CN109443719A - A kind of drill vibration signal on-line virtual testing method and its system - Google Patents
A kind of drill vibration signal on-line virtual testing method and its system Download PDFInfo
- Publication number
- CN109443719A CN109443719A CN201811295029.1A CN201811295029A CN109443719A CN 109443719 A CN109443719 A CN 109443719A CN 201811295029 A CN201811295029 A CN 201811295029A CN 109443719 A CN109443719 A CN 109443719A
- Authority
- CN
- China
- Prior art keywords
- data
- signal
- vibration signal
- drill
- drill bit
- 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.)
- Granted
Links
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 kind of drill vibration signal on-line virtual testing method and its system, the step of method is: S1, sensor acquire the vibration signal and up-delivering signal conditioning module of tested drill bit；S2, signal conditioning module and data collecting card pre-process vibration signal；Signal is uploaded to calculation processing center by S3, data collecting card, carries out data format conversion to vibration signal；S4: calculation processing center carries out feature extraction to input data using VMD-TEO Energy-Entropy algorithm and judges the operating status of drill bit using GA-SVM algorithm classification；S5: when drill bit breaks down, control system issues instruction, keeps drill bit out of service.Its system includes piezoelectric acceleration sensor, signal conditioning module, data collecting card, calculation processing center；Piezoelectric acceleration sensor acquires the vibration signal and up-delivering signal conditioning module of tested drill bit drill string, and signal conditioning module and data collecting card upload calculation processing center to Signal Pretreatment, and by serial communication modular.
Description
Technical field
The present invention relates to the technical fields of the bit wear of the industries drilling well such as petroleum, mine, geological prospecting detection, specifically
It says and is related to a kind of drill vibration signal on-line virtual testing method and its system.
Background technique
Reducing drilling cost is always the major issue that the industries such as petroleum, mine, geological prospecting face, and drilling bit is most
It is all to be made of diamond, it is expensive, therefore it is to reduce to bore that how the abrasion of Detection & Controling drill bit, which reduces drill bit cost,
The key factor of well cost.Present invention is mainly applied to detect the degree of wear of the drill bit in drilling process.In previous work
In Cheng Yingyong, the mode of experience is mostly used for the test of drill bit, these mode accuracys rate are low, and detection efficiency is low.The present invention is
Vibration signal of the method using LabVIEW and MATLAB to drill bit is detected, is ground using different in drill bit operation process
The data of damage degree can quickly judge the quality of drill bit to the training of VMD-TEO Energy-Entropy algorithm of support vector machine, this
Detection method accuracy rate is high, speed is fast, and efficiency is higher.
Now widely used drill bit failures detection device is all traditional hardware testing instrument, mostly uses hardware circuit to vibration
Signal is handled, and is then shown to vibration signal waveforms.The waveform of display is needed to be carried out by the inspector of profession
Judgement, requirement of this drill bit detection mode to inspector be relatively high, and the efficiency and accuracy rate of detection can examined members
It influences.Traditional test equipment mostly uses enclosed construction, is difficult to be modified it and extend, the adaptation for different objects
Property is poor.Virtual test system of the invention is developed jointly using LabVIEW and MATLAB, can effectively will be intelligent
Algorithm combines realization fault diagnosis with virtual instrument technology.Virtual test system uses software design, the processing for signal
With better flexibility；Test macro is integrated with signal acquisition, conversion, data processing, intelligent decision, as the result is shown etc. a variety of
Function need to only modify to software when in face of different test objects, substantially increase test macro using flexible and
Testing efficiency and accuracy rate.
Empirical mode decomposition (Empirical Mode is mostly used for mechanical fault diagnosis field at present
Decomposition, EMD) method extracts characteristic quantity.EMD is a kind of adaptive signal processing method, can be by fault-signal point
Solution is multiple IMF modal components, then recycles the IMF component extraction fault signature for decomposing and obtaining.Since EMD is a kind of recurrence
Formula mode decomposition method accumulates error constantly because of the propagation characteristic of envelope evaluated error in decomposable process, will lead to decomposition
Result in there is modal overlap phenomenon.
