CN110428000A - A kind of milling process energy efficiency state clustering method - Google Patents
A kind of milling process energy efficiency state clustering method Download PDFInfo
- Publication number
- CN110428000A CN110428000A CN201910694126.6A CN201910694126A CN110428000A CN 110428000 A CN110428000 A CN 110428000A CN 201910694126 A CN201910694126 A CN 201910694126A CN 110428000 A CN110428000 A CN 110428000A
- Authority
- CN
- China
- Prior art keywords
- state
- energy efficiency
- milling
- milling process
- cutting
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/005—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles during manufacturing process
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P80/00—Climate change mitigation technologies for sector-wide applications
- Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Machine Tool Sensing Apparatuses (AREA)
Abstract
The invention discloses a kind of milling process energy efficiency state clustering methods based on temperature field temperature image, mainly milling process temperature field temperature image is obtained using thermal imaging system, image processing techniques is combined with milling process, establishes the clustering method of milling process energy efficiency state.Method includes the following steps: step 1, the selection and extraction of temperature image under different milling conditions;Step 2 establishes the corresponding relationship of milling process Yu temperature image, and analyzes included mechanistic information;Step 3: clustering method is established.It can classify to cutting process energy efficiency state using this method, realize that the purpose of the high and low different energy efficiency states of identification cutting process, power-saving technology and intellectual monitoring to manufacture system have reference value.
Description
Technical field
The invention belongs to mechanical manufacturing field more particularly to a kind of milling process energy efficiency state clustering methods.
Background technique
Machining energy efficiency evaluation problem, which has the cost savings and increased quality of manufacturing process, accelerates impetus, right
The sustainable development of human and environment has long-range positive effect, is intelligence manufacture and green manufacturing crossing domain important subject
One of with research hotspot.From the proposition of cutting efficiency concept, so far after nearly development in 30 years, the measure for improving efficiency is main
It is the improvement for focusing machine tool system and technical process, it is specifically excellent around grinding machine structure design, technological procedure analysis or system parameter
The expansion of the research contents such as change is explored.Since cutting process is a complicated system engineering, for numerous processing
Product, manufacturing process is multifarious, and the energy of actual cut process consumption is one and is superimposed upon cutting system intrinsic energy
Dynamic variable on the basis of consumption is difficult designing so the cutting efficiency in cutting system actual moving process is also time-varying
It is accurately modeled and is predicted by offline mode with the process planning stage, even if being known as the cutting based on energy efficient design
System of processing can not also provide quantitative real-time energy efficiency evaluation index online.Even if by the machining of efficiency optimization design
System, due to each component (main shaft, feed drive system etc.) performance degradation be it is nonsynchronous, can also enable system of processing entirety
Effect can not be permanently retained in expected high energy efficiency state.
If cutting process efficiency is regarded as a kind of state (similar conditions of machine tool, cutting tool state, processing quality state etc.)
And cutting process energy efficiency state can be monitored online, there certainly will be important references valence to high energy efficiency problem of cutting is solved
Value, however there is presently no effective, reliable monitoring means, this just provides a new research topic for researcher and " sentences
Whether a disconnected running system of processing is in high energy efficiency state ".
Therefore the new challenge for facing intelligence manufacture and green manufacturing, needs to be grasped cutting process energy efficiency state on-line monitoring side
Method, only combines the demand of the cutting system energy, efficiency, quality and economy, just can guarantee that cutting processing system is run
In high energy efficiency, high quality status, meet the demand of economic, environmental protection, quality.Therefore, those skilled in the art causes
Power is in developing a kind of milling process energy efficiency state recognition methods.
Summary of the invention
In view of the above drawbacks of the prior art, the technical problem to be solved by the present invention is to how application image processing
Method can be classified and be identified to energy efficiency state, and the source of low-energy-efficiency problem occurred to cutting processing system can carry out
It positions and traces to the source, can be monitored online for the energy efficiency state of cutting processing system and theoretical foundation and potential technology solution are provided,
How the energy efficiency state of milling process is differentiated from milling process temperature image.
To achieve the above object, the present invention provides a kind of milling process energy efficiency state clustering method, feature exists
In, comprising the following steps:
Step 1, design milling process energy efficiency state test, to the milling process application thermal imaging under different machining conditions
Instrument carries out Image Acquisition, obtains multiple groups temperature image, the milling state pair according to different moments where any a certain milling process
Temperature image is selected, and is extracted one by one to the general character milling state of all different machining conditions.
Any milling process is all divided into identical five kinds of states in different moments by step 2, and both state one, cutter were initial
Cut workpiece;State two, cutter, which initially cuts workpiece, expires the transition stage of knife cutting workpiece to cutter;State three, cutter is full
Knife cutting workpiece;State four, cutter expire knife cutting workpiece to the transition stage for finally cutting out workpiece;State five, cutter is cut
Workpiece out.Each state corresponds to a width temperature image, and every piece image all includes white, red, yellow, and green and blue five kinds of colors, divides
Not from machine tool chief axis, milling workpiece, milling cutter, material chip and milling area environmental analysis milling process mechanism and temperature image
Information.
