CN110340733A - A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method - Google Patents
A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method Download PDFInfo
- Publication number
- CN110340733A CN110340733A CN201910655163.6A CN201910655163A CN110340733A CN 110340733 A CN110340733 A CN 110340733A CN 201910655163 A CN201910655163 A CN 201910655163A CN 110340733 A CN110340733 A CN 110340733A
- Authority
- CN
- China
- Prior art keywords
- cutter
- image
- cutting
- online
- place
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q11/00—Accessories fitted to machine tools for keeping tools or parts of the machine in good working condition or for cooling work; Safety devices specially combined with or arranged in, or specially adapted for use in connection with, machine tools
- B23Q11/0042—Devices for removing chips
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
- B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
- B23Q17/0957—Detection of tool breakage
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
- B23Q17/248—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves using special electromagnetic means or methods
- B23Q17/249—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves using special electromagnetic means or methods using image analysis, e.g. for radar, infrared or array camera images
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Optics & Photonics (AREA)
- Machine Tool Sensing Apparatuses (AREA)
Abstract
The invention discloses a kind of damages of Clean Cutting environment bottom tool online with in-place detection system and method, and system includes online anticipation and identifying system, in-place detection system, the cleaning systems for cleaning cutter, computer；Online anticipation and identifying system are used for the vibratory output of real-time measurement cutting position, the changing value of main shaft voltage and electric current, the changing value of cutting force, acoustical signal；In-place detection system for acquiring the side edge image and shear blade image of cutter in place；The computer is used to carry out the signal of online anticipation and identifying system acquisition the extraction and data fusion of temporal signatures and frequency domain character；Whether preliminary judgement is abnormal；Further image is handled, determines the type of impairment and degree of injury of stage property.The method of the present invention tentatively prejudges knife damage by anticipation online and identifying system, then differentiates to the type of impairment and degree of injury of cutter, realizes the accurate detection of tool wear.
Description
Technical field
The invention belongs to Tool monitoring technical field, especially a kind of Clean Cutting environment bottom tool damage is examined in place online
Examining system and method.
Background technique
In automated machine process, monitoring cutting-tool wear state is to reduce manufacturing cost, reduces manufacturing environment danger
Evil guarantees that manufacturing system normally and efficiently run with product quality.Processing of the tool wear to product
Precision, processing quality and processing efficiency significantly affect, if tool wear is not found in time, it will lead to part report
Useless, lathe is shut down, lathe damage or even workpiece fracture and hurt sb.'s feelings, and directly affects processing benefit and economic benefit.Industrial statistics table
Bright, tool wear is to cause the primary factor of machine failure, and thus caused downtime accounts for numerically-controlled machine tool total down-time
1/5~1/3.Therefore, cutting-tool wear state monitoring is carried out for digital-control processing system batch processes, is intelligence manufacture background
Under problem in the urgent need to address.
China Patent Publication No. CN102564314A discloses a kind of for detecting the orthogonal vision of wear condition of end mill
Detection system, the detection system need to shut down disassembly cutter and detect in the lab, occupy production hour, be unfavorable for flexible manufacturing
The raising of system production benefit and economic benefit, and can not detect cutter mill/breakage state change process in process；
China Patent Publication No. CN106840028A discloses a kind of on-position measure method and apparatus of tool wear, by vision-based detection system
System is installed on machine tool side to detect, and vision detection system is fixed, and cannot achieve multi-angle and adopts to cutter progress image
Collection, tool wear region detection is not comprehensive, and does not account for automated cleaning cutting detection environment, lacks vision system protective device
And cutter cleaning apparatus；China Patent Publication No. CN107553219A discloses a kind of based on multiple types sensor composite signal
Tool Wear Monitoring method, due to the complicated multiplicity of machining condition, the disturbing factor for influencing testing result is more, is easy due to other
Failure factor causes cutter mill/breakage erroneous judgement, causes the waste of cutter.In conclusion current detection method has one
Fixed deficiency, lacks a set of efficiency and precision and takes into account, and the tool wear monitoring system of high degree of automation, seriously hinder flexibility
The raising of manufacture system productivity effect and economic benefit.
Summary of the invention
The purpose of the present invention is to provide a kind of Clean Cutting environment bottom tool damage online with in-place detection system and side
Method, to realize the accurate detection of tool wear.
The technical solution for realizing the aim of the invention is as follows:
A kind of damage of Clean Cutting environment bottom tool online and in-place detection system, including prejudge online and identifying system,
In-place detection system, data collecting card, image pick-up card, the cleaning systems for cleaning cutter, computer；
It is described it is online anticipation and identifying system include vibrating sensor, voltage sensor, force snesor, current sensor,
Acoustic emission sensor；The vibrating sensor is used for the vibratory output of cutting position during real-time measurement machine tooling；The electricity
Pressure sensor and current sensor are respectively used to the changing value of main shaft voltage and electric current during real-time measurement machine tooling；It is described
Force snesor is used for the changing value of cutting force during real-time measurement machine tooling；The acoustic emission sensor is used for real-time measurement
Acoustical signal when Tool in Cutting；
The in-place detection system includes robot, vision system；Cutting for machine tooling center is arranged in the robot
Cut position side nearby；In robot end, the robot drives vision system to acquire knife in place for the vision system setting
The side edge image and shear blade image of tool；Signal that the data collecting card, image pick-up card respectively detect each sensor and
The image transmitting of vision system acquisition is to computer；
The computer is equipped with data processing module and image processing module, and the data processing module is used for each sensing
The vibration of device acquisition, electric current, voltage, cutting force, acoustic emission signal carries out the extraction of temporal signatures and frequency domain character and data are melted
It closes；Using the data characteristics of extraction as the longitudinal axis, the time is that horizontal axis generates change curve, if in process, curve is more than setting threshold
Value, then issue early warning, and lathe is shut down；Described image processing module, for handling the tool image that vision system acquires,
Cutting Tool Damage degree is calculated, discriminates whether to need to change cutter, and show on a display screen；When computer-made decision prejudges and knows online
When the data exception of other system acquisition, lathe is shut down, and cleaning systems start to clean cutter, cutter rotation, and vision system starts to adopt
Collect the side edge image and shear blade image of cutter；Further determine the type of impairment and degree of injury of cutter.