Variation mode decomposition (Variational Mode Decomposition, VMD) is a kind of novel compared to EMD
Signal processing method, have solid theoretical basis.VMD uses non-recursive isolation, and this method is searched by iteration
The mode of Variation Model optimal solution is sought to determine the centre frequency and bandwidth of each modal components, so as to adaptively in frequency
Domain subdivision signal and each component is efficiently separated, overcome modal overlap phenomenon present in EMD, making to decompose has better noise
Robustness and modal separation effect.
Summary of the invention
In view of the above-mentioned defects in the prior art, the present invention provide a kind of drill vibration signal on-line virtual testing method and its
System, the purpose is to solve the complexity of collecting and detecting device present in the drill bit failures detection device of the prior art, to operator
More demanding, the problems such as using flexible is poor.The present invention decomposes fault-signal using VMD method, overcomes EMD institute
Existing modal overlap phenomenon, making to decompose has better noise robustness and modal separation effect.
To achieve the above object, the technical solution used in the present invention is:
A kind of the step of drill vibration signal on-line virtual testing method, this method, is:
S1, sensor is set on tested drill bit and makes sensor connection signal conditioning module, sensor acquires tested drill bit
The vibration signal is simultaneously uploaded to signal conditioning module by vibration signal；
S2, signal conditioning module and data collecting card connect, and signal conditioning module is by data by filtering and passing after enhanced processing
It is defeated by data collecting card；Data collecting card handles signal, converts digital signal for the analog signal of acquisition；
S3, data collecting card connection serial communication module are simultaneously connect by serial communication module with calculation processing center, at calculating
The built-in drill vibration signal software systems developed based on LabVIEW and MATLAB in reason center, the data collecting card will be digital
Signal passes to calculation processing center by serial communication module, is handled by calculation processing center；
S4: calculation processing center pre-processes initial data and is stored and converted to data format；Generate input number
According to；The VMD-TEO algorithm of support vector machine pair in MATLABScript node Calling MATLAB is utilized under LabVIEW platform
Input data carries out classification processing；
Variation mode decomposition (VMD) method is first passed through to decompose input data；Optimal five are filtered out using kurtosis criterion
A signal component calculates the teager energy operator (TEO) of this five components, then calculates TEO Energy-Entropy again, constitutes
TEO Energy-Entropy matrix；Using TEO Energy-Entropy matrix as the input of the support vector machines (GA-SVM) of genetic algorithm optimization, utilize
GA-SVM classifies to it；Judge the operating status of drill bit；
S5: calculation processing center connects tested drill bit, works as drill bit by the outer connected control system of serial communication module, the control system
When failure, control system issues instruction, keeps drill bit out of service；Otherwise drill bit continues to run, until pressing stop button
Until.
As improvement to above-mentioned technical proposal, the VMD-TEO algorithm of support vector machine step of the step S4 are as follows:
S41, input data is read, 2048 signaling points is taken to constitute input signal, initialization is carried out to the parameter of VMD and is set
It sets: mode number, penalty factor, bandwidth；
S42, to input signalVariation mode decomposition is carried out, obtains each IMF component, calculate the center frequency of each component
Rate judges whether adjacent centre frequency is similar；If similar, determine that mode number isIf dissmilarity is with mode numberRepeat this step；
S43, it is based on kurtosis criterion, the kurtosis value of each IMF component after VMD is decomposed is calculated, and filter out wherein
The maximum five IMF components of kurtosis value；
S44, five IMF components for being screened in step S43 utilize the teager energy operator formula meter of discrete signal
It calculates, the instantaneous amplitude of each componentWith instantaneous amplitude:
Then each IMF component is calculatedInstantaneous energy spectrumWith Energy-Entropy probability distribution:
Therefore TEO Energy-Entropy are as follows:
To constitute the eigenmatrix of one five dimension；
S45, building GA-SVM classifier, classify to five dimensional feature matrixes of step S44 output.
