(t)}。
As a further improvement to the above solution, the GA-SVM classifier construction process of step S45 is:
s451, constructing a GA-SVM classifier model, firstly, training the constructed GA-SVM model, constructing a training set and a testing set for the GA-SVM model, respectively taking 100 groups of data of the drill bit under normal state, medium abrasion and severe abrasion, wherein the signal point number of each group of data is 2048, and performing VMD-TEO energy entropy feature extraction on the data to obtain 300 groups of feature matrixes of 1 × 5, wherein the normal state is marked with '1', the medium abrasion is marked with '2', and the severe abrasion is marked with '3'. 70 groups of data are taken to form a training set in each state, and the other 90 groups are taken as a test set to train the GA-SVM model. And saving the model for the trained GA-SVM classifier.
As a further improvement to the above scheme, the invention also provides an online virtual test system for the vibration signal of the drill bit, which comprises a piezoelectric acceleration sensor, a signal conditioning module, a data acquisition card and a calculation processing center which are sequentially connected; the piezoelectric acceleration sensor is arranged on a drill column of a tested drill bit, collects vibration signals of the drill column of the tested drill bit and uploads the vibration signals to the signal conditioning module, the signal conditioning module filters and amplifies the vibration signals collected by the piezoelectric acceleration sensor and then transmits the vibration signals to the data acquisition card, the data acquisition card converts analog signals transmitted by the signal conditioning module into digital signals and uploads the digital signals to the calculation processing center through the serial port communication module, the calculation processing center comprises a data processing module, a drill bit vibration signal software system developed based on a LabVIEW platform is arranged in the data processing module, the digital signals uploaded by the serial port communication module are preprocessed and stored, data formats are converted and decomposed, the optimal five signal components are screened out according to kurtosis criteria, and TEO energy entropies of the five components are calculated, and forming a TEO energy entropy matrix. In order to ensure the accuracy of the test signal of the sensor, the monitoring effect of the measuring point on the drill bit drill string is good.
As a further improvement to the above scheme, the drill vibration signal online virtual test system further comprises a serial communication module, wherein the serial communication module comprises a serial initialization unit and a data reading unit; the data input into the calculation processing center firstly enter a serial communication module, and the serial communication module is divided into a serial initialization unit and a data reading unit; before the serial port initialization unit operates, data initialization is carried out on the resource name, the baud rate, the data bit, the parity check, the stop stream and the data stream corresponding to the VISA serial port, and the data reading unit reads data transmitted by the lower computer through the serial port communication module.
As a further improvement to the above scheme, the data processing module is divided into a data generation unit and an intelligent algorithm unit; the data generation unit firstly converts the format of the vibration data to generate a usable data format of matlab; the intelligent algorithm unit is used for extracting and classifying the characteristics of input data by utilizing a VMD-TEO support vector machine algorithm so as to judge the running state of the drill bit.
As a further improvement of the scheme, the drill bit vibration signal online virtual test system further comprises a control system, and the control system is connected with the drill bit to be tested and is connected with the computing processing center through a serial port communication module. The control system is any control system known in the art.
As a further improvement to the scheme, the online virtual testing system for the drill bit vibration signals further comprises a data monitoring module, and the data monitoring module comprises a data storage unit, a data query unit, a result display unit and an early warning unit.
Compared with the prior art, the invention has the following beneficial effects:
