CN107398612A - A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis - Google Patents
A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis Download PDFInfo
- Publication number
- CN107398612A CN107398612A CN201710823416.7A CN201710823416A CN107398612A CN 107398612 A CN107398612 A CN 107398612A CN 201710823416 A CN201710823416 A CN 201710823416A CN 107398612 A CN107398612 A CN 107398612A
- Authority
- CN
- China
- Prior art keywords
- cluster analysis
- host computer
- electric spark
- cluster
- electric
- 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
- B23H—WORKING OF METAL BY THE ACTION OF A HIGH CONCENTRATION OF ELECTRIC CURRENT ON A WORKPIECE USING AN ELECTRODE WHICH TAKES THE PLACE OF A TOOL; SUCH WORKING COMBINED WITH OTHER FORMS OF WORKING OF METAL
- B23H7/00—Processes or apparatus applicable to both electrical discharge machining and electrochemical machining
- B23H7/14—Electric circuits specially adapted therefor, e.g. power supply
- B23H7/20—Electric circuits specially adapted therefor, e.g. power supply for programme-control, e.g. adaptive
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Mechanical Engineering (AREA)
- Electrical Discharge Machining, Electrochemical Machining, And Combined Machining (AREA)
Abstract
A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis, are related to electrical discharge machining.The detection means includes electric spark machine tool, electrical signal detection collector, cluster analysis device, servo controller and host computer；The electrical signal detection collector includes amplifying circuit and isolation optocoupler etc.；Cluster analysis device includes STM32 single-chip microcomputers and serial port module etc., electric spark machine tool passes through cable connection amplifying circuit, isolation module is sequentially connected amplifying circuit, single-chip microcomputer, serial port module and host computer, and the host computer connects motion control card, motion control card connection electric spark machine tool.The electric signal of the electric spark machine tool carries out cluster analysis by the input of electrical signal detection collector in cluster analysis device, SERVO CONTROL is carried out in the cluster result output termination servo controller of cluster analysis device, the relevant signal output of servo controller connects host computer, by being shown, being controlled and being stored in host computer.
Description
Technical field
The present invention relates to electrical discharge machining, is examined more particularly, to a kind of electric spark clearance discharge condition based on cluster analysis
Survey apparatus and method.
Background technology
Electrical discharge machining is to reach removal workpiece by instantaneous high-temperature caused by the pulse feature spark discharge of electrode and workpiece
The special process method on surface, to reach the requirement with definite shape and surface roughness.Electrical discharge machining is not strong by workpiece
Degree, the influence of hardness, and without the cutting force of macroscopic view.Be particularly suitable for use in the processing of unmanageable complicated shape.In aviation, boat
My god, automobile, electronics, mould all there is very big application.
The discharge condition in gap generally has zero load, spark, three kinds of short circuit.It is more perfect to also have stable electric arc and unstable
Electric arc totally five kinds of discharge conditions.There are voltage, no current when general unloaded, short circuit has electric current, no-voltage, and spark has voltage to have electricity again
Signal is flowed, can now distinguish spark, unstable arc, stable electric arc again according to voltage x current.
That is, different gap states correspond to different voltage and current signals, that is, different features be present.
Found by the literature search to prior art, a kind of electricity based on wavelet transformation disclosed in Chinese patent 200910045050
Spark gap condition checkout gear and method, voltage can only be tested and analyzed, moreover, can only also be divided into intrinsic several
Kind state.Cluster analysis is a kind of analysis method for the multivariate statistics classification classified to data.It can not supervised
Under conditions of data are classified automatically and reasonably.It need not give a set goal, can automatic mining go out number
According to hiding feature.
The signals such as processing method, the voltage x current of different conditions such as present electric spark drilling, milling, wire cutting are all different,
During SERVO CONTROL, every kind of processing method is required for rule of thumb selected threshold under different processing environments, and the present invention being capable of root
According to the gap electric signal automatic identification gap state and clearance features during processing, foundation is provided for SERVO CONTROL, improves processing effect
Rate.
The content of the invention
It is an object of the invention to provide a kind of electric spark clearance based on cluster analysis that foundation is provided for SERVO CONTROL
Discharge condition detection apparatus and method.