Step 3 establishes clustering method based on color of image region and two aspect content of milling process, proposes temperature
The matched high and low classification method of milling process energy efficiency state of image array characteristic value.Each color region of temperature image is divided
Indescribably take, obtain the data matrix of different colours, process-color image Matrix Solving norm to all different milling conditions and
Spectral radius establishes matching condition, is iterated operation, finally determines high and low different energy efficiency states.
Further, the foundation for evaluating milling process energy efficiency state is the milling process temperature figure obtained using thermal imaging system
Picture.
Further, the matrix that image array is five kinds of states and five kinds of colors is defined are as follows:
Θ in formulaiIndicating energy efficiency state temperature figure vector matrix, i indicates milling condition,Indicate that region vector, j indicate
Color of image, 1 corresponding white area, 2 corresponding red areas, 3 corresponding yellow areas, 4 corresponding green areas, 5 corresponding blue regions
Domain, k indicate milling state, 1 corresponding states one, 2 corresponding states two, 3 corresponding states three, 4 corresponding states four, 5 corresponding states five.
Further, it is expressed as respectively from the cluster that milling state and color region are established:
Image array can be expressed as from color region,
It can be expressed as from image array in milling state,
It can be expressed as from image array in milling state,
It can be seen that divide to the image group of different milling conditions can by the expression-form of above-mentioned two matroid
The energy efficiency state for reacting different moments milling process, is embodied in the row vector of two kinds of expression matrix forms, set expression is
A kind of complete milling process under machining condition.
Further, the energy efficiency state assorting process of foundation can indicate are as follows:
Step 1, solution matrix A and its transposed matrix norm.
Matrix A and its transposed matrix indicate are as follows:
Norm is obtained,
Step 2, solution matrix A and its transposed matrix spectral radius.
It obtains,
It obtains,
Step 3 is ranked up different machining conditions by norm and spectral radius value.
Further, the energy efficiency state classifying rules of foundation is expressed as:
Classification state is set as [0,1] two class, implied meaning high energy efficiency state and low-energy-efficiency state, sets matching degree,
Setting
Corresponding milling condition i is 0 state, and output data is stored in bottom half.Similarly,
Corresponding machining condition i is 1 state, and output data is stored in upper half.What appearance mixed | | A | |iWith | | AT||i
Sequence is re-started, continues subregion by step 3, is iterated matching, until being divided into 0,1 liang of class.
Further, the high and low energy efficiency state decision rule of foundation indicates are as follows:
State is determined by spectral radius,
ρ(A)≤||A||
Following matching condition is set,
If | | A | |iandρi(A) ∈ bottom half, it is determined that 0 state;
If | | AT||iandρi(AT) ∈ bottom half, it is determined that 0 state;
IfThen enter interative computation;
IfThen enter interative computation;
If | | A | |iandρi(A) ∈ upper half, it is determined that 1 state;
If | | AT||iandρi(AT) ∈ upper half, it is determined that 1 state;
IfThen enter interative computation;
IfThen enter interative computation.
In this way, temperature image state matrix can be divided into two major classes, if corresponding parameter is low-energy-efficiency state in 0 class, 1
Corresponding parameter is high energy efficiency state in class, then classification terminates, and vice versa.It, then can be with if there are still hybrid parameters in 0,1 class
Obtain several parameter groups in 0,1 class at the corresponding parameter group of extreme value and neighbouring extreme value, still can get in this way high energy efficiency state and
Low-energy-efficiency state, then clustering terminates.
Further, the contrast standard of the high and low energy efficiency state of milling process is with the experience of cutting ratio energy or instantaneous efficiency public affairs
Formula is reference frame,
P (t) in formulacFor any time machine cut power, P (t) is any time lathe input power, ηcFor instantaneous energy
Effect.P is lathe input power, ηsFor Machine Tool Main Drive system energy transfer efficiency, Z is unit time material removing rate, and E is cutting
Than energy.Wherein Z can be expressed during the cutting process by cutting data element, Z=apfv
A in formulapFor cutting-in (back engagement of the cutting edge), f is feed rate, and v is cutting speed.
In better embodiment of the invention, the compression of image is not only realized, but also has expanded color image color
Extraction, conventional method is using red, yellow, blue, and this method uses 5 kinds of colors, and sets 5 kinds of states, makes matrix side
Battle array.
In another better embodiment of the invention, the method for proposition establishes image square matrix, and image is made to solve feature
It is worth, avoids the occurrence of singular matrix and be possibly realized.
The beneficial effects of the present invention are make using the technique study milling process energy efficiency state of temperature field temperature image
Technique study cutting process state based on image procossing is possibly realized.