the drill bit vibration signal online virtual test system is developed by using a LabVIEW platform, the advantages of a virtual instrument are fully utilized, the use flexibility of the test system is improved, the volume of the test system is reduced, the test system is more convenient to carry, and the test can be carried out on an engineering site. In addition, the vibration data are classified by adopting a VMD-TEO support vector machine algorithm in MATLAB called by a MATLABScript node in a LabVIEW platform, the method has obvious advantages, improves the intelligent degree of the drill vibration signal test system, reduces the requirements on a test device operator, enables non-professionals to operate the test system quickly and skillfully, and greatly improves the test efficiency and accuracy. Specifically, the method comprises the following steps: 1. simple and practical structure, small volume, convenient carrying, low maintenance workload and low cost. 2. The test system uses an intelligent algorithm, so that the intelligent degree of the system is improved, and the requirement on a user is reduced. 3. The working efficiency and accuracy of the test system are improved. 4. The test system has high use flexibility and strong applicability, can flexibly change the software program setting according to the different characteristics of the tested device, and is convenient and fast to change the software program.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of the system components of the present invention;
FIG. 2 is a schematic diagram of a computing processing center and a serial communication module;
FIG. 3 is a flow chart of the system operation;
FIG. 4 is a flow chart of a VMD-TEO energy operator based support vector machine algorithm;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived from the embodiments of the present invention by a person skilled in the art without any creative effort, should be included in the protection scope of the present invention.
As shown in fig. 1, 2, 3 and 4, the drill vibration signal online virtual test method of the invention comprises the following steps:
s1, arranging a piezoelectric acceleration sensor 2 above a drill string of the tested drill bit 1, connecting the piezoelectric acceleration sensor 2 with a signal conditioning module 3, and acquiring a vibration signal of the drill string of the tested drill bit 1 by the piezoelectric acceleration sensor 2 and uploading the vibration signal to the signal conditioning module 3;
s2, the signal conditioning module 3 is connected with the data acquisition card 4, and the signal conditioning module 3 transmits the data to the data acquisition card 4 after filtering and amplifying; the data acquisition card 4 processes the signals and converts the acquired analog signals into digital signals;
s3, connecting the data acquisition card 4 with the calculation processing center 6 through the serial communication module 5, wherein the calculation processing center 6 is internally provided with a drill bit vibration signal software system developed based on LabVIEW and MATLAB, and the data acquisition card 4 transmits the digital signal to the calculation processing center 6 through the serial communication module 5 and is processed by the data processing module of the calculation processing center 6;
s4: the calculation processing center 6 preprocesses and stores the original data and converts the data format; generating input data; and calling a VMD-TEO support vector machine algorithm in MATLAB by using a MATLABScript node under a LabVIEW platform to classify the input data. Decomposing data by a Variational Modal Decomposition (VMD) method; screening out the optimal five signal components by utilizing a kurtosis criterion, calculating a Teager Energy Operator (TEO) of the five components, and then calculating the TEO energy entropy to form a TEO energy entropy matrix; taking the TEO energy entropy matrix as the input of a genetic algorithm optimized support vector machine (GA-SVM), and classifying the TEO energy entropy matrix by utilizing the GA-SVM; judging the running state of the drill bit;
s5: the calculation processing center 6 is externally connected with a control system 8 through a serial port communication module 7, the control system 8 is connected with the drill bit 1 to be tested, and when the drill bit fails, the control system sends out an instruction to stop the drill bit from running; otherwise, the drill bit continues to run until the stop button is pressed.
The VMD-TEO support vector machine algorithm of the step S4 comprises the following steps:
s41, reading input data, taking 2048 signal points to form an input signal x (t), and initializing parameters of the VMD, where k is 2, α is 2000, and τ is 0;
s42, carrying out variation modal decomposition on the input signal x (t) to obtain each IMF component ui(t) calculating the center frequency of each component, and judging whether the adjacent center frequencies are similar; if the similarity is found, determining the mode number to be k-1, and if the similarity is not found, repeating the step by using the mode number to be k + 1;
s43 based on kurtosis criterion
Figure GDA0002395724270000101
Calculating kurtosis values of IMF components after VMD decomposition, and screening five IMF components with the maximum kurtosis value;
s44, for the five IMF components screened in the step S43, using the text of the discrete signalr energy operator formula calculation, instantaneous amplitude A and instantaneous amplitude omega of each componentc：
ψD[x(tq)]＝x2(tq)-x(tq-1)x(tq+1)
Figure GDA0002395724270000102
Figure GDA0002395724270000103
Each IMF component u is then calculatediInstantaneous energy spectrum T ofiAnd energy EiProbability distribution P ofi：
Ti(t)＝ψ[ui(t)]＝(Ai(t)ωi(t))2，(i＝1,2,…,5)
Figure GDA0002395724270000104
Figure GDA0002395724270000105
The TEO energy entropy is therefore:
Figure GDA0002395724270000111
thus forming a five-dimensional feature matrix e1,e2,e3,e4,e5}；
And S45, constructing a GA-SVM classifier, and classifying the five-dimensional feature matrix output in the step S44.
S411, a Variational Modal Decomposition (VMD), wherein the VMD algorithm decomposes the input signal into a plurality of Intrinsic Mode Functions (IMFs) under a variational framework, and different IMFs are concentrated on corresponding central frequencies (omega)k) Nearby, the center frequency and the bandwidth of each IMF component are continuously updated in the process of solving the variation model;
the variation problem is as follows:
Figure GDA0002395724270000112
Figure GDA0002395724270000113
in the above formula: f represents the input signal: x (t); { uk}＝{u1,u2,…,ukThe method comprises the steps of (1) decomposing an input signal to obtain k IMF components; { omega [ [ omega ] ]k}＝{ω1,ω2,…,ωk-is the center frequency of each IMF component;
the constraint condition can be kept strict by utilizing a Lagrange multiplication operator lambda (t), and the augmented Lagrange function is expressed as follows:
Figure GDA0002395724270000114
alternate update with alternate multiplier Algorithm (AMDD) for the above equation
Figure GDA0002395724270000115
And
Figure GDA0002395724270000116
the method is used for solving the optimal solution of the augmented Lagrange function, and comprises the following specific steps:
a initialization
Figure GDA0002395724270000117
And n;
b, executing a cycle: n is n +1
c updating according to
Figure GDA0002395724270000121
And ωk：
Figure GDA0002395724270000122
Figure GDA0002395724270000123
d, updating lambda by using the following formula:
Figure GDA0002395724270000124
e, judging whether the iteration customization condition is met:
Figure GDA0002395724270000125
if not, returning to the step b, repeating the steps b, c and d;
if the condition is satisfied, the output of the end iteration is
Figure GDA0002395724270000126
Then converting the frequency domain signal into a time domain signal u through Fourier transformk(t) to obtain k IMF components { u } respectively1(t),u2(t),…,uk(t)}。
The construction process of the GA-SVM classifier of the step S45 is as follows:
s451, constructing a GA-SVM classifier model, firstly, training the constructed GA-SVM model, constructing a training set and a testing set for the GA-SVM model, respectively taking 100 groups of data of the drill bit under a normal state, medium abrasion and severe abrasion, wherein the signal point number of each group of data is 2048, and performing VMD-TEO energy entropy feature extraction on the data to obtain 300 groups of feature matrixes of 1 × 5, wherein the normal state is marked with '1', the medium abrasion is marked with '2', and the severe abrasion is marked with '3'; in each state, 70 groups of data are taken to form a training set, and the other 90 groups of data are taken as a test set to train the GA-SVM model; and saving the model for the trained GA-SVM classifier.