The electric spark clearance discharge condition detection apparatus based on cluster analysis includes electric spark machine tool, electric signal
Detect collector, cluster analysis device, servo controller and host computer；The electrical signal detection collector include amplifying circuit and every
From optocoupler etc.；Cluster analysis device includes STM32 single-chip microcomputers and serial port module etc., and electric spark machine tool is amplified by cable connection
Circuit, isolation module are sequentially connected amplifying circuit, single-chip microcomputer, serial port module and host computer, and the host computer connects motion control
Card, motion control card connection electric spark machine tool.The electric signal of the electric spark machine tool is gathered by electrical signal detection
Device input carries out cluster analysis in cluster analysis device, is carried out in the cluster result output termination servo controller of cluster analysis device
SERVO CONTROL, the relevant signal output of servo controller connect host computer, by being shown, being controlled and being stored in host computer.
The single-chip microcomputer can use STM32 single-chip microcomputers.
The discharge condition in electrical spark working man-hour is divided into five types：Open-circuit condition, normal spark discharge condition, transferred arc
Discharge condition, stable arc discharge state and short-circuit condition, various states correspond to different voltage x current states, pass through respectively
The detection method and device that the present invention relates to, can exactly by interpolar pulse be divided into above-mentioned spark discharge, transferred arc,
Stable electric arc, open circuit, short-circuit five classes.
The electric spark clearance discharge condition detection method based on cluster analysis concretely comprises the following steps：
1) electrical discharge machining is carried out first, and the gap electric signal of processing is input in electrical signal detection collector and sampled,
Corresponding data are obtained, the electric signal is electric current, voltage；
2) the gap electrical signal data collected is input in cluster analysis device and carries out cluster analysis, while be sent to
Position machine makes display and storage；
3) in cluster analysis device, the electrical signal data of each time interval is one group, and by data prediction, removal is made an uproar
Sound point, each gap electrical signal data are a data point；
4) by clustering algorithm, each single data point as a class, the distance between each point will be calculated, away from
Cluster is classified as near point；
5) select clusters number, carry out the calculating in step 4), obtain cluster to the end, the point of each distribution by cluster,
Become the class of several aggregations, each class corresponds to a kind of gap state, the machining state in each time interval is divided into self-defined
Several classes, and obtained the center of class；
6) cluster result of step 5) is input in servo controller, makes SERVO CONTROL.
The present invention is proposed using cluster analysis to identify gap state different in electrical discharge machining, and according to gap-like
State carries out SERVO CONTROL.
On the basis of the identification of spark discharge state-detection and discharge condition characteristic, for classifying in electrical discharge machining
Threshold value is difficult to situation about determining, the present invention is proposed using cluster analysis to identify each state in electrical discharge machining, and with
This is according to progress SERVO CONTROL.Cluster analysis can cluster according to the feature of image watermarking to data, the number of like attribute
According to that will be classified as one kind, electric signal corresponding to different states (electric current, voltage) is polymerized to different classes.Can be preferably by electric fire
Colored state classifies, and so clearly can carry out SERVO CONTROL to each state.
Electric spark clearance discharge condition detection method based on cluster analysis involved in the present invention, by gap telecommunications
Number sampled, and cluster analysis is carried out to sampled signal.Oneself classification number of needs and the central point of each class are obtained, and
Reference point using this central point as each state, there is provided controlled accordingly to servo controller.
The present invention provides a kind of quantization means to processing gap state detection, and each add can be obtained by experimental data
Work status information.By carrying out real-time cluster analysis to data, each section of process gap electric signal can be classified
And its central point is obtained, provide foundation for follow-up SERVO CONTROL.It is also possible to the data of whole section of process are gathered
Class, the distribution center's point and its distribution situation of each state are obtained, foundation is provided for processing result analysis.
Brief description of the drawings
Fig. 1 is that the structure of the electric spark clearance discharge condition detection apparatus embodiment of the present invention based on cluster analysis is shown
It is intended to.
Fig. 2 is voltage waveform view corresponding to spark discharge state.
Fig. 3 is current waveform schematic diagram corresponding to spark discharge state.
In figs 2 and 3, it should be tried one's best during processing and keep it in spark discharge state.