The present invention obtains and analyzes a large amount of temperature field temperature image patterns by a large number of experiments, greatly exceed routine
Sample size used by image procossing.
The present invention can establish cutting process temperature image library, instruct cutting process various states, as lathe load condition,
Cutter galls the state of the cutting processes such as damage state, cutting quality state.
Technical effect
It is described further below with reference to technical effect of the attached drawing to design of the invention, specific structure and generation, with
It is fully understood from the purpose of the present invention, feature and effect.
Detailed description of the invention
Fig. 1 is the 5 kinds of states and corresponding temperature image graph that milling process of the invention defines;
Fig. 2 is the temperature image moment system of battle formations of the invention;
Fig. 3 is the matrix diagram of low-energy-efficiency state temperature figure original image of the invention;
Fig. 4 is the matrix diagram after low-energy-efficiency state temperature figure image procossing of the invention;
Fig. 5 is that each color element of low-energy-efficiency state temperature figure of the invention extracts the result figure after decomposing;
Fig. 6 is the matrix diagram of high energy efficiency state temperature figure original image of the invention;
Fig. 7 is the matrix diagram after high energy efficiency state temperature figure image procossing of the invention;
Fig. 8 is that each color element of high energy efficiency state temperature figure of the invention extracts the result figure after decomposing;
Fig. 9 is cluster algorithm flow chart of the invention;
Figure 10 is different milling condition classification results figures of the invention.
Specific embodiment
Multiple preferred embodiments of the invention are introduced below with reference to Figure of description, keep its technology contents more clear and just
In understanding.The present invention can be emerged from by many various forms of embodiments, and protection scope of the present invention not only limits
The embodiment that Yu Wenzhong is mentioned.
In the accompanying drawings, the identical component of structure is indicated with same numbers label, everywhere the similar component of structure or function with
Like numeral label indicates.The size and thickness of each component shown in the drawings are to be arbitrarily shown, and there is no limit by the present invention
The size and thickness of each component.Apparent in order to make to illustrate, some places suitably exaggerate the thickness of component in attached drawing.
As shown in Figure 1,5 kinds of states that milling process of the present invention defines include: state one, cutter initially cuts workpiece;Shape
State two, cutter, which initially cuts workpiece, expires the transition stage of knife cutting workpiece to cutter;State three, cutter expire knife cutting work
Part;State four, cutter expire knife cutting workpiece to the transition stage for finally cutting out workpiece;State five, cutter cuts out workpiece.
State one, main shaft: initial cuts, main shaft load is smaller, and temperature rise is low;Workpiece: workpiece is placed in environment, temperature with
Environment temperature is similar, and cutter starts contact workpiece, and workpiece and environment heat exchange are very fast, and temperature rise is lower;Cutter: tool temperature and ring
Border temperature is approximate, and cutter starts contact workpiece, and part cutting edge is cut, and true cutting output is small, load it is small, cutter temperature rise compared with
It is small;Chip: Chip Morphology is also imperfect under this state, the process gradually grown up is presented, carrying heat is little, with certain speed
During outflow, heat exchange is carried out with external environment, heat loss is fast;Environment: original state ambient temperature-stable, it is unchanged.
Easily there is cutter flutter, impact and disrepair phenomenon, and system stability since cutter dynamic balance property is unstable in this state
There are fluctuation, with than that can judge efficiency, unit time material removal volume is less, load lower, there are different degrees of vibrations
Dynamic and noise, efficiency are lower.
State two, main shaft: less, slightly there is light green color state representation in temperature figure for temperature rise variation;Workpiece: there is obvious become
Change, cutting process is cut according to true cutting output, and workpiece temperature at cutting position is high, and radiating rate is slow, cuts position
It sets and white, red, yellow and green is around presented, blue is presented in the workpiece portion far from cutting position, this is Heat transfer law
Embodiment;Cutter: the trend being stepped up is presented in temperature;Chip: having significant difference with state one, and chip carries heat and increases,
But most heats have carried out quick heat exchange with the high-speed motion of chip and the external world, and color shows as yellow and green
Color;Environment: the change of temperature field of environment is little;The load of this state cutter increases, and cutting tends to gradually stable process, the highest temperature
Degree appears in cutting position, although most heats, heat loss of the cutting position heat loss compared with chip are taken away in chip
Slower, cutting position is also easy to produce built-up edge and tool wear phenomenon, and true cutting output is caused to change, this is also the unit time
The reason of practical removal volume of material and theoretical removing body accumulate in difference, while being also the changing original of material removal volume
Cause, so that influencing efficiency there is time-varying state, with than that can judge efficiency, this state is high compared with original state efficiency, the reason is that
In the case where load variation less, unit time material removal volume is increased.