The invention also provides an online virtual test system for the vibration signals of the drill bit, which comprises a piezoelectric acceleration sensor 2, a signal conditioning module 3, a data acquisition card 4 and a calculation processing center 6 which are connected in sequence; the piezoelectric acceleration sensor 2 is arranged on a drill string of the drill bit 1 to be detected, collects vibration signals of the drill string of the drill bit to be detected and uploads the vibration signals to the signal conditioning module 3, the signal conditioning module 3 filters and amplifies the vibration signals collected by the piezoelectric acceleration sensor 2 and then transmits the vibration signals to the data acquisition card 4, the data acquisition card 4 converts analog signals transmitted by the signal conditioning module 3 into digital signals and uploads the digital signals to the calculation processing center through the serial communication module 5, the calculation processing center 6 is internally provided with a data processing module of a drill bit vibration signal software system developed based on a LabVIEW platform, the digital signals uploaded by the serial communication module 5 are preprocessed and stored, data formats are converted and decomposed, optimal five signal components are screened out through kurtosis criteria, and TEO energy entropies of the five components are calculated, and forming a TEO energy entropy matrix. The drill bit vibration signal online virtual test system further comprises a control system 8, wherein the control system 8 is connected with the drill bit 1 to be tested and is connected with the calculation processing center 6 through a serial port communication module 7.
The data input into the calculation processing center 6 firstly enter the serial communication module 5, and the serial communication module 5 is divided into a serial initialization unit and a data reading unit; before the serial port initialization unit operates, data initialization is carried out on the resource name, the baud rate, the data bit, the parity check, the stop stream and the data stream corresponding to the VISA serial port, and the data reading unit reads data transmitted by the lower computer through the serial port communication module.
The data processing module is divided into a data generating unit and an intelligent algorithm unit; the data generation unit firstly converts the format of the vibration data to generate a usable data format of matlab; the intelligent algorithm unit is used for extracting and classifying the characteristics of input data by utilizing a VMD-TEO support vector machine algorithm so as to judge the running state of the drill bit.
The drill bit vibration signal online virtual test system further comprises a data monitoring module, and the data monitoring module comprises a data storage unit, a data query unit, a result display unit and an early warning unit. The data monitoring module can store and query data, display the classification result of the data processing module and give early warning. The early warning unit contains three pilot lamps, and bright green lamp when the drill bit is slight wearing and tearing, bright yellow lamp when moderate wearing and tearing, bright red lamp when serious wearing and tearing. If the drill bit is seriously worn, an alarm is given and a control command is automatically sent by a control module of the software to control a controller of the drill bit so as to stop the drill bit.
The drill bit vibration signal online virtual test system is developed by using a LabVIEW platform, the advantages of a virtual instrument are fully utilized, the use flexibility of the test system is improved, the volume of the test system is reduced, the test system is more convenient to carry, and the test can be carried out on an engineering site. In addition, the vibration data are classified by adopting a VMD-TEO energy entropy-based support vector machine algorithm in MATLAB called by a MATLABScript node in a LabVIEW platform, the method has obvious advantages, improves the intelligent degree of a drill bit vibration signal test system, reduces the requirement on a test device operator, enables non-professionals to operate the test system quickly and skillfully, and greatly improves the test efficiency and accuracy. Specifically, the method comprises the following steps: 1. simple and practical structure, small volume, convenient carrying, low maintenance workload and low cost. 2. The test system uses an intelligent algorithm, so that the intelligent degree of the system is improved, and the requirement on a user is reduced. 3. The working efficiency and accuracy of the test system are improved. 4. The test system has high use flexibility and strong applicability, can flexibly change the software program setting according to the different characteristics of the tested device, and is convenient and fast to change the software program.