Compared with prior art, the present invention its remarkable advantage is:
(1) present invention is using with comprehensive cutter mill/damage testing method in place, realizing Cutting Tool Damage detection effect online
The dual promotion of rate and reliability: online anticipation and identifying system tentatively prejudge cutter mill/breakage, when preliminary judgement knife
After having mill/breakage, tool image is acquired by in-place detection system, cutter mill/breakage Precision measurement is carried out, finally determines knife
Have mill/damaged degree, cutter mill/damaged type and whether need to change cutter, reduces cutter mill/damage testing error rate；?
Line prejudges process in real time and Precision measurement in place is made to acquire tool image without frequent interval, only arrives cutter mill/breakage in anticipation
After carry out Image Acquisition, reduce the data collection capacity of Precision measurement in place, reduce machine stop times, effectively extend lathe and set
Standby non-failure operation time, the productivity effect and economic benefit of flexible manufacturing system greatly improved.
(2) present invention carries out Image Acquisition using soft robot driving camera camera system, and flexibility preferably, can be preferable
Adaptation lathe machining center in narrow space, the feature more than obstacle, and the degree of automation is higher, realizes cutter mill/breakage
Intellectualized detection.
(3) present invention is by setting starting cleaning high-pressure spray gun, lens wiper and motorized window, when can effectively prevent detection
Chip is speckled on cutter, while being kept the cleaning of camera lens and being prevented chip in process from scratching camera lens and cutting fluid, water
Vapour etc. is attached on camera lens, detects environment for the clean cutting that cutter mill/damage testing provides；Improve detection accuracy.
(4) present invention uses Frequency detection method, and preceding the 60% of the cutter rated life time, high-precision detection system in place is daily
It is detected before booting processing or after process finishing, when after line anticipation and identifying system are alarmed or after the cutter rated life time
In 40% process time section, high-precision detection system detection frequency in place is numerous, before per pass manufacturing procedure starts or can tie
The high-precision detection of vision in place is carried out after beam.Avoiding online anticipation and identifying system anticipation error influences processing quality, mentions simultaneously
High detection efficiency.
Detailed description of the invention
Fig. 1 is that detection system illustrates general structure schematic diagram.
Fig. 2 is vision system structural schematic diagram.
Fig. 3 is cleaning systems structural schematic diagram.
Fig. 4 is lens wiper structural schematic diagram.
Fig. 5 is that industrial camera acquires cutter side edge image schematic diagram.
Fig. 6 is that industrial camera acquires cutter shear blade image schematic diagram.
Fig. 7 is present system workflow schematic diagram.
Specific embodiment
With reference to the accompanying drawing and specific embodiment is described further the present invention.
In conjunction with Fig. 1-Fig. 6, a kind of Clean Cutting environment bottom tool damage of the invention online and in-place detection system, including
Online anticipation and identifying system, in-place detection system, data collecting card 15, image pick-up card 13, cleaning systems, computer 14；
The online anticipation and identifying system include vibrating sensor 1, voltage sensor 2, force snesor 3, current sense
Device 4, acoustic emission sensor 5；The vibrating sensor 1 is mounted on the cutting position at machine tooling center 6 nearby or at knife handle, uses
Vibratory output during real-time measurement machine tooling near cutting position；The voltage sensor 2 is installed with current sensor 4
Changing value at the control cabinet at machine tooling center 6, for main shaft voltage and electric current during real-time measurement machine tooling；Institute
State force snesor 3 be installed at the workbench at machine tooling center 6 or machine tool chief axis clamping cutter 16 at, be used for real-time measurement machine
The changing value of cutting force in bed process；The acoustic emission sensor 5 is fixed on cutting for machine tooling center 6 by magnet base
It cuts near position, acoustical signal when being cut for real-time measurement cutter 16；
The signal that each sensor detects is transferred to computer 14 by the data collecting card 15；The in-place detection system
Including robot 78, vision system；Side near the cutting position at machine tooling center 6 is arranged in the robot 78；It is described
Vision system setting in 78 end of robot, the robot 78 drive vision system acquire in place cutter 16 side edge image and
Shear blade image；
The cleaning systems guarantee the accuracy of vision system acquisition image for cleaning cutter 16；
The computer 14 is equipped with data processing module and image processing module, and the data processing module is used for each biography
The extraction and data of vibration, electric current, voltage, cutting force, acoustic emission signal progress temporal signatures and frequency domain character that sensor acquires
Fusion；Using the data characteristics of extraction as the longitudinal axis, the time is that horizontal axis generates change curve, if in process, curve is more than setting
Threshold value, then issue early warning, and lathe is shut down；Described image processing module, 16 image of cutter for acquiring to vision system carry out
Processing calculates Cutting Tool Damage degree, discriminates whether to need to change cutter, and show on a display screen；When computer 14 determines online
When the data exception of anticipation and identifying system acquisition, lathe is shut down, and cleaning systems start to clean cutter 16, and cutter 16 rotates, depending on
Feel system starts to acquire the side edge image and shear blade image of cutter 16；Further determine the type of impairment and damage journey of cutter 16
Degree.