As further improvement of these options, the VMD algorithm steps of the step S41 are as follows:
Input signal is decomposed into multiple intrinsic mode letters under variation frame by S411, variation mode decomposition (VMD), VMD algorithm
Number (IMF), different IMF concentrate on its corresponding centre frequency () near, the centre frequency and bandwidth of each IMF component
All constantly updated during solving Variation Model；
Variational problem is as follows:
In above formula:Represent input signal:；For the k IMF component decomposed to input signal；For each IMF
The centre frequency of component；
Introduce secondary penalty factorTo guarantee that signal can be decomposed accurately under the influence of Gaussian noise；Utilize Lagrange multiplier
OperatorConstraint condition can be made to keep stringency, Augmented Lagrangian Functions are expressed as follows:
Above formula is alternately updated using alternately Multiplier Algorithm (AMDD)、And, to solve augmentation Lagrange
The optimal solution of function, the specific steps are as follows:
A: initialization、、With；
B: circulation is executed:
C: it updates according to the following formulaWith:
D: following formula pair is utilizedIt updates:
E: judge whether to meet iteration customization condition:
If being unsatisfactory for return step b, step b, c, d are repeated；
If meeting condition, terminating iteration output is, time-domain signal is then converted by Fourier transformation by frequency-region signal, to obtain k IMF component and be respectively。
As further improvement of these options, the GA-SVM classifier building process of the step S45 are as follows:
S451, building GA-SVM sorter model, it is necessary first to the GA-SVM model of building is trained, training GA-SVM mould
Type needs to establish training set and test set, take respectively drill bit in normal state, moderate abrasion and the data that are seriously worn it is each
100 groups, the signal points of every group of data are 2048, and VMD-TEO Energy-Entropy feature extraction is carried out to it, obtains 300 groups's
Eigenmatrix, label is 1 " under normal condition, and label is 2 " under moderate abrasion condition, and label is 3 " under heavy wear conditions.
70 groups of data composing training collection are taken under every kind of state, remaining 90 groups are trained GA-SVM model as test set.For
GA-SVM classifier preservation model after training.
As further improvement of these options, the present invention simultaneously provides a kind of drill vibration signal on-line virtual testing
System, the system include sequentially connected piezoelectric acceleration sensor, signal conditioning module, data collecting card, in calculation processing
The heart；The piezoelectric acceleration sensor is arranged on the drill string of tested drill bit, and the vibration signal of the tested drill bit drill string of acquisition simultaneously will
The vibration signal uploads to signal conditioning module, the vibration signal that the signal conditioning module acquires piezoelectric acceleration sensor
Data collecting card, the analog signal that the data collecting card transmits signal conditioning module are transferred to after being filtered amplification
It is converted into digital signal, and calculation processing center is uploaded by serial communication modular, the calculation processing center includes at data
Module is managed, the built-in drill vibration signal software systems based on LabVIEW platform exploitation of the data processing module are led to by serial ports
The digital signal that letter module uploads is pre-processed and is stored and converted to data format, decomposed, and is sieved by kurtosis criterion
Five optimal signal components are selected, and calculate the TEO Energy-Entropy of this five components, constitute TEO Energy-Entropy matrix.To guarantee
Sensor tests the accuracy of signal, and usual measuring point selection monitoring effect on drill bit drill string is preferable.
As further improvement of these options, the drill vibration signal on-line virtual testing system further includes serial ports
Communication module, serial communication module include initialization of (a) serial ports unit and data-reading unit；Input the data at calculation processing center
Serial communication module is initially entered, serial communication module is divided into initialization of (a) serial ports unit and data-reading unit；Initialization of (a) serial ports
First to resource name corresponding to VISA serial ports, baud rate, data bit, even-odd check, stopping stream and data before unit operation
Stream carries out data initialization, and data-reading unit reads the data that slave computer is transmitted by serial communication module.
As further improvement of these options, data processing module is divided into data generating unit and intelligent algorithm unit；Number
Vibration data is first subjected to format conversion according to generation unit, generates data format workable for matlab；Intelligent algorithm unit
Effect is using VMD-TEO algorithm of support vector machine to input data progress feature extraction and classification, and then to the fortune of drill bit
Row situation judges.
As further improvement of these options, the drill vibration signal on-line virtual testing system further includes control
System, the control system connect with tested drill bit and connect calculation processing center by serial communication modular.The control system
Any control system being known in the art.
As further improvement of these options, the drill vibration signal on-line virtual testing system further includes data
Monitoring module, data monitoring module include data storage element, data query unit, as the result is shown unit, prewarning unit.