Claims (8)

1. An online virtual test method for a drill vibration signal is characterized by comprising the following steps:
s1, arranging a sensor on the drill bit to be measured, connecting the sensor with a signal conditioning module, and acquiring a vibration signal of the drill bit to be measured by the sensor and uploading the vibration signal to the signal conditioning module;
s2, the signal conditioning module is connected with the data acquisition card, and the signal conditioning module transmits the data to the data acquisition card after filtering and amplifying; the data acquisition card processes the signals and converts the acquired analog signals into digital signals;
s3, connecting a data acquisition card with a serial communication module and connecting the data acquisition card with a calculation processing center through the serial communication module, wherein a drill bit vibration signal software system developed based on LabVIEW and MATLAB is arranged in the calculation processing center, and the data acquisition card transmits a digital signal to the calculation processing center through the serial communication module and the digital signal is processed by the calculation processing center;
s4: the computing processing center preprocesses and stores the original data and converts the data format; generating input data; under a LabVIEW platform, an MATLABScript node is utilized to call a VMD-TEO support vector machine algorithm in MATLAB to classify input data;
firstly, decomposing input data by a variational modal decomposition method; screening out the optimal five signal components by utilizing a kurtosis criterion, calculating a teager energy operator of the five components, and then calculating TEO energy entropy to form a TEO energy entropy matrix; taking the TEO energy entropy matrix as the input of a support vector machine optimized by a genetic algorithm, and classifying the TEO energy entropy matrix by using the support vector machine; judging the running state of the drill bit;
the VMD-TEO support vector machine algorithm comprises the following steps:
s41, reading input data, taking 2048 signal points to form an input signal x (t), and initializing parameters of the VMD, where k is 2, α is 2000, and τ is 0;
s42, carrying out variation modal decomposition on the input signal x (t) to obtain each IMF component ui(t) calculating the center frequency of each component, and judging whether the adjacent center frequencies are similar; if the similarity is found, determining the mode number to be k-1, and if the similarity is not found, repeating the step by using the mode number to be k + 1;
s43 based on kurtosis criterion
Figure FDA0002395724260000021
Calculating kurtosis values of IMF components after VMD decomposition, and screening five IMF components with the maximum kurtosis value;
s44, for the five screened in step S43The IMF components are calculated by using TEO energy operator formula of discrete signal, and the instantaneous amplitude A and the instantaneous amplitude omega of each componentc：
ψD[x(tq)]＝x2(tq)-x(tq-1)x(tq+1)
Figure FDA0002395724260000022
Figure FDA0002395724260000023
Each IMF component u is then calculatediInstantaneous energy spectrum T ofiAnd energy EiProbability distribution P ofi：
Ti(t)＝ψ[ui(t)]＝(Ai(t)ωi(t))2，(i＝1,2,…,5)
Figure FDA0002395724260000024
Figure FDA0002395724260000025
The TEO energy entropy is therefore:
Figure FDA0002395724260000026
thus forming a five-dimensional feature matrix e1,e2,e3,e4,e5}；
S45, constructing a GA-SVM classifier, and classifying the five-dimensional feature matrix output in the step S44;
s5: the calculation processing center is externally connected with a control system through a serial port communication module, the control system is connected with a drill bit to be tested, and when the drill bit fails, the control system sends an instruction to stop the drill bit from running; otherwise, the drill bit continues to run until the stop button is pressed.
2. The online virtual drill bit vibration signal testing method according to claim 1, wherein the VMD algorithm of step S41 comprises the steps of:
s411, performing variational modal decomposition, wherein a variational modal decomposition algorithm decomposes an input signal into a plurality of intrinsic modal functions under a variational framework, and different intrinsic modal functions are concentrated in corresponding central frequencies omegakNearby, the center frequency and the bandwidth of each IMF component are continuously updated in the process of solving the variation model;
the variation problem is as follows:
Figure FDA0002395724260000031
Figure FDA0002395724260000032
in the above formula: f represents the input signal: x (t); { uk}＝{u1,u2,…,ukThe method comprises the steps of (1) decomposing an input signal to obtain k IMF components; { omega [ [ omega ] ]k}＝{ω1,ω2,…,ωk-is the center frequency of each IMF component;
the constraint condition can be kept strict by utilizing a Lagrange multiplication operator lambda (t), and the augmented Lagrange function is expressed as follows:
Figure FDA0002395724260000033
alternate update with alternate multiplier Algorithm (AMDD) for the above equation
Figure FDA0002395724260000034
And
Figure FDA0002395724260000035
the method is used for solving the optimal solution of the augmented Lagrange function, and comprises the following specific steps:
a initialization