The computer 14 is used to control the operation at machine tooling center 6 and robot 78, and controls cutter and rotate a circle
The frequency, the side edge image taking to guarantee cutter 16 is complete；The computer 14 is equipped with data processing module and image procossing
Module；Vibration, electric current, voltage, cutting force, acoustic emission signal progress of the data processing module for being acquired to each sensor
The extraction and data fusion of temporal signatures and frequency domain character；Using the data characteristics of extraction as the longitudinal axis, the time is that horizontal axis generates variation
Curve then issues early warning if in process, curve is more than threshold value, and lathe is shut down；Described image processing module is used for vision
16 image of cutter of system acquisition is handled, and is calculated Cutting Tool Damage degree, is discriminated whether to need to change cutter, and be shown in display
On screen.
Further, the data processing module includes feature extraction and integrated unit, curve generation unit, the first judgement
Unit；
Vibration, electric current, voltage, cutting force and the acoustic emission signal that the feature extraction and integrated unit are used to acquire
Cutter mill/breakage state temporal signatures, frequency domain character can be objectively responded by extracting by formula (1)-(8):
Temporal signatures:
xMax=max (| xi|) (3)
Frequency domain character:
xiRepresent collected monitoring signals in the certain time period of i-th of sensor during the cutting process, i=1,2,
3,…,N；N is number of probes；μ is the mean value of collected monitoring signals in certain time period during the cutting process, is
The static part of monitoring signals reflects the variation tendency of monitoring signals；xRMSTo be adopted in certain time period during the cutting process
The root mean square of the monitoring signals collected indicates to reflect monitoring in the average energy of a certain section of given monitoring in time signal
The intensity of signal；xMaxFor the sum of the maximum value of collected monitoring signals in certain time period during the cutting process, prison is indicated
Survey the maximum instantaneous amplitude of signal；xStdFor the standard deviation of collected monitoring signals in certain time period during the cutting process,
The dynamic part of monitoring signals reflects the degree that monitoring signals fluctuate near mean value；xSkeIt is a certain for during the cutting process
The degree of bias of collected monitoring signals in period reflects monitoring signals using mean value as the degree of asymmetry of line of symmetry；xKurFor
The kurtosis of collected monitoring signals in certain time period in cutting process reflects the transient phenomenon of monitoring signals and steady
Property；fiIt represents collected monitoring signals in given certain time period and passes through Fast Fourier Transform (FFT) (FFT) from time-domain signal
The frequency spectrum that (i.e. original signal) converts, P (fi) power spectral densities of monitoring signals is represented, M represents the power of monitoring signals
The length of spectrum；xFCIt is the static part of frequency spectrum for the center of gravity of frequency of monitoring signals；xFVFor the frequency variance of monitoring signals, frequency spectrum
Divide dynamic state part, reflect the degree of fluctuation of the frequency spectrum of monitoring signals near center of gravity of frequency.
The curve generation unit is used for using the data characteristics of extraction as the longitudinal axis, and the time is horizontal axis, and real-time display is through feature
Extract the curve changed over time with each temporal signatures, the frequency domain character merged.
First judging unit be used for determine above-mentioned curve characteristic value at a time whether beyond setting threshold
Value issues early warning if being more than threshold value, and lathe is shut down.
Further, described image processing module includes image processing unit, damage measurement unit, the second judging unit；
Described image processing unit be used to carry out the tool image of acquisition gray processing, noise reduction, median filtering, at binaryzation
Reason and edge detection extract tool wear profile.
The damage measurement unit calculates mill/breakage area and maximum for measuring to cutter mill/damaged area
Mill/breakage bandwidth and average mill/breakage bandwidth.Such as the Pixel Dimensions of image are set as M × N, according to optical amplifier times
Number, can calculate the real area S and side length a × b of a pixel on uncalibrated image, can calculate mill/breakage area and maximum
Mill/breakage bandwidth and average mill/breakage bandwidth.
Second judging unit differentiates cutter mill damaged type (side edge damage or shear blade damage according to tool wear position
Wound), and whether setting value is reached to determine whether to need to change cutter according to tool abrasion.
In order to avoid prejudge online and identifying system anticipation error, influence processing quality, while in order to improve detection efficiency,
Using Frequency detection method, i.e., within lathe actual processing initial stage, the preceding 60% process time section of cutter rated life time,
The high-precision detection system in position can carry out the high-precision detection of vision in place before booting processing daily or to cutter after process finishing, and work as
After line anticipation and identifying system alarm or in the rear 40% process time section of cutter rated life time, high-precision detection system in place
Detect frequency it is numerous, can before per pass manufacturing procedure starts or after carry out the high-precision detection of vision in place.
Further, the vision system 7 includes connector 75, camera support 74, industrial camera 70, camera lens 71, annular
Light source 72；The camera support 74 is fixed on the execution end of robot 78 by connector 75；The industrial camera 70 is fixed
On camera support 74；The camera lens 71 is connect with industrial camera 70, and the annular light source 72 is mounted on 71 front end of camera lens；It is described
Robot 78 drives the image of the acquisition cutter 16 in place of industrial camera 70.
Further, the robot 78 uses the existing soft robot purchased.