Compared with prior art, beneficial effect obtained by the present invention is:
Drill vibration signal on-line virtual testing system of the invention, is developed using LabVIEW platform, takes full advantage of Virtual instrument
The advantage of device improves the using flexible of test macro, reduces the volume of test macro, it is made to easily facilitate carrying, can
It is tested in engineering site.In addition the MATLAB that the present invention is called using the MATLABScript node in LabVIEW platform
In classified based on VMD-TEO algorithm of support vector machine to vibration data, the advantage of this method is obvious, improves
The intelligence degree of drill vibration signal test system reduces the requirement to test device operator, makes layman
Can quick skilled operation test macro, and the efficiency of its test and accuracy rate are also greatly improved.It is specific: 1,
It is simple and practical in structure, small in size it is easy to carry, maintenance load is low, at low cost.2, test macro uses intelligent algorithm, improves
The intelligence degree of system reduces the requirement to user.3, the working efficiency of test macro and accuracy rate are promoted.4, it tests
System using flexible degree height, strong applicability can flexibly change software program setting according to the difference of tested device characteristic, soft
The change of part program is convenient.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention without any creative labor, may be used also for those of ordinary skill in the art
To obtain other drawings based on these drawings.
Fig. 1 is system composed structure schematic diagram of the invention；
Fig. 2 is the composed structure schematic diagram at calculation processing center, serial communication modular；
Fig. 3 is system operational flow diagram；
Fig. 4 is the algorithm of support vector machine flow chart based on VMD-TEO energy operator；
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
As shown in Figure 1,2,3, 4, the step of drill vibration signal on-line virtual testing method of the invention, this method, is:
S1, piezoelectric acceleration sensor 2 is set above the drill string of tested drill bit 1 and makes the connection letter of piezoelectric acceleration sensor 2
Number conditioning module 3, the vibration signal that piezoelectric acceleration sensor 2 acquires tested 1 drill string of drill bit simultaneously upload to the vibration signal
Signal conditioning module 3；
S2, signal conditioning module 3 are connect with data collecting card 4, and signal conditioning module 3 is by data after filtering and enhanced processing
It is transferred to data collecting card 4；Data collecting card 4 handles signal, converts digital signal for the analog signal of acquisition；
S3, data collecting card 4 are connect by serial communication module 5 with calculation processing center 6, are based on built in calculation processing center 6
The drill vibration signal software systems of LabVIEW and MATLAB exploitation, the data collecting card 4 lead to digital signal by serial ports
News module 5 passes to calculation processing center 6, is handled by the data processing module at calculation processing center 6；
S4: calculation processing center 6 pre-processes initial data and is stored and converted to data format；Generate input number
According to；The VMD-TEO algorithm of support vector machine pair in MATLABScript node Calling MATLAB is utilized under LabVIEW platform
Input data carries out classification processing.Variation mode decomposition (VMD) method is first passed through to decompose data；It is sieved using kurtosis criterion
Five optimal signal components are selected, the teager energy operator (TEO) of this five components is calculated, then calculates TEO again
Energy-Entropy constitutes TEO Energy-Entropy matrix；Using TEO Energy-Entropy matrix as the support vector machines (GA-SVM) of genetic algorithm optimization
Input, classified using GA-SVM to it；Judge the operating status of drill bit；
S5: by the outer connected control system 8 of serial communication module 7, which connects tested drill bit 1 at calculation processing center 6,
When drill bit breaks down, control system issues instruction, keeps drill bit out of service；Otherwise drill bit continues to run, and stops until pressing
Only until button.
The VMD-TEO algorithm of support vector machine step of the step S4 are as follows:
S41, input data is read, 2048 signaling points is taken to constitute input signal, initialization is carried out to the parameter of VMD and is set
It sets: mode number, penalty factor, bandwidth；
S42, to input signalVariation mode decomposition is carried out, obtains each IMF component, calculate the center frequency of each component
Rate judges whether adjacent centre frequency is similar；If similar, determine that mode number isIf dissmilarity is with mode numberRepeat this step；
S43, it is based on kurtosis criterion, the kurtosis value of each IMF component after VMD is decomposed is calculated, and filter out wherein
The maximum five IMF components of kurtosis value；
S44, for step when S43 in five IMF components being screened, utilize the teager energy operator formula meter of discrete signal
It calculates, the instantaneous amplitude of each componentWith instantaneous amplitude:
Then each IMF component is calculatedInstantaneous energy spectrumWith energyProbability distribution:
Therefore TEO Energy-Entropy are as follows:
To constitute the eigenmatrix of one five dimension；
S45, building GA-SVM classifier, classify to five dimensional feature matrixes of step S44 output.