Figure FDA0002395724260000036
And n;
b, executing a cycle: n is n +1
c updating according to
Figure FDA0002395724260000037
And ωk：
Figure FDA0002395724260000038
Figure FDA0002395724260000041
d, updating lambda by using the following formula:
Figure FDA0002395724260000042
e, judging whether the iteration customization condition is met:
Figure FDA0002395724260000043
if not, returning to the step b, repeating the steps b, c and d;
if the condition is satisfied, the output of the end iteration is
Figure FDA0002395724260000044
Then converting the frequency domain signal into a time domain signal u through Fourier transformk(t) to obtain k IMF components { u } respectively1(t),u2(t),…,uk(t)}。
3. The on-line virtual drill vibration signal testing method as claimed in claim 2, wherein the GA-SVM classifier constructing process of step S45 is:
s451, constructing a GA-SVM classifier model, firstly, training the constructed GA-SVM model, constructing a training set and a testing set for the GA-SVM model, respectively taking 100 groups of data of the drill bit under a normal state, medium abrasion and severe abrasion, wherein the signal point number of each group of data is 2048, and performing VMD-TEO energy entropy feature extraction on the data to obtain 300 groups of feature matrixes of 1 × 5, wherein the normal state is marked with '1', the medium abrasion is marked with '2', and the severe abrasion is marked with '3'; in each state, 70 groups of data are taken to form a training set, and the other 90 groups of data are taken as a test set to train the GA-SVM model; and saving the model for the trained GA-SVM classifier.
4. The on-line virtual test system for the drill vibration signals is characterized by comprising a piezoelectric acceleration sensor, a signal conditioning module, a data acquisition card and a calculation processing center which are sequentially connected; the piezoelectric acceleration sensor is arranged above a drill column of a tested drill bit, collects vibration signals of the drill column of the tested drill bit and uploads the vibration signals to the signal conditioning module, the signal conditioning module filters and amplifies the vibration signals collected by the piezoelectric acceleration sensor and then transmits the vibration signals to the data acquisition card, the data acquisition card converts analog signals transmitted by the signal conditioning module into digital signals and uploads the digital signals to the calculation processing center through the serial port communication module, the calculation processing center comprises a data processing module, a drill bit vibration signal software system developed based on a LabVIEW platform is arranged in the data processing module, the digital signals uploaded by the serial port communication module are preprocessed and stored, data formats are converted and decomposed, the optimal five signal components are screened out according to kurtosis criteria, and TEO energy entropies of the five components are calculated, and forming a TEO energy entropy matrix.
5. The on-line virtual test system for the drill bit vibration signal as claimed in claim 4, further comprising a control system, wherein the control system is connected with the drill bit to be tested and connected with the computing processing center through a serial communication module.
6. The online virtual test system for the drill bit vibration signal as recited in claim 4, wherein the data processing module is divided into a data generation unit and an intelligent algorithm unit; the data generation unit firstly converts the format of the vibration data to generate a usable data format of matlab; the intelligent algorithm unit is used for extracting and classifying the characteristics of input data by utilizing a VMD-TEO support vector machine algorithm so as to judge the running state of the drill bit.
7. The online virtual test system for the drill bit vibration signal as recited in claim 4, further comprising a serial communication module, wherein the serial communication module comprises a serial initialization unit and a data reading unit; the data input into the calculation processing center firstly enter a serial communication module, and the serial communication module is divided into a serial initialization unit and a data reading unit; before the serial port initialization unit operates, data initialization is carried out on the resource name, the baud rate, the data bit, the parity check, the stop stream and the data stream corresponding to the VISA serial port, and the data reading unit reads data transmitted by the lower computer through the serial port communication module.
8. The on-line virtual test system for the drill bit vibration signal as recited in claim 4, further comprising a data monitoring module, wherein the data monitoring module comprises a data storage unit, a data query unit, a result display unit and an early warning unit.
CN201811295029.1A 2018-11-01 2018-11-01 Drill bit vibration signal online virtual test method and system thereof Expired - Fee Related CN109443719B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811295029.1A CN109443719B (en) 2018-11-01 2018-11-01 Drill bit vibration signal online virtual test method and system thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811295029.1A CN109443719B (en) 2018-11-01 2018-11-01 Drill bit vibration signal online virtual test method and system thereof