Further, when acquiring the side edge image of cutter 16, cutter need to rotate a circle, and a point n times rotate, and 2 π of each rotation/
The angle of n, n > 2, the side edge image taking to guarantee cutter 16 are complete；
Further, the cleaning systems include motorized window 9, Pneumatic cleaning high-pressure spray gun 11, air compressor 10, band spray
The rubber hose 12 of mouth, lens protecting cap 73；The motorized window 9 is installed near the cutting position at machine tooling center 6, setting
Between vision system and stage property 16, closed carrying out cutter mill when Precision measurement/breakage in place be automatically that robot 78 is opened/
It closes, vision system 7 can be made to be isolated with processing environment, prevent the chip scratch camera lens 71 when processing, cutting fluid from drenching camera lens 71, water
Vapour is attached on camera lens 71.When carrying out Image Acquisition, motorized window 9 is opened, and robot 78 protrudes into machine tooling center 6, band
Dynamic industrial camera 70 completes the Image Acquisition of 16 side edge of cutter, shear blade, and acquisition Wan Hou robot 78 drives industrial camera 70 to return to
In situ, motorized window 9 is closed；The Pneumatic cleaning high-pressure spray gun 11 is connected on air compressor 10, and the rubber hose 12 is connected to
At gun slot, 12 nozzle one end of rubber hose is fixed on robot 78 by fixed frame 77 and executes end, carries out to cutter 16
Before Image Acquisition, rubber hose 12 and robot 78 are servo-actuated, achieve the purpose that clean cutter 16；The lens protecting cap 73 is installed
The dirts such as contamination dust, chip and cutting fluid are avoided for protecting camera lens 71 and annular light source 72 in 72 front end of annular light source
Dirt.
Further, the cleaning systems further include lens wiper 8, and the lens wiper 8 is fixed near robot 78, and electricity
It is isolated dynamic window 9 with processing environment；The lens wiper 8 includes wiping mirror cotton 81, bracket 82, and the wiping mirror cotton 81 is fixed on branch
82 upper end of frame；After camera lens 71 and annular light source 72 are infected with dust or dirt, robot 78 drives camera lens 71 in cleaning cotton
Reciprocal wiping back and forth on 81；
In conjunction with Fig. 7, based on above-mentioned detection system, the invention also provides a kind of damages of Clean Cutting environment bottom tool to exist
Line in position detecting method, specifically includes the following steps:
Step 1, multi-sensor cooperation prejudge cutter mill/breakage in real time online:
Step 1.1, acquisition data: operation lathe carries out actual processing, utilizes vibrating sensor, current sensor, voltage
Sensor, cutting force snesor, acoustic emission sensor acquire the vibration of lathe, electric current, voltage, cutting force, sound emission letter in real time
Number；
It step 1.2, feature extraction and merges: the vibration of acquisition, electric current, voltage, cutting force and acoustic emission signal is pressed
Formula (1)-(8), which are extracted, can objectively respond cutter mill/breakage state temporal signatures, frequency domain character:
Temporal signatures:
xMax=max (| xi|) (3)
Frequency domain character:
xiRepresent collected monitoring signals in the certain time period of i-th of sensor during the cutting process, i=1,2,
3,…,N；N is number of probes；μ is the mean value of collected monitoring signals in certain time period during the cutting process, is
The static part of monitoring signals reflects the variation tendency of monitoring signals；xRMSTo be adopted in certain time period during the cutting process
The root mean square of the monitoring signals collected indicates to reflect monitoring in the average energy of a certain section of given monitoring in time signal
The intensity of signal；xMaxFor the sum of the maximum value of collected monitoring signals in certain time period during the cutting process, prison is indicated
Survey the maximum instantaneous amplitude of signal；xStdFor the standard deviation of collected monitoring signals in certain time period during the cutting process,
The dynamic part of monitoring signals reflects the degree that monitoring signals fluctuate near mean value；xSkeIt is a certain for during the cutting process
The degree of bias of collected monitoring signals in period reflects monitoring signals using mean value as the degree of asymmetry of line of symmetry；xKurFor
The kurtosis of collected monitoring signals in certain time period in cutting process reflects the transient phenomenon of monitoring signals and steady
Property；fiIt represents collected monitoring signals in given certain time period and passes through Fast Fourier Transform (FFT) (FFT) from time-domain signal
The frequency spectrum that (i.e. original signal) converts, P (fi) power spectral densities of monitoring signals is represented, M represents the power of monitoring signals
The length of spectrum；xFCIt is the static part of frequency spectrum for the center of gravity of frequency of monitoring signals；xFVFor the frequency variance of monitoring signals, frequency spectrum
Divide dynamic state part, reflect the degree of fluctuation of the frequency spectrum of monitoring signals near center of gravity of frequency；
Step 1.3, online anticipation: using the data characteristics of extraction as the longitudinal axis, the time is horizontal axis, and real-time display is through feature extraction
The curve changed over time with each temporal signatures, the frequency domain character merged then issues if in process, curve is more than threshold value
Early warning, lathe are shut down；
Step 2, after being cleaned to cutter, vision system detects cutter mill/breakage in place:
Step 2.1, soft robot drive camera to acquire complete cutter side edge, shear blade image, acquire cutter side edge figure
When picture, cutter need to rotate a circle, and a point n times rotate, and rotate 2 π/n angle, n > 2, to guarantee that cutter side edge image taking is complete every time
It is whole, image storing path is set in the image capture software of computer, and the image of acquisition is transferred to by gigabit Ethernet line
Image pick-up card is transmitted further to computer, and acquisition Wan Hou robot drives industrial camera to return to original position, and motorized window is closed；
Step 2.2 carries out gray processing, noise reduction, intermediate value filter using tool image of the image processing and analysis software to acquisition
Wave, binary conversion treatment and edge detection extract tool wear profile.If the Pixel Dimensions of image are M × N, according to optical amplifier
Multiple can calculate the real area S and side length a × b of a pixel on uncalibrated image, measure to cutter mill/damaged area,
Calculate mill/breakage area and maximum mill/breakage bandwidth and average mill/breakage bandwidth；
Whether step 3, the tool abrasion according to high-precision detection in place and position differentiate cutter mill damaged type and need
Replace cutter.