Input signal is decomposed into multiple eigen modes under variation frame by S411, variation mode decomposition (VMD), VMD algorithm
State function (IMF), different IMF concentrate on its corresponding centre frequency () near, the centre frequency of each IMF component and
Bandwidth is all constantly updated during solving Variation Model；
Variational problem is as follows:
In above formula:Represent input signal:；For the k IMF component decomposed to input signal；For each IMF
The centre frequency of component；
Introduce secondary penalty factorTo guarantee that signal can be decomposed accurately under the influence of Gaussian noise；Utilize Lagrange multiplier
OperatorConstraint condition can be made to keep stringency, Augmented Lagrangian Functions are expressed as follows:
Above formula is alternately updated using alternately Multiplier Algorithm (AMDD)、And, to solve augmentation Lagrange
The optimal solution of function, the specific steps are as follows:
A: initialization、、With；
B: circulation is executed:
C: it updates according to the following formulaWith:
D: following formula pair is utilizedIt updates:
E: judge whether to meet iteration customization condition:
If being unsatisfactory for return step b, step b, c, d are repeated；
If meeting condition, terminating iteration output is, time-domain signal is then converted by Fourier transformation by frequency-region signal, to obtain k IMF component and be respectively。
The GA-SVM classifier building process of the step S45 are as follows:
S451, building GA-SVM sorter model, it is necessary first to the GA-SVM model of building is trained, training GA-SVM mould
Type needs to establish training set and test set, take respectively drill bit in normal state, moderate abrasion and the data that are seriously worn it is each
100 groups, the signal points of every group of data are 2048, and VMD-TEO Energy-Entropy feature extraction is carried out to it, obtains 300 groups's
Eigenmatrix, label is 1 " under normal condition, and label is 2 " under moderate abrasion condition, and label is 3 " under heavy wear conditions；
70 groups of data composing training collection are taken under every kind of state, remaining 90 groups are trained GA-SVM model as test set；For
GA-SVM classifier preservation model after training.
The present invention simultaneously provides a kind of drill vibration signal on-line virtual testing system, which includes sequentially connected pressure
Electric acceleration transducer 2, signal conditioning module 3, data collecting card 4, calculation processing center 6；The piezoelectric acceleration sensor 2
It is arranged on the drill string of tested drill bit 1, acquire the vibration signal of tested drill bit drill string and the vibration signal is uploaded into signal tune
Module 3 is managed, the signal conditioning module 3 is transmitted after the vibration signal that piezoelectric acceleration sensor 2 acquires is filtered amplification
To data collecting card 4, the analog signal that signal conditioning module 3 transmits is converted digital signal by the data collecting card 4,
And calculation processing center is uploaded by serial communication modular 5, the calculation processing center 6 is built-in to be developed based on LabVIEW platform
Drill vibration signal software systems data processing module, the digital signal that is uploaded by serial communication modular 5 is carried out in advance
Reason and storage simultaneously convert data format, are decomposed, and five optimal signal components are filtered out by kurtosis criterion, and calculate
The TEO Energy-Entropy of this five components out constitutes TEO Energy-Entropy matrix.The drill vibration signal on-line virtual testing system is also
Including control system 8, the control system 8 connect with tested drill bit 1 and connects calculation processing center by serial communication modular 7
6。
The data at input calculation processing center 6 initially enter serial communication module 5, and serial communication module 5 divides at the beginning of serial ports
Beginningization unit and data-reading unit；First to resource name corresponding to VISA serial ports, baud before the operation of initialization of (a) serial ports unit
Rate, even-odd check, stops stream and data stream progress data initialization at data bit, and data-reading unit passes through serial communication mould
Block reads the data that slave computer is transmitted.
Data processing module is divided into data generating unit and intelligent algorithm unit；Vibration data is first carried out lattice by data generating unit
Formula conversion, generates data format workable for matlab；The effect of intelligent algorithm unit is calculated using VMD-TEO support vector machines
Method carries out feature extraction and classification to input data, and then judges to the operation conditions of drill bit.