State three, main shaft: temperature change is little, slightly shallow on color intensity compared with previous state;Workpiece: temperature field
The variation of temperature figure is the most obvious, and red, yellow and green is presented in the workpiece overwhelming majority, and red specific gravity is more, this shows cutting
Cutting position has a large amount of heat transfer to workpiece, and for workpiece heat-sinking capability not as good as chip and cutter, main cause is that workpiece is in solid
Determine the state of clamping;Cutter: since this state is full knife cutting, cutter is completely in workpiece machining surface regional scope,
Cutting process is transferred to the heat of cutter, and state heat dissipation effect is deteriorated than before under this state, because over time and space
State before cutter is exposed to environment division relatively has occurred that the change of basic form, this just considerably increases tool wear
A possibility that generating with built-up edge, unit time material removal volume is also time-varying;Chip: it is compared with state second is that distinguishing,
With the propulsion of cutting process, the dynamic characteristic of the state chip is variation, the work that state one and two chip of state are subject to
The cutting and friction for firmly mostling come from cutter, have the friction with workpiece concurrently, and have arrived state three, the friction of chip and workpiece
Power significantly increases, and affects the direction of chip campaign and the heat of carrying, and the chip being attached on workpiece and cutter also starts to increase
It is more, cutting region temperature field is had an impact;Environment: cutting region blue portion color intensity is thin out, and environment temperature is varied;This
State cutter reaches stable cutting state, and workpiece material is by a large amount of, quick removal, and system entirety cutting temperature is all in acceleration
The stage of liter, this state are to consider the most important phase of cutting system element (machine tool chief axis load and stability, Tool in Cutting
Energy, workpiece cutting ability and material build-in attribute, the reasonability of cutting parameter).With than that can judge efficiency, this state is that efficiency is high
Stage.
State four, main shaft: temperature rise is compared with state one, two, three it has been seen in that having the state substantially changeing, main shaft is at one section
Rotation function acting, functional component loss all produce a large amount of heat in time, and main transmission energy transfer efficiency changes
Become;Workpiece: it can be seen that four workpiece temperature figure feature of state is similar with state two, but from workpiece bulk temperature state, respectively
Partial region temperature is higher than state two, this is because the heat that workpiece is hoarded is also more and more, outside with the progress of cutting process
Boundary's environment temperature is also riseing, and heat loss ability dies down, and more important is it can be seen that cutting position heat from state two
It spends diffusion zone and scalloped profile is presented, temperature boundary is obvious and clear, and has arrived state four, it can be seen that this Partial Feature is
Significantly different, yellow, green and red area obscurity boundary are also mingled with RED sector in yellow and green area, this is because
Caused by two o'clock reason, first is that the heat transfer process of workpiece area and ambient enviroment difference occurs, and second is that part is splashed
Chip be scattered in the region of workpiece machined surface, work surface and other surfaces, cause workpiece bulk temperature field to present
This phenomenon;Cutter: this state tool temperature starts to be declined, because cutterhead, in rotary course, cutter is at a time
It far from cutting workpiece, is directly contacted with environment, heat scatters and disappears fast when relatively expiring knife cutting;Chip: compared with the stream of three chip of state
It does well more at random, the reason is that it is more unstable with the friction of workpiece and the kinetic characteristics of cutter, and chip carrying
Heat is also not fixed, and cutting-tool wear state, cutter mark and workpiece deformation all constantly influence the heat of chip;Environment: this shape
State and three temperature field environment difference of state are little;This state cutter mechanical characteristic is similar with state two, the removal of unit time material
Volume is gradually declining, the reason is that cutter is imperfect cutting to the cutting of workpiece, with than that can judge efficiency, and this state efficiency
To be lower than state two, the reason is that system temperature rise aggravates, the generation and loss of heat are increased;
State five, main shaft: by the cutting of front, machine-tool spindle system temperature rise fever reaches peak value, can be with from temperature figure
Find out that end-state main transmission becomes green state from blue color states before;Workpiece: comparing this state and state one, discovery
Significant difference is had already appeared, presents and is incremented by the temperature profile of cutting terminal from cutting beginning by complete cutting process workpiece
Rule, workpiece storage have contained heat, this was determined by the workpiece heat dissipation time;Cutter: this state tool temperature is dropped compared with previous state
It is low;Chip: this state Chip Morphology attenuates, becomes smaller, and outflow direction is also unstable, and carrying heat, state is reduced obviously than before,
Until final cutting terminates chipless outflow;Environment: environment temperature changes significantly, and by complete cutting process, cutting system is hard
Heat exchange has occurred in the heat and cutting region environment that part component part generates, and nearby temperature expands to far from cutting region environment for cutting region
It dissipates;This state is similar with the cutting process of state one, and efficiency is lower compared with state one, and reason declines comprising system temperature rise, main shaft performance
It moves back, tool wear, workpiece flexible deformation.
As shown in Fig. 2, the present invention establishes milling process temperature image array, box representing matrix in figure, row in matrix
Element is followed successively by 5 kinds of states of definition.