In order to avoid prejudge online and identifying system anticipation error, influence processing quality, while in order to improve detection efficiency,
Using Frequency detection method, i.e., within lathe actual processing initial stage, the preceding 60% process time section of cutter rated life time,
The high-precision detection system in position can be switched on daily process before or process finishing after the high-precision detection of vision in place is carried out to cutter, and when online
After anticipation and identifying system alarm or in the rear 40% process time section of cutter rated life time, high-precision detection system inspection in place
Measured frequency frequency conversion is numerous, can before per pass manufacturing procedure starts or after carry out the high-precision detection of vision in place.
Claims (10)
1. a kind of damage of Clean Cutting environment bottom tool is online and in-place detection system, which is characterized in that including prejudging online and
Identifying system, in-place detection system, data collecting card (15), image pick-up card (13), the cleaning systems by cleaning cutter, based on
Calculation machine (14)；
The online anticipation and identifying system include vibrating sensor (1), voltage sensor (2), force snesor (3), electric current biography
Sensor (4), acoustic emission sensor (5)；The vibrating sensor (1) is for cutting position during real-time measurement machine tooling
Vibratory output；The voltage sensor (2) and current sensor (4) are respectively used to main shaft voltage during real-time measurement machine tooling
With the changing value of electric current；The force snesor (3) is used for the changing value of cutting force during real-time measurement machine tooling；The sound
Emission sensor (5) is used for acoustical signal when real-time measurement Tool in Cutting；
The in-place detection system includes robot (78), vision system；The robot (78) is arranged at machine tooling center
Cutting position nearby side；In robot (78) end, the robot (78) drives vision system for the vision system setting
The side edge image and shear blade image of system acquisition cutter in place；The data collecting card (15), image pick-up card (13) respectively will be each
The image transmitting for signal and the vision system acquisition that sensor detects gives computer (14)；
The computer (14) is equipped with data processing module and image processing module, and the data processing module is used for each sensing
The vibration of device acquisition, electric current, voltage, cutting force, acoustic emission signal carries out the extraction of temporal signatures and frequency domain character and data are melted
It closes；Using the data characteristics of extraction as the longitudinal axis, the time is that horizontal axis generates change curve, if in process, curve is more than setting threshold
Value, then issue early warning, and lathe is shut down；Described image processing module, for handling the tool image that vision system acquires,
Cutting Tool Damage degree is calculated, discriminates whether to need to change cutter, and show on a display screen；When computer (14) determine online anticipation
And identifying system acquisition data exception when, lathe is shut down, and cleaning systems start to clean cutter, and cutter rotation, vision system opens
The side edge image and shear blade image of beginning acquisition cutter；Further determine the type of impairment and degree of injury of cutter.
2. detection system according to claim 1, which is characterized in that the data processing module includes feature extraction and melt
Close unit, curve generation unit, the first judging unit；
Vibration, electric current, voltage, cutting force and the acoustic emission signal that the feature extraction and integrated unit are used to acquire are extracted
Reflect cutter mill/breakage state temporal signatures, frequency domain character；
The curve generation unit is used for using the data characteristics of extraction as the longitudinal axis, and the time is horizontal axis, and real-time display is through feature extraction
The curve changed over time with each temporal signatures, the frequency domain character merged；
First judging unit be used for determine above-mentioned curve characteristic value at a time whether beyond setting threshold value, if
More than threshold value, then early warning is issued, lathe is shut down.
3. detection system according to claim 1, which is characterized in that described image processing module includes image procossing list
Member, damage measurement unit, the second judging unit；
Described image processing unit be used to carry out the tool image of acquisition gray processing, noise reduction, median filtering, binary conversion treatment with
And edge detection extracts tool wear profile；
The damage measurement unit for being measured to cutter mill/damaged area, calculate mill/breakage area and it is maximum grind/
Damaged bandwidth and average mill/breakage bandwidth；
Second judging unit differentiated according to tool wear position cutter grind damaged type, and according to tool abrasion whether
Reach setting value to determine whether to need to change cutter.
4. detection system according to claim 1, which is characterized in that the vision system (7) includes connector (75), phase
Machine support (74), industrial camera (70), camera lens (71), annular light source (72)；The camera support (74) passes through connector (75)
It is fixed on the execution end of robot (78)；The industrial camera (70) is fixed on camera support (74)；The camera lens (71)
It is connect with industrial camera (70), the annular light source (72) is arranged in camera lens (71) front end；The robot (78) drives industry
Camera (70) acquires the image of cutter in place.
5. detection system according to claim 1, which is characterized in that the cleaning systems include motorized window (9), pneumatic clear
Wash high-pressure spray gun (11), air compressor (10), the rubber hose (12) with nozzle, lens protecting cap (73)；The motorized window
(9) it is arranged between vision system and stage property；The Pneumatic cleaning high-pressure spray gun (11) is connected on air compressor (10), described
Rubber hose (12) is connected at gun slot, and rubber hose (12) nozzle one end is fixed on robot (78) by fixed frame (77) and holds
Row end；The lens protecting cap (73) is installed on annular light source (72) front end, for protecting camera lens (71) and annular light source
(72)。
6. detection system according to claim 6, which is characterized in that the cleaning systems further include lens wiper (8), described
Lens wiper (8) includes wiping mirror cotton (81), bracket (82), and the wiping mirror cotton (81) is fixed on bracket (82) upper end；When camera lens (71)
After being infected with dust or dirt with annular light source (72), robot (78) drives camera lens (71) to wipe on cleaning cotton (81).