The drill vibration signal on-line virtual testing system further includes data monitoring module, and data monitoring module includes number
According to storage element, data query unit, as the result is shown unit, prewarning unit.Data monitoring module is able to achieve the storage to data
With inquiry, and the result that can classify to data processing module be shown, and make early warning.Prewarning unit indicates there are three containing
Lamp gives a green light when drill bit mild wear, and moderate bright orange lamp when wearing sends out a warning when being seriously worn.If drill bit is seriously worn
When, it can sound an alarm and control command is issued by the control module of software automatically the controller of drill bit is controlled, make to bore
Head stops working.
Drill vibration signal on-line virtual testing system of the invention, is developed using LabVIEW platform, takes full advantage of void
The advantage of quasi- instrument, improves the using flexible of test macro, reduces the volume of test macro, easily facilitate it and take
Band can be tested in engineering site.In addition the present invention is called using the MATLABScript node in LabVIEW platform
The algorithm of support vector machine based on VMD-TEO Energy-Entropy in MATLAB classifies to vibration data, the advantage of this method
It is obvious, the intelligence degree of drill vibration signal test system is improved, the requirement to test device operator is reduced,
Make layman also can quick skilled operation test macro, and the efficiency of its test and accuracy rate also obtained it is very big
It is promoted.It is specific: 1, it is simple and practical in structure, small in size it is easy to carry, maintenance load is low, at low cost.2, test macro uses
Intelligent algorithm improves the intelligence degree of system, reduces the requirement to user.3, the working efficiency and standard of test macro
True rate is promoted.4, test macro using flexible degree height, strong applicability can flexibly be changed according to the difference of tested device characteristic
Software program setting, software program change are convenient.
Claims (9)
1. a kind of drill vibration signal on-line virtual testing method, which is characterized in that the step of this method is:
S1, sensor is set on tested drill bit and makes sensor connection signal conditioning module, sensor acquires tested drill bit
The vibration signal is simultaneously uploaded to signal conditioning module by vibration signal；
S2, signal conditioning module and data collecting card connect, and signal conditioning module is by data by filtering and passing after enhanced processing
It is defeated by data collecting card；Data collecting card handles signal, converts digital signal for the analog signal of acquisition；
S3, data collecting card connection serial communication module are simultaneously connect by serial communication module with calculation processing center, at calculating
The built-in drill vibration signal software systems developed based on LabVIEW and MATLAB in reason center, the data collecting card will be digital
Signal passes to calculation processing center by serial communication module, is handled by calculation processing center；
S4: calculation processing center pre-processes initial data and is stored and converted to data format；Generate input number
According to；The VMD-TEO algorithm of support vector machine pair in MATLABScript node Calling MATLAB is utilized under LabVIEW platform
Input data carries out classification processing；
Variation mode decomposition method is first passed through to decompose input data；Five optimal signals are filtered out using kurtosis criterion
Component calculates the teager energy operator of this five components, then calculates TEO Energy-Entropy again, constitutes TEO Energy-Entropy square
Battle array；Using TEO Energy-Entropy matrix as the input of the support vector machines of genetic algorithm optimization, classified using support vector machines to it；
Judge the operating status of drill bit；
S5: calculation processing center connects tested drill bit, works as drill bit by the outer connected control system of serial communication module, the control system
When failure, control system issues instruction, keeps drill bit out of service；Otherwise drill bit continues to run, until pressing stop button
Until.
2. drill vibration signal on-line virtual testing method according to claim 1, which is characterized in that the step S4's
VMD-TEO algorithm of support vector machine step are as follows:
S41, input data is read, 2048 signaling points is taken to constitute input signal, Initialize installation is carried out to the parameter of VMD:
Mode number, penalty factor, bandwidth；
S42, to input signalVariation mode decomposition is carried out, obtains each IMF component, the centre frequency of each component is calculated,
Judge whether adjacent centre frequency is similar；If similar, determine that mode number isIf dissmilarity is with mode number
Repeat this step；
S43, it is based on kurtosis criterion, the kurtosis value of each IMF component after VMD is decomposed is calculated, and filter out wherein
The maximum five IMF components of kurtosis value；
S44, for step when S43 in five IMF components being screened, calculated using the TEO energy operator formula of discrete signal,
The instantaneous amplitude of each componentWith instantaneous amplitude:
Then each IMF component is calculatedInstantaneous energy spectrumWith energyProbability distribution:
Therefore TEO Energy-Entropy are as follows:
To constitute the eigenmatrix of one five dimension；
S45, building GA-SVM classifier, classify to five dimensional feature matrixes of step S44 output.