As shown in figure 3, the present invention is extracted the original temperature image of low-energy-efficiency state by test, box indicates square in figure
Battle array.
As shown in figure 4, the present invention carries out color boundaries processing to original temperature image, high energy efficiency state temperature figure is obtained
As treated form, box representing matrix in figure, image includes 5 kinds of colors, is followed successively by purple, red from high to low by temperature
Color, yellow, green and blue, clear in order to show, purple therein is the white characterized in original image.
As shown in figure 5, extracting respectively to 5 kinds of colors, low-energy-efficiency state temperature picture breakdown matrix, side in figure are established
Frame representing matrix, wherein row element successively represents 5 kinds of milling shapes of state one, state two, state three, state four and state five
State, column element successively represent 5 kinds of purple, red, yellow, green and blue temperature field color regions, and 0 in matrix indicates the shape
State is without corresponding color region.
As shown in fig. 6, the present invention is extracted the original temperature image of high energy efficiency state by test, box indicates square in figure
Battle array.
As shown in fig. 7, the present invention carries out color boundaries processing to the original temperature image of high energy efficiency state, high energy efficiency is obtained
Form after state temperature image procossing, box representing matrix in figure, image include 5 kinds of colors, from high to low successively by temperature
Clear in order to show for purple, red, yellow, green and blue, purple therein is the white characterized in original image.
As shown in figure 8, extracting respectively to 5 kinds of colors, high energy efficiency state temperature picture breakdown matrix, side in figure are established
Frame representing matrix, wherein row element successively represents 5 kinds of milling shapes of state one, state two, state three, state four and state five
State, column element successively represent 5 kinds of purple, red, yellow, green and blue temperature field color regions, and 0 in matrix indicates the shape
State is without corresponding color region.
As shown in figure 9, the present invention provides a kind of, the milling process energy efficiency state cluster based on temperature field temperature image is divided
Analysis method, which comprises the following steps:
Step 1, design milling process energy efficiency state test, to the milling process application thermal imaging system under different machining conditions
Image Acquisition is carried out, multiple groups temperature image is obtained, the milling state according to different moments where any a certain milling process is to heat
Degree image is selected, and is extracted one by one to the general character milling state of all different machining conditions.
Step 2, any milling process is all divided into identical five kinds of states in different moments, both state one, cutter were initial
Cut workpiece;State two, cutter, which initially cuts workpiece, expires the transition stage of knife cutting workpiece to cutter;State three, cutter is full
Knife cutting workpiece;State four, cutter expire knife cutting workpiece to the transition stage for finally cutting out workpiece;State five, cutter is cut
Workpiece out.Each state corresponds to a width temperature image, and every piece image all includes white, red, yellow, and green and blue five kinds of colors, divides
Not from machine tool chief axis, milling workpiece, milling cutter, material chip and milling area environmental analysis milling process mechanism and temperature image
Information.
Step 3, clustering method is established based on color of image region and two aspect content of milling process, proposes temperature figure
As the matched high and low classification method of milling process energy efficiency state of matrix exgenvalue.Each color region of temperature image is distinguished
It extracts, obtains the data matrix of different colours, to the process-color image Matrix Solving norm and spectrum of all different milling conditions
Radius establishes matching condition, is iterated operation, finally determines high and low different energy efficiency states.
The foundation of evaluation milling process energy efficiency state is the milling process temperature image obtained using thermal imaging system.
Define the matrix that image array is five kinds of states and five kinds of colors are as follows:
Θ in formulaiIndicating energy efficiency state temperature figure vector matrix, i indicates milling condition,Indicate that region vector, j indicate
Color of image, 1 corresponding white area, 2 corresponding red areas, 3 corresponding yellow areas, 4 corresponding green areas, 5 corresponding blue regions
Domain, k indicate milling state, 1 corresponding states one, 2 corresponding states two, 3 corresponding states three, 4 corresponding states four, 5 corresponding states five.
Be expressed as respectively from the cluster that milling state and color region are established: image array can indicate from color region
For, it can be expressed as from image array in milling state,
It can be seen that divide to the image group of different milling conditions can by the expression-form of above-mentioned two matroid
The energy efficiency state for reacting different moments milling process, is embodied in the row vector of two kinds of expression matrix forms, set expression is
A kind of complete milling process under machining condition.
The energy efficiency state assorting process of foundation can indicate are as follows:
Step 1, solution matrix A and its transposed matrix norm.
Matrix A and its transposed matrix indicate are as follows:
Norm is obtained,
Step 2, solution matrix A and its transposed matrix spectral radius.
It obtains,
It obtains,
Step 3 is ranked up different machining conditions by norm and spectral radius value.