7. the detection method of detection system according to claim 1-7, which comprises the following steps:
Step 1, multi-sensor cooperation prejudge cutter mill/breakage in real time online:
Step 1.1, acquisition data: the vibration of acquisition lathe, electric current, voltage, cutting force, acoustic emission signal in real time；
It step 1.2, feature extraction and merges: the vibration of acquisition, electric current, voltage, cutting force and acoustic emission signal being extracted anti-
Reflect cutter mill/breakage state temporal signatures, frequency domain character；
Step 1.3, online anticipation: using the data characteristics of extraction as the longitudinal axis, the time is horizontal axis, real-time display through feature extraction with melt
The curve that each temporal signatures, the frequency domain character closed changes over time then issues early warning if in process, curve is more than threshold value,
Lathe is shut down；
Step 2, after being cleaned to cutter, vision system detects cutter mill/breakage in place:
Step 2.1, the complete cutter side edge of acquisition, shear blade image；
Step 2.2 proposes tool image progress gray processing, noise reduction, median filtering, binary conversion treatment and the edge detection of acquisition
Take tool wear profile；Cutter mill/damaged area is measured, mill/breakage area and maximum mill/breakage bandwidth are calculated
And average mill/breakage bandwidth；
Whether step 3, the tool abrasion according to high-precision detection in place and position differentiate cutter mill damaged type and need to change
Cutter.
8. detection method according to claim 8, which is characterized in that temporal signatures:
xMax=max (| xi|)(3)
Frequency domain character:
xiRepresent collected monitoring signals in the certain time period of i-th of sensor during the cutting process, i=1,2,3 ...,
N；N is number of probes；μ is the mean value of collected monitoring signals in certain time period during the cutting process；xRMSTo cut
The root mean square of collected monitoring signals in certain time period during cutting；xMaxFor certain time period during the cutting process
The sum of the maximum value of interior collected monitoring signals；xStdFor monitoring collected in certain time period during the cutting process letter
Number standard deviation；xSkeFor the degree of bias of collected monitoring signals in certain time period during the cutting process；xKurTo cut
The kurtosis of collected monitoring signals in certain time period in the process；fiRepresent collected prison in given certain time period
Survey the frequency spectrum that signal is converted by Fast Fourier Transform (FFT) from time-domain signal, P (fi) represent the power spectrums of monitoring signals
Degree, M represent the length of the power spectrum of monitoring signals；xFCFor the center of gravity of frequency of monitoring signals；xFVFor the frequency side of monitoring signals
Difference.
9. detection method according to claim 8, it is characterised in that in order to avoid prejudging online and identifying system prejudges out
Mistake influences processing quality, while in order to improve detection efficiency, using Frequency detection method, i.e., at lathe actual processing initial stage,
In the preceding 60% process time section of cutter rated life time, high-precision detection system in place can be switched on daily processes preceding or process finishing
The high-precision detection of vision in place is carried out to cutter afterwards, and when after line anticipation and identifying system are alarmed or after the cutter rated life time
In 40% process time section, high-precision detection system detection frequency in place is numerous, before per pass manufacturing procedure starts or can tie
The high-precision detection of vision in place is carried out after beam.
10. detection method according to claim 8, which is characterized in that when acquisition cutter side edge image, cutter need to rotate one
Week, a point n times rotate, and rotate 2 π/n angle, n > 2 every time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910655163.6A CN110340733A (en) | 2019-07-19 | 2019-07-19 | A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910655163.6A CN110340733A (en) | 2019-07-19 | 2019-07-19 | A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110340733A true CN110340733A (en) | 2019-10-18 |
Family
ID=68179313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910655163.6A Pending CN110340733A (en) | 2019-07-19 | 2019-07-19 | A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110340733A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110900659A (en) * | 2019-12-11 | 2020-03-24 | 广东博智林机器人有限公司 | Robot collision detection method and device, electronic equipment and storage medium |
| CN111047578A (en) * | 2019-12-12 | 2020-04-21 | 深圳精匠云创科技有限公司 | Cutter detection device and detection method thereof |
| CN111060603A (en) * | 2019-12-19 | 2020-04-24 | 江苏理工学院 | Sleeper beam damage monitoring method, system and device |
| CN111122587A (en) * | 2020-01-19 | 2020-05-08 | 南京理工大学 | Cutter damage detection method based on visual feature extraction |
| CN111347292A (en) * | 2020-02-21 | 2020-06-30 | 青岛理工大学 | System and method for monitoring and controlling state of cutter of numerical control machine tool |
| CN111692968A (en) * | 2020-07-27 | 2020-09-22 | 上海威研精密科技有限公司 | On-machine multi-view visual detector for micro milling cutter and detection method thereof |