3. drill vibration signal on-line virtual testing method according to claim 2, which is characterized in that the step S41's
VMD algorithm steps are as follows:
Input signal is decomposed into multiple eigen modes under variation frame by S411, variation mode decomposition, variation mode decomposition algorithm
State function, different intrinsic mode functions concentrate on its corresponding centre frequencyNear, the centre frequency of each IMF component with
And bandwidth is all constantly updated during solving Variation Model；
Variational problem is as follows:
In above formula:Represent input signal:；Input signal is decomposed to obtainA IMF component；It is each
The centre frequency of IMF component；
Introduce secondary penalty factorTo guarantee that signal can be decomposed accurately under the influence of Gaussian noise；Utilize Lagrange multiplier
OperatorConstraint condition can be made to keep stringency, Augmented Lagrangian Functions are expressed as follows:
Above formula is alternately updated using alternately Multiplier Algorithm (AMDD)、And, to solve augmentation Lagrange letter
Several optimal solutions, the specific steps are as follows:
A: initialization、、With；
B: circulation is executed:
C: it updates according to the following formulaWith:
D: following formula pair is utilizedIt updates:
E: judge whether to meet iteration customization condition:
If being unsatisfactory for return step b, step b, c, d are repeated；
If meeting condition, terminating iteration output is, time-domain signal is then converted by Fourier transformation by frequency-region signal, to obtain k IMF component and be respectively。
4. drill vibration signal on-line virtual testing method according to claim 2, which is characterized in that the step S45's
GA-SVM classifier building process are as follows:
S451, building GA-SVM sorter model, it is necessary first to the GA-SVM model of building is trained, training GA-SVM mould
Type needs to establish training set and test set, take respectively drill bit in normal state, moderate abrasion and the data that are seriously worn it is each
100 groups, the signal points of every group of data are 2048, and VMD-TEO Energy-Entropy feature extraction is carried out to it, obtains 300 groups's
Eigenmatrix, label is 1 " under normal condition, and label is 2 " under moderate abrasion condition, and label is 3 " under heavy wear conditions；
70 groups of data composing training collection are taken under every kind of state, remaining 90 groups are trained GA-SVM model as test set；For
GA-SVM classifier preservation model after training.
5. a kind of drill vibration signal on-line virtual testing system, which is characterized in that the system includes that sequentially connected piezoelectricity adds
Velocity sensor, signal conditioning module, data collecting card, calculation processing center；The piezoelectric acceleration sensor is arranged in quilt
It surveys above the drill string of drill bit, acquire the vibration signal of tested drill bit drill string and the vibration signal is uploaded into signal conditioning module,
The vibration signal that piezoelectric acceleration sensor acquires is filtered after amplification and is transferred to data acquisition by the signal conditioning module
Card, the analog signal that the data collecting card transmits signal conditioning module are converted into digital signal, and logical by serial ports
Believe that module uploads calculation processing center, the calculation processing center includes data processing module, base built in the data processing module
In the drill vibration signal software systems of LabVIEW platform exploitation, the digital signal uploaded by serial communication modular is carried out pre-
Processing and storage simultaneously convert data format, are decomposed, and five optimal signal components are filtered out by kurtosis criterion, and count
The TEO Energy-Entropy of this five components is calculated, TEO Energy-Entropy matrix is constituted.
6. drill vibration signal on-line virtual testing system according to claim 5, which is characterized in that the drill vibration
Signal on-line virtual testing system further includes control system, and the control system connect with tested drill bit and passes through serial communication mould
Block connects calculation processing center.
7. drill vibration signal on-line virtual testing system according to claim 5, which is characterized in that data processing module
It is divided into data generating unit and intelligent algorithm unit；Vibration data is first carried out format conversion by data generating unit, is generated
Data format workable for matlab；The effect of intelligent algorithm unit is using VMD-TEO algorithm of support vector machine to input number
According to progress feature extraction and classification, and then the operation conditions of drill bit is judged.