The energy efficiency state classifying rules of foundation is expressed as: set classification state as [0,1] two class, implied meaning high energy efficiency state and
Low-energy-efficiency state sets matching degree,
Setting
Corresponding milling condition i is 0 state, and output data is stored in bottom half.Similarly,
Corresponding machining condition i is 1 state, and output data is stored in upper half.What appearance mixed | | A | |iWith | | AT||i
Sequence is re-started, continues subregion by step 3, is iterated matching, until being divided into 0,1 liang of class.
The high and low energy efficiency state decision rule established indicates are as follows: state is determined by spectral radius,
ρ(A)≤||A||
Following matching condition is set,
If | | A | |iandρi(A) ∈ bottom half, it is determined that 0 state;
If | | AT||iandρi(AT) ∈ bottom half, it is determined that 0 state;
IfThen enter interative computation;
IfThen enter interative computation;
If | | A | |iandρi(A) ∈ upper half, it is determined that 1 state;
If | | AT||iandρi(AT) ∈ upper half, it is determined that 1 state;
IfThen enter interative computation;
IfThen enter interative computation.
In this way, temperature image state matrix can be divided into two major classes, if corresponding parameter is low-energy-efficiency state in 0 class, 1
Corresponding parameter is high energy efficiency state in class, then classification terminates, and vice versa.It, then can be with if there are still hybrid parameters in 0,1 class
Obtain several parameter groups in 0,1 class at the corresponding parameter group of extreme value and neighbouring extreme value, still can get in this way high energy efficiency state and
Low-energy-efficiency state, then clustering terminates.
The contrast standard of the high and low energy efficiency state of milling process with the empirical equation of cutting ratio energy or instantaneous efficiency be with reference to according to
According to,
P (t) in formulacFor any time machine cut power, P (t) is any time lathe input power, ηcFor instantaneous energy
Effect.P is lathe input power, ηsFor Machine Tool Main Drive system energy transfer efficiency, Z is unit time material removing rate, and E is cutting
Than energy.Wherein Z can be expressed during the cutting process by cutting data element,
Z=apfv
A in formulapFor cutting-in (back engagement of the cutting edge), f is feed rate, and v is cutting speed.
The compression of image is not only realized by this method, but also has expanded the extraction of color image color, conventional method
It is using red, yellow, blue, and this method uses 5 kinds of colors, and sets 5 kinds of states, makes matrix square matrix.The method of proposition is built
Image square matrix has been found, has made it possible that image solves characteristic value, avoids the occurrence of singular matrix.Using the side of temperature field temperature image
Method research milling process energy efficiency state, makes it possible the technique study cutting process state based on image procossing.By a large amount of
Test, obtains and analyzes a large amount of temperature field temperature image patterns, greatly exceeds sample number used by normal image is handled
Amount.It can establish cutting process temperature image library, instruct cutting process various states, as lathe load condition, cutter gall damage
The state of the cutting processes such as state, cutting quality state.
As shown in Figure 10, the present invention provides the milling process energy efficiency state classification results based on temperature field temperature image.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that the ordinary skill of this field is without wound
The property made labour, which according to the present invention can conceive, makes many modifications and variations.Therefore, all technician in the art
Pass through the available technology of logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Scheme, all should be within the scope of protection determined by the claims.
Claims (8)
1. a kind of milling process energy efficiency state clustering method, which comprises the following steps:
Step 1, design milling process energy efficiency state test, to the milling process application thermal imaging system under different machining conditions into
Row Image Acquisition obtains multiple groups temperature image, and the milling state according to different moments where any a certain milling process is to temperature
Image is selected, and is extracted one by one to the general character milling state of all different machining conditions;
Any milling process is all divided into identical five kinds of states in different moments by step 2, and both state one, cutter were initially cut
Workpiece;State two, cutter, which initially cuts workpiece, expires the transition stage of knife cutting workpiece to cutter;State three, cutter expire cutterhead
Cutting workpiece;State four, cutter expire knife cutting workpiece to the transition stage for finally cutting out workpiece;State five, cutter cuts out work
Part;Each state corresponds to a width temperature image, and every piece image all includes white, red, yellow, and green and blue five kinds of colors, respectively from
Machine tool chief axis, milling workpiece, milling cutter, material chip and milling area environmental analysis milling process mechanism and temperature image letter
Breath;
Step 3 establishes clustering method based on color of image region and two aspect content of milling process, proposes temperature image
The matched high and low classification method of milling process energy efficiency state of matrix exgenvalue;Each color region of temperature image is mentioned respectively
It takes, obtains the data matrix of different colours, to the process-color image Matrix Solving norm and spectrum half of all different milling conditions
Diameter establishes matching condition, is iterated operation, finally determines high and low different energy efficiency states.
2. milling process energy efficiency state clustering method as described in claim 1, which is characterized in that evaluation milling process energy
The foundation of effect state is the milling process temperature image obtained using thermal imaging system.