Publications (2)

Publication Number Publication Date
CN109443719A CN109443719A (en) 2019-03-08
CN109443719B true CN109443719B (en) 2020-05-19

Family

ID=65550637

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811295029.1A Expired - Fee Related CN109443719B (en) 2018-11-01 2018-11-01 Drill bit vibration signal online virtual test method and system thereof

Country Status (1)

Country Link
CN (1) CN109443719B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110161125A (en) * 2019-06-17 2019-08-23 哈尔滨工业大学 The Aeroengine Smart monitoring method combined based on acceleration with sound emission cognition technology
CN110542828A (en) * 2019-07-17 2019-12-06 佛山电力设计院有限公司 Distribution network fault positioning method and system
CN111191502B (en) * 2019-11-22 2023-04-18 中国地质大学（武汉） Stick-slip and jump drill abnormal working condition identification method based on drill string vibration signal
CN112765945A (en) * 2021-01-26 2021-05-07 徐州徐工矿业机械有限公司 Mechanical vibration data processing system and method based on virtual instrument
CN113887322A (en) * 2021-09-08 2022-01-04 安徽建筑大学 LED lamp power supply driving fault diagnosis method and system based on light output time-frequency characteristics

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000036592A1 (en) * 1998-12-16 2000-06-22 Sarnoff Corporation Improved noise spectrum tracking for speech enhancement
CN202049027U (en) * 2011-01-21 2011-11-23 天津工业大学 Tester for spindle vibration testing based on LabVIEW (laboratory virtual instrumentation engineering workbench)
CN104634571A (en) * 2015-02-06 2015-05-20 北京航空航天大学 Fault diagnosis method for rolling bearing based on LCD-MF (Local Characteristic Scale Decomposition )-(Multifractal)
CN107831012A (en) * 2017-10-11 2018-03-23 温州大学 A kind of Method for Bearing Fault Diagnosis based on Walsh conversion with Teager energy operators
CN107944199A (en) * 2017-12-22 2018-04-20 浙江工业大学 A kind of gearbox fault recognition methods based on spectral trends and variation mode decomposition

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108597540A (en) * 2018-04-11 2018-09-28 南京信息工程大学 A kind of speech-emotion recognition method based on variation mode decomposition and extreme learning machine

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000036592A1 (en) * 1998-12-16 2000-06-22 Sarnoff Corporation Improved noise spectrum tracking for speech enhancement
CN202049027U (en) * 2011-01-21 2011-11-23 天津工业大学 Tester for spindle vibration testing based on LabVIEW (laboratory virtual instrumentation engineering workbench)
CN104634571A (en) * 2015-02-06 2015-05-20 北京航空航天大学 Fault diagnosis method for rolling bearing based on LCD-MF (Local Characteristic Scale Decomposition )-(Multifractal)
CN107831012A (en) * 2017-10-11 2018-03-23 温州大学 A kind of Method for Bearing Fault Diagnosis based on Walsh conversion with Teager energy operators
CN107944199A (en) * 2017-12-22 2018-04-20 浙江工业大学 A kind of gearbox fault recognition methods based on spectral trends and variation mode decomposition

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于EEMD-TEO熵的高速列车故障诊断方法;靳行等;《西南交通大学学报》;20180430;第53卷(第2期);第359-366页 *
基于变分模态分解和Teager能量算子的滚动轴承故障特征提取;马增强;《振动与冲击》;20160731;第35卷(第13期);第134-139页 *

Also Published As

Publication number Publication date
CN109443719A (en) 2019-03-08

Similar Documents

Publication Publication Date Title
CN109443719B (en) Drill bit vibration signal online virtual test method and system thereof
CN102809493B (en) Abnormal sound diagnosis device
CN104132250B (en) A kind of pipe leakage characteristic vector pickup method based on improvement wavelet packet
CN102488518B (en) Electroencephalogram detection method and device by utilizing fluctuation index and training for promotion
CN109489977B (en) KNN-AdaBoost-based bearing fault diagnosis method
CN110018670A (en) A kind of industrial process unusual service condition prediction technique excavated based on dynamic association rules
CN103720468B (en) Be applied to artifact identification method and the device of dynamic electrocardiogram (ECG) data
CN107095669A (en) A kind of processing method and system of epileptic's EEG signals
CN105139029A (en) Activity recognition method and activity recognition device for persons serving sentences
US20180177415A1 (en) Cardiovascular disease detection
CN108830328A (en) Merge the microseismic signals SMOTE recognition methods and monitoring system of spatial knowledge
CN116517860A (en) Ventilator fault early warning system based on data analysis
CN114167838A (en) Multi-scale health assessment and fault prediction method for servo system
CN110956155B (en) Fuzzy clustering identification method for fully mechanized mining face operation procedure based on CO data
CN110765983A (en) Coal pillar type rock burst risk evaluation method based on microseismic frequency spectrum characteristics
CN107367552A (en) Damage propatagtion method of real-time based on FCM algorithms under the influence of time varying temperature
CN103234746A (en) Device and method for online diagnosing faults of wind turbine generator gear case
CN106708009A (en) Ship dynamic positioning measurement system multiple-fault diagnosis method based on support vector machine clustering
CN116186946A (en) Hydraulic system fault diagnosis method and system based on diagnosis model
CN109558041B (en) GPU acceleration-based blade tip clearance signal acquisition, processing and transmission method
CN202596709U (en) Oil well fault intelligent diagnosing device
CN114611548A (en) State classification method and system based on vibration analysis
CN116523722A (en) Environment monitoring analysis system with machine learning capability
CN102494717A (en) Engineering machinery fault diagnosis system based on sensor data wireless transmission
CN114492146B (en) Bolt group loosening positioning and quantitative analysis method and system based on transfer learning

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200519

Termination date: 20201101

CF01 Termination of patent right due to non-payment of annual fee