8. drill vibration signal on-line virtual testing system according to claim 5, which is characterized in that the drill vibration
Signal on-line virtual testing system further includes serial communication module, and serial communication module includes that initialization of (a) serial ports unit and data are read
Take unit；The data at input calculation processing center initially enter serial communication module, and serial communication module is divided into initialization of (a) serial ports
Unit and data-reading unit；First to resource name corresponding to VISA serial ports, baud rate, number before the operation of initialization of (a) serial ports unit
According to bit, even-odd check, stop stream and data stream progress data initialization, data-reading unit passes through serial communication module and reads
The data that slave computer is transmitted.
9. drill vibration signal on-line virtual testing system according to claim 5, which is characterized in that the drill vibration
Signal on-line virtual testing system further includes data monitoring module, and data monitoring module includes data storage element, data query
Unit, as the result is shown unit, prewarning unit.
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 true CN109443719A (en) | 2019-03-08 |
| CN109443719B 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) |
Cited By (5)
| 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 |
| CN111191502A (en) * | 2019-11-22 | 2020-05-22 | 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 (6)
| 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 |
| 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 |
-
2018
- 2018-11-01 CN CN201811295029.1A patent/CN109443719B/en not_active Expired - Fee Related
Patent Citations (6)
| 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 |
| 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 |
Non-Patent Citations (2)
| Title |
|---|
| 靳行等: "基于EEMD-TEO熵的高速列车故障诊断方法", 《西南交通大学学报》 * |
| 马增强: "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", 《振动与冲击》 * |
Cited By (6)
| 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 |
| CN111191502A (en) * | 2019-11-22 | 2020-05-22 | 中国地质大学（武汉） | Stick-slip and jump drill abnormal working condition identification method based on drill string vibration signal |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109443719B (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109443719A (en) | A kind of drill vibration signal on-line virtual testing method and its system | |
| CN110287552B (en) | Motor bearing fault diagnosis method and system based on improved random forest algorithm | |
| CN100485342C (en) | Integrated supporting vector machine mixed intelligent diagnosing method for mechanical fault | |
| CN109489977B (en) | KNN-AdaBoost-based bearing fault diagnosis method | |
| CN101718634A (en) | Equipment state comprehensive dynamic alarming method based on multivariate probability model | |
| CN104731081B (en) | For the method and system device of high-power locomotive vehicle-mounted data dump analysis | |
| CN109272123A (en) | It is a kind of based on convolution-Recognition with Recurrent Neural Network sucker rod pump operating condition method for early warning | |
| CN104596575B (en) | Blower fan tower barrel safe operation monitoring device and its application method | |
| CN105424364A (en) | Diagnostic method and device of train bearing failure | |
| CN111946559B (en) | Method for detecting structures of wind turbine foundation and tower | |
| CN106840381A (en) | A kind of method of testing of blasting vibration attenuation effect | |
| CN108108880A (en) | A kind of mineral processing production index Visualized Analysis System and method | |
| CN106777622A (en) | The method and system of the electromechanical equipment on-line fault diagnosis based on artificial intelligence | |
| CN110008350A (en) | A kind of pump Ankang knowledge base lookup method based on Bayesian inference | |
| CN108830328A (en) | Merge the microseismic signals SMOTE recognition methods and monitoring system of spatial knowledge | |
| CN109635879A (en) | A kind of Malfunction Diagnosis for Coal-Mining Machine system that parameter is optimal | |
| CN106604193A (en) | Defect detection method and system applied to receiver | |
| CN102494882A (en) | Device for online monitoring and fault diagnosis of spring of vibrating screen for mining and method thereof | |
| CN203159066U (en) | Operation state recorder of crane | |
| CN110058576A (en) | Equipment fault prognostics and health management method based on big data | |
| CN107727392B (en) | State index evaluation and optimization method based on signal detection and ROC analysis | |
| CN104765311A (en) | Pumping well monitoring and early warning device | |
| CN108829084A (en) | A kind of Fault diagnosis model distributed intelligence method and device thereof based on particle filter | |
| CN205280141U (en) | Nearly bank wind, wave, stream and water level monitoring integrated device | |
| CN106777840A (en) | A kind of machinery design parameter optimizes system |
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 |