3. milling process energy efficiency state clustering method as described in claim 1, which is characterized in that defining image array is
The matrix of five kinds of states and five kinds of colors are as follows:
Θ in formulaiIndicating energy efficiency state temperature figure vector matrix, i indicates milling condition,Indicate that region vector, j indicate image face
Color, 1 corresponding white area, 2 corresponding red areas, 3 corresponding yellow areas, 4 corresponding green areas, 5 corresponding blue regions, k table
Show milling state, 1 corresponding states one, 2 corresponding states two, 3 corresponding states three, 4 corresponding states four, 5 corresponding states five.
4. milling process energy efficiency state clustering method as claimed in claim 3, which is characterized in that respectively from the milling
The cluster that state and the color region are established:
Image array can be expressed as on the color region,
Image array can be expressed as in the milling state,
5. milling process energy efficiency state clustering method as claimed in claim 4, which is characterized in that the energy efficiency state of foundation
Assorting process can indicate are as follows:
Step 1, solution matrix A and its transposed matrix number;
Matrix A and its transposed representation are as follows:
Norm is obtained,
Step 2, solution matrix A and its transposition spectral radius;
It obtains,
It obtains,
Step 3 is ranked up different machining conditions by norm and spectral radius value;
6. milling process energy efficiency state clustering method as claimed in claim 5, which is characterized in that the energy efficiency state of foundation
Classifying rules is expressed as:
Classification state is set as [0,1] two class, implied meaning high energy efficiency state and low-energy-efficiency state, sets matching degree,
Setting
Corresponding milling condition i is 0 state, and output data is stored in bottom half;Similarly,
Corresponding machining condition i is 1 state, and output data is stored in upper half;What appearance mixed | | A | |iWith | | AT||iAgain
It is ranked up, continues subregion by step 3, be iterated matching, until being divided into 0,1 liang of class.
7. milling process energy efficiency state clustering method as claimed in claim 6, which is characterized in that the high and low energy of foundation
Effect state decision rule indicates are as follows:
State is determined by spectral radius,
ρ(A)≤||A||
Following matching condition is set,
If | | A | |iandρi(A) ∈ bottom half, it is determined that 0 state;
If | | AT||iandρi(AT) ∈ bottom half, it is determined that 0 state;
IfBottom half then enters interative computation;
IfBottom half then enters interative computation;
If | | A | |iandρi(A) ∈ upper half, it is determined that 1 state;
If | | AT||iandρi(AT) ∈ upper half, it is determined that 1 state;
IfUpper half then enters interative computation;
IfUpper half then enters interative computation;
In this way, temperature image state matrix can be divided into two major classes, if corresponding parameter is low-energy-efficiency state in 0 class, in 1 class
Corresponding parameter is high energy efficiency state, then classification terminates, and vice versa;If in 0,1 class, there are still hybrid parameters, then can obtain
The corresponding parameter group of extreme value and several parameter groups at extreme value, still can get high energy efficiency state and low energy in this way in 0,1 class
Effect state, then clustering terminates.
8. milling process energy efficiency state clustering method as claimed in claim 7, which is characterized in that milling process is high and low
The contrast standard of energy efficiency state using the empirical equation of cutting ratio energy or instantaneous efficiency as reference frame,
P (t) in formulacFor any time machine cut power, P (t) is any time lathe input power, ηcFor instantaneous efficiency;P
For lathe input power, ηsFor Machine Tool Main Drive system energy transfer efficiency, Z is unit time material removing rate, and E is cutting ratio
Energy;Wherein Z can be expressed during the cutting process by cutting data element,
Z=apfv
A in formulapFor cutting-in (back engagement of the cutting edge), f is feed rate, and v is cutting speed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910694126.6A CN110428000B (en) | 2019-07-30 | 2019-07-30 | Milling process energy efficiency state clustering analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910694126.6A CN110428000B (en) | 2019-07-30 | 2019-07-30 | Milling process energy efficiency state clustering analysis method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110428000A true CN110428000A (en) | 2019-11-08 |
CN110428000B CN110428000B (en) | 2022-07-08 |
Family
ID=68413140
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910694126.6A Active CN110428000B (en) | 2019-07-30 | 2019-07-30 | Milling process energy efficiency state clustering analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110428000B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111158313A (en) * | 2020-01-14 | 2020-05-15 | 上海交通大学 | Method for modeling energy consumption and optimizing machining process of numerical control machine tool |
CN112270282A (en) * | 2020-11-03 | 2021-01-26 | 华北电力大学 | Power signal filtering method and system by utilizing matrix spectral mode |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5463470A (en) * | 1991-10-09 | 1995-10-31 | Fuji Photo Film Co., Ltd. | Methods of collecting photometric image data and determining light exposure by extracting feature image data from an original image |