| CN113427321A (en) * | 2021-07-22 | 2021-09-24 | 南京理工大学 | Real-time detection device and detection method for cutter used in machining center |
| CN113752088A (en) * | 2021-09-22 | 2021-12-07 | 南京理工大学 | Tool magazine integrated type tool damage detection system and method based on machine vision |
| CN113843659A (en) * | 2021-09-22 | 2021-12-28 | 南京理工大学 | Machine vision-based spindle integrated type cutter damage frequency conversion on-machine detection system and method |
| CN113916983A (en) * | 2021-09-30 | 2022-01-11 | 南京中科煜宸激光技术有限公司 | Device for detecting damage of scraper of powder spreading type metal 3D printing equipment |
| CN114378639A (en) * | 2021-12-30 | 2022-04-22 | 科大讯飞华南人工智能研究院（广州）有限公司 | Tool life prediction method and device, storage medium and electronic device |
| CN114880814A (en) * | 2022-07-08 | 2022-08-09 | 南通恒强轧辊有限公司 | Roller remanufacturing auxiliary optimization method based on big data |
| CN115302846A (en) * | 2022-10-10 | 2022-11-08 | 苏州鼎佳精密科技股份有限公司 | Die-cutting machine |
| CN115635366A (en) * | 2022-10-12 | 2023-01-24 | 哈尔滨工程大学 | Large machine tool blade acoustic emission online monitoring system and method |
| CN116690314A (en) * | 2023-08-01 | 2023-09-05 | 合肥智谷机电科技有限公司 | Cutter damage detection device based on machine vision detects and compares |
| CN117030857A (en) * | 2023-10-07 | 2023-11-10 | 中科航迈数控软件(深圳)有限公司 | Tool flaw detection method based on phase control ultrasonic waves and related equipment |
| CN117300729A (en) * | 2023-10-12 | 2023-12-29 | 山东润龙精密机床有限公司 | Efficient and precise machining method for normal round holes of special-shaped products based on three-axis machine tool |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1133882A (en) * | 1997-07-14 | 1999-02-09 | Toshiba Mach Co Ltd | Tool breakage judging method |
| CN102528561A (en) * | 2012-02-28 | 2012-07-04 | 上海大学 | On-line automatic detection device for detecting wear condition of rotary cutter in the whole processing cycle |
| CN102564314A (en) * | 2011-12-06 | 2012-07-11 | 上海交通大学 | Orthogonal vision detection system for detecting wear condition of end mill |
| CN106840028A (en) * | 2016-12-23 | 2017-06-13 | 湖北文理学院 | The on-position measure method and apparatus of tool wear |
| CN107398778A (en) * | 2017-07-31 | 2017-11-28 | 詹白勺 | HSC Milling Tools wear vision in-situ monitor system and method |
| CN206855141U (en) * | 2017-04-01 | 2018-01-09 | 深圳市蓝海永兴实业有限公司 | A kind of milling cutter wears on-line measuring device |
| CN207058207U (en) * | 2017-07-31 | 2018-03-02 | 台州市质量技术监督检测研究院 | The fine cutter wear depth on-position measure device of high-speed cutting |
| CN108414382A (en) * | 2018-02-08 | 2018-08-17 | 大连理工大学 | A kind of software end effector to the detection of TBM cutting-tool wear states |
| CN109623655A (en) * | 2019-02-25 | 2019-04-16 | 长沙理工大学 | A kind of stone state in-situ monitor system and method |
| CN109732406A (en) * | 2019-01-07 | 2019-05-10 | 北京信息科技大学 | A kind of intelligent tool method for diagnosing faults |
| CN210678020U (en) * | 2019-07-19 | 2020-06-05 | 南京理工大学 | Online and in-place detection system for cutter damage in clean cutting environment |
-
2019
- 2019-07-19 CN CN201910655163.6A patent/CN110340733A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH1133882A (en) * | 1997-07-14 | 1999-02-09 | Toshiba Mach Co Ltd | Tool breakage judging method |
| CN102564314A (en) * | 2011-12-06 | 2012-07-11 | 上海交通大学 | Orthogonal vision detection system for detecting wear condition of end mill |
| CN102528561A (en) * | 2012-02-28 | 2012-07-04 | 上海大学 | On-line automatic detection device for detecting wear condition of rotary cutter in the whole processing cycle |
| CN106840028A (en) * | 2016-12-23 | 2017-06-13 | 湖北文理学院 | The on-position measure method and apparatus of tool wear |
| CN206855141U (en) * | 2017-04-01 | 2018-01-09 | 深圳市蓝海永兴实业有限公司 | A kind of milling cutter wears on-line measuring device |
| CN107398778A (en) * | 2017-07-31 | 2017-11-28 | 詹白勺 | HSC Milling Tools wear vision in-situ monitor system and method |
| CN207058207U (en) * | 2017-07-31 | 2018-03-02 | 台州市质量技术监督检测研究院 | The fine cutter wear depth on-position measure device of high-speed cutting |
| CN108414382A (en) * | 2018-02-08 | 2018-08-17 | 大连理工大学 | A kind of software end effector to the detection of TBM cutting-tool wear states |
| CN109732406A (en) * | 2019-01-07 | 2019-05-10 | 北京信息科技大学 | A kind of intelligent tool method for diagnosing faults |
| CN109623655A (en) * | 2019-02-25 | 2019-04-16 | 长沙理工大学 | A kind of stone state in-situ monitor system and method |
| CN210678020U (en) * | 2019-07-19 | 2020-06-05 | 南京理工大学 | Online and in-place detection system for cutter damage in clean cutting environment |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110900659A (en) * | 2019-12-11 | 2020-03-24 | 广东博智林机器人有限公司 | Robot collision detection method and device, electronic equipment and storage medium |
| CN111047578A (en) * | 2019-12-12 | 2020-04-21 | 深圳精匠云创科技有限公司 | Cutter detection device and detection method thereof |
| CN111047578B (en) * | 2019-12-12 | 2023-07-18 | 深圳精匠云创科技有限公司 | Tool detection device and detection method thereof |
| CN111060603A (en) * | 2019-12-19 | 2020-04-24 | 江苏理工学院 | Sleeper beam damage monitoring method, system and device |
| CN111122587A (en) * | 2020-01-19 | 2020-05-08 | 南京理工大学 | Cutter damage detection method based on visual feature extraction |
| CN111122587B (en) * | 2020-01-19 | 2022-06-28 | 南京理工大学 | Cutter damage detection method based on visual feature extraction |