Embodiment
Embodiments of the invention are elaborated below in conjunction with the accompanying drawings：The present embodiment using technical solution of the present invention before
Put and implemented, give detailed embodiment and specific operating process, but protection scope of the present invention is not limited to down
The embodiment stated.
As shown in figure 1, the electric spark clearance discharge condition detection apparatus embodiment bag of the present invention based on cluster analysis
Include：Electric spark machine tool 1, electrical signal detection collector 2, cluster analysis device 3, servo controller 4 and host computer 5.Wherein, it is electric
Signal detection collector 2 includes amplifying circuit, isolation optocoupler etc.；Cluster analysis device 3 includes STM32 single-chip microcomputers, serial port module etc..
Connected mode is：For electric spark machine tool 1 by cable connection amplifying circuit, isolation module connects amplifying circuit, then connects STM32
Single-chip microcomputer, host computer is then connected by serial port module, motion control card is connected by host computer, finally connected from motion control card
Electric spark machine tool 1.
As shown in Figures 2 and 3, the discharge condition in electrical spark working man-hour is divided into five types：Open-circuit condition, normal spark electric discharge
State, transient arc discharge state, stable arc discharge state and short-circuit condition, various states correspond to different voltage respectively
Current status, by the detection method and device that the present invention relates to, interpolar pulse can be divided into above-mentioned spark exactly
Electric discharge, transferred arc, stable electric arc, open circuit, short-circuit five classes.
The electric spark method of servo-controlling based on cluster analysis that the present embodiment is related to, passes through the height in STM32 single-chip microcomputers
Fast modulus conversion chip is sampled to the gap voltage signal in electrical spark working man-hour, and obtained data are sent to by single-chip microcomputer
Host computer, and control the motion of motion control card controlled motor to process gap to control.
Specific implementation step is：
1. single-chip microcomputer will be input to after the amplified isolation of the voltage signal in electrical spark working man-hour, AD is carried out in single-chip microcomputer and is adopted
Sample, voltage signal is obtained, and send the data to host computer and monitored in real time.
The precision of analog-to-digital conversion is 12, sample frequency 1MHz, for most of electrical discharge machining, this sampling
Speed accurately reacts gap state enough.
2. passing through data prediction to the voltage signal collected, remove noise spot, the volume of algorithm is carried out on MATLAB
Write and the data to collecting cluster, obtain 5 center particles, it is corresponding with 5 kinds of states in micro EDM gap.
This step needs the mass data of previous experiments, writes clustering algorithm program by MATLAB, finds out clearance features.
Its main thought is that each sample point is first regarded as a class, and what is then constantly repeated closes two wherein closest classes
And until reaching the number of final class, that is, five classes.Five center particles have thus been obtained as gap state to join
According to.
3. step 1. in data, every 1024 groups of sampled points, which are asked, is once averagely used as a sample point, calculates this sample
The distance of the center barycenter for 5 classes that the Current Voltage value of point obtains to cluster, state now is grouped into that of distance minimum
In individual class, in this, as the foundation of control strategy.
Be exactly specifically obtained sample point from step 2. in which center particle it is near, which class just belonged to, is also just belonged to
In processing gap state corresponding to that class.Corresponding different conditions take different control strategies.Due to dma module in single-chip microcomputer
Utilization, memory space can be read and writen in the case where not consuming cpu resource, ensure that the real-time of system
Property.
4. step 3. in obtain different status informations, send different instructions to motion control card, make motion control card
Make different control strategies.Using larger speed rollback strategy during for short circuit, play a part of small distance cutter lifting；Stable
During arc discharge can medium velocity retract, strongly avoid electrode calcination；Make transition electric by micro- cutter lifting in transient arc discharge
Arc discharge is adjusted to spark discharge, avoids the deterioration of processing.In normal spark discharge using the feeding of small speed, make as far as possible
Gap is maintained at spark discharge state.In open circuit using the feeding of big speed.