CN104615077A (en) * | 2015-01-07 | 2015-05-13 | 重庆大学 | Efficient energy-saving optimizing method for numerical control milling processing process parameters based on Taguchi method |
CN105631481A (en) * | 2016-01-07 | 2016-06-01 | 西安交通大学 | Ferrograph abrasive particle composite characteristic construction method based on heredity programming |
CN107341503A (en) * | 2017-06-02 | 2017-11-10 | 上海交通大学 | The recognition methods of multi-source energy efficiency state is directed in a kind of working angles |
CN107507194A (en) * | 2017-08-07 | 2017-12-22 | 广东电网有限责任公司珠海供电局 | A kind of insulator chain fault detection method based on infrared image temperature distributing rule and BP neural network |
-
2019
- 2019-07-30 CN CN201910694126.6A patent/CN110428000B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5463470A (en) * | 1991-10-09 | 1995-10-31 | Fuji Photo Film Co., Ltd. | Methods of collecting photometric image data and determining light exposure by extracting feature image data from an original image |
CN104615077A (en) * | 2015-01-07 | 2015-05-13 | 重庆大学 | Efficient energy-saving optimizing method for numerical control milling processing process parameters based on Taguchi method |
CN105631481A (en) * | 2016-01-07 | 2016-06-01 | 西安交通大学 | Ferrograph abrasive particle composite characteristic construction method based on heredity programming |
CN107341503A (en) * | 2017-06-02 | 2017-11-10 | 上海交通大学 | The recognition methods of multi-source energy efficiency state is directed in a kind of working angles |
CN107507194A (en) * | 2017-08-07 | 2017-12-22 | 广东电网有限责任公司珠海供电局 | A kind of insulator chain fault detection method based on infrared image temperature distributing rule and BP neural network |
Non-Patent Citations (2)
Title |
---|
孙岳,于占江,许金凯,于化东: "基于振源识别的高速微铣削机床状态研究", 《振动与冲击》 * |
计时鸣,张立彬,袁巨龙,周晓军: "基于计算机泛视觉的刀具状态监测方法研究", 《中国会议》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111158313A (en) * | 2020-01-14 | 2020-05-15 | 上海交通大学 | Method for modeling energy consumption and optimizing machining process of numerical control machine tool |
CN111158313B (en) * | 2020-01-14 | 2022-11-29 | 上海交通大学 | Method for modeling energy consumption and optimizing machining process of numerical control machine tool |
CN112270282A (en) * | 2020-11-03 | 2021-01-26 | 华北电力大学 | Power signal filtering method and system by utilizing matrix spectral mode |
CN112270282B (en) * | 2020-11-03 | 2021-12-10 | 华北电力大学 | Power signal filtering method and system by utilizing matrix spectral mode |
Also Published As
Publication number | Publication date |
---|---|
CN110428000B (en) | 2022-07-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110428000A (en) | A kind of milling process energy efficiency state clustering method | |
Diaz-Rozo et al. | Machine learning-based CPS for clustering high throughput machining cycle conditions | |
Patange et al. | Milling cutter condition monitoring using machine learning approach | |
Ou et al. | Intelligent analysis of tool wear state using stacked denoising autoencoder with online sequential-extreme learning machine | |
Hu et al. | Tool wear monitoring in milling of titanium alloy Ti–6Al–4 V under MQL conditions based on a new tool wear categorization method | |
Saravanamurugan et al. | Chatter prediction in boring process using machine learning technique | |
CN111273607B (en) | Spark-based numerical control machine tool running state monitoring method | |
Mohammadi et al. | Machine learning for quality prediction in abrasion-resistant material manufacturing process | |
CN113369994A (en) | Cutter state monitoring method in high-speed milling process | |
Nguyen et al. | Optimal feature selection using genetic algorithm for mechanical fault detection of induction motor | |
Qian et al. | Open-world social event classification | |
Unver et al. | Exploring the potential of transfer learning for chatter detection | |
Fu et al. | Symmetric graph regularized constraint propagation | |
Chen et al. | Monitoring grinding wheel redress-life using support vector machines | |
Dimić et al. | Improvement of the accuracy of prediction using unsupervised discretization method: educational data set case study | |
Campatelli | Analysis of the environmental impact for a turning operation of AISI 1040 steel | |
Duo et al. | Surface roughness assessment on hole drilled through the identification and clustering of relevant external and internal signal statistical features | |
CN109396956A (en) | A kind of chain digital control gear hobbing machine hobboing cutter state intelligent monitoring method | |
Sun et al. | Research on robot target recognition based on deep learning | |
CN111611946A (en) | Micro milling cutter abrasion on-line monitoring method based on principal component analysis and BP neural network | |
Shao | Recognition of control chart patterns using decision tree of multi-class SVM | |
Dutta et al. | Assessment of machining features for tool condition monitoring in face milling using an artificial neural network | |
Martins et al. | Neural networks models for wear patterns recognition of single-point dresser | |
Zaqout et al. | Pixel-based skin color detection technique | |
Watanabe et al. | Development of a multi-granularity design support system using hierarchical clustering for non-dominated solutions |
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 |