| CN111347292A (en) * | 2020-02-21 | 2020-06-30 | 青岛理工大学 | System and method for monitoring and controlling state of cutter of numerical control machine tool |
| WO2021164137A1 (en) * | 2020-02-21 | 2021-08-26 | 青岛理工大学 | Cutting tool state monitoring and control system and method for numerical control machine tool |
| CN111692968B (en) * | 2020-07-27 | 2021-10-29 | 上海威研精密科技有限公司 | On-machine multi-view visual detector for micro milling cutter and detection method thereof |
| CN111692968A (en) * | 2020-07-27 | 2020-09-22 | 上海威研精密科技有限公司 | On-machine multi-view visual detector for micro milling cutter and detection method thereof |
| CN113427321A (en) * | 2021-07-22 | 2021-09-24 | 南京理工大学 | Real-time detection device and detection method for cutter used in machining center |
| CN113752088A (en) * | 2021-09-22 | 2021-12-07 | 南京理工大学 | Tool magazine integrated type tool damage detection system and method based on machine vision |
| CN113843659A (en) * | 2021-09-22 | 2021-12-28 | 南京理工大学 | Machine vision-based spindle integrated type cutter damage frequency conversion on-machine detection system and method |
| CN113752088B (en) * | 2021-09-22 | 2024-04-19 | 南京理工大学 | Tool magazine integrated tool damage detection system and method based on machine vision |
| CN113916983A (en) * | 2021-09-30 | 2022-01-11 | 南京中科煜宸激光技术有限公司 | Device for detecting damage of scraper of powder spreading type metal 3D printing equipment |
| CN113916983B (en) * | 2021-09-30 | 2024-07-16 | 南京中科煜宸激光技术有限公司 | Device for detecting damage of scraper of powder-spreading type metal 3D printing equipment |
| CN114378639A (en) * | 2021-12-30 | 2022-04-22 | 科大讯飞华南人工智能研究院（广州）有限公司 | Tool life prediction method and device, storage medium and electronic device |
| CN114378639B (en) * | 2021-12-30 | 2024-01-09 | 科大讯飞华南人工智能研究院（广州）有限公司 | Tool life prediction method and device, storage medium and electronic equipment |
| CN114880814A (en) * | 2022-07-08 | 2022-08-09 | 南通恒强轧辊有限公司 | Roller remanufacturing auxiliary optimization method based on big data |
| CN115302846A (en) * | 2022-10-10 | 2022-11-08 | 苏州鼎佳精密科技股份有限公司 | Die-cutting machine |
| CN115635366A (en) * | 2022-10-12 | 2023-01-24 | 哈尔滨工程大学 | Large machine tool blade acoustic emission online monitoring system and method |
| CN116690314A (en) * | 2023-08-01 | 2023-09-05 | 合肥智谷机电科技有限公司 | Cutter damage detection device based on machine vision detects and compares |
| CN116690314B (en) * | 2023-08-01 | 2023-10-24 | 合肥智谷机电科技有限公司 | Cutter damage detection device based on machine vision detects and compares |
| CN117030857A (en) * | 2023-10-07 | 2023-11-10 | 中科航迈数控软件(深圳)有限公司 | Tool flaw detection method based on phase control ultrasonic waves and related equipment |
| CN117300729A (en) * | 2023-10-12 | 2023-12-29 | 山东润龙精密机床有限公司 | Efficient and precise machining method for normal round holes of special-shaped products based on three-axis machine tool |
| CN117300729B (en) * | 2023-10-12 | 2024-04-30 | 山东润龙精密机床有限公司 | Efficient and precise machining method for normal round holes of special-shaped products based on three-axis machine tool |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110340733A (en) | A kind of damage of Clean Cutting environment bottom tool online with in-place detection system and method | |
| CN108655826B (en) | Cutter abrasion visual inspection device and visual inspection method for numerical control turning | |
| CN102528561B (en) | On-line automatic detection device for detecting wear condition of rotary cutter in the whole processing cycle | |
| RU2545501C2 (en) | Method and device for analysing vibrations and sample database therefor and use of sample database | |
| CN111830906B (en) | On-machine monitoring system for failure state of rotary cutter and detection method thereof | |
| US20090129882A1 (en) | Methods, Systems, and Devices for Monitoring Tools in a Dental Milling Machine | |
| KR102120753B1 (en) | Prognostic method of tool life using vibration properties analysis | |
| CN102226688A (en) | Automatic detection system of engine piston ring | |
| CN108627524B (en) | Roller grinding surface cleaning and machine vision detection system | |
| CN202317881U (en) | On-line anti-collision monitoring device for centerless internal grinder | |
| CN205362716U (en) | Intelligent cutter | |
| JP4182399B2 (en) | Machine tool operation information collection system | |
| CN210678020U (en) | Online and in-place detection system for cutter damage in clean cutting environment | |
| CN108827976A (en) | A kind of amenities stain vision detection system | |
| CN106932410A (en) | A kind of high accuracy minute surface automatic checkout equipment and its method of work | |
| CN211696243U (en) | On-line processing and detecting device for roller | |
| CN106093058A (en) | A kind of press quality amount detecting device and detection method | |
| CN116087224A (en) | Photoelectric detection device and photoelectric detection method | |
| CN208580040U (en) | A kind of amenities stain vision detection system | |
| CN111442720A (en) | Online machining detection device and method for roller | |
| CN203149187U (en) | Machine vision image acquisition reflection device and image acquisition system | |
| CN202041472U (en) | Automatic detecting system of engine piston ring | |
| CN108032064B (en) | The pressure monitoring device of press machine | |
| CN102310265B (en) | Method and device for detecting error for laser cutting processing system | |
| CN219766387U (en) | Damping device for accurate measurement of shearing material identification 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 |