Claims (3)
1. a kind of electric spark clearance discharge condition detection apparatus based on cluster analysis, it is characterised in that including electric discharge machining apparatus
Bed, electrical signal detection collector, cluster analysis device, servo controller and host computer；The electrical signal detection collector includes putting
Big circuit and isolation optocoupler etc.；Cluster analysis device includes single-chip microcomputer and serial port module, and electric spark machine tool passes through cable connection
Amplifying circuit, isolation module are sequentially connected amplifying circuit, single-chip microcomputer, serial port module and host computer, the host computer connection motion
Control card, motion control card connection electric spark machine tool；The electric signal of the electric spark machine tool passes through electrical signal detection
Collector input carries out cluster analysis in cluster analysis device, and the cluster result output of cluster analysis device is terminated in servo controller
SERVO CONTROL is carried out, the relevant signal output of servo controller connects host computer, by being shown, being controlled and being deposited in host computer
Storage.
A kind of 2. electric spark clearance discharge condition detection apparatus based on cluster analysis as claimed in claim 1, it is characterised in that
The single-chip microcomputer uses STM32 single-chip microcomputers.
3. the electric spark clearance discharge condition detection method based on cluster analysis, it is characterised in that it is concretely comprised the following steps：
1) electrical discharge machining is carried out first, and the gap electric signal of processing is input in electrical signal detection collector and sampled, is obtained
Corresponding data, the electric signal are electric current, voltage；
2) the gap electrical signal data collected is input in cluster analysis device and carries out cluster analysis, while be sent to host computer
Make display and storage；
3) in cluster analysis device, the electrical signal data of each time interval is one group, by data prediction, removes noise
Point, each gap electrical signal data are a data point；
4) by clustering algorithm, each single data point as a class, the distance between each point will be calculated, distance closely
Point be classified as cluster；
5) clusters number is selected, carries out the calculating in step 4), obtains cluster to the end, the point of each distribution is become by cluster
The class of several aggregations, each class correspond to a kind of gap state, the machining state in each time interval are divided into customized several
Individual class, and obtained the center of class；
6) cluster result of step 5) is input in servo controller, makes SERVO CONTROL.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710823416.7A CN107398612A (en) | 2017-09-13 | 2017-09-13 | A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710823416.7A CN107398612A (en) | 2017-09-13 | 2017-09-13 | A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107398612A true CN107398612A (en) | 2017-11-28 |
Family
ID=60388243
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710823416.7A Pending CN107398612A (en) | 2017-09-13 | 2017-09-13 | A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107398612A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108213621A (en) * | 2017-12-13 | 2018-06-29 | 清华大学 | A kind of electrode gap control method for EDM Grinding |
| CN109202192A (en) * | 2018-10-31 | 2019-01-15 | 清华大学 | Fine electric spark servo scanning machined parameters preferred method and system |
| CN110328419A (en) * | 2019-07-22 | 2019-10-15 | 南京理工大学 | No resistive spark pulse power and its processing and clearance discharge condition recognition methods |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203759190U (en) * | 2014-03-20 | 2014-08-06 | 陕西公众智能科技有限公司 | Distribution network cable partial discharge on-line monitoring and discharge position positioning system |
| CN105522238A (en) * | 2014-10-28 | 2016-04-27 | 佛山科学技术学院 | Discharging gap state detection module based on pulse sequence analysis |
| CN106513879A (en) * | 2016-12-21 | 2017-03-22 | 北京仿真中心 | Spark discharge state recognition and detection method based on chaos theory |
| CN106735636A (en) * | 2017-03-31 | 2017-05-31 | 山东豪迈机械科技股份有限公司 | Electrical discharge machining discharging headlamp circuit and method |
| CN106808035A (en) * | 2016-12-02 | 2017-06-09 | 北京建筑大学 | Electrical discharge machining self-adaptation control method, apparatus and system |
-
2017
- 2017-09-13 CN CN201710823416.7A patent/CN107398612A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203759190U (en) * | 2014-03-20 | 2014-08-06 | 陕西公众智能科技有限公司 | Distribution network cable partial discharge on-line monitoring and discharge position positioning system |
| CN105522238A (en) * | 2014-10-28 | 2016-04-27 | 佛山科学技术学院 | Discharging gap state detection module based on pulse sequence analysis |
| CN106808035A (en) * | 2016-12-02 | 2017-06-09 | 北京建筑大学 | Electrical discharge machining self-adaptation control method, apparatus and system |
| CN106513879A (en) * | 2016-12-21 | 2017-03-22 | 北京仿真中心 | Spark discharge state recognition and detection method based on chaos theory |
| CN106735636A (en) * | 2017-03-31 | 2017-05-31 | 山东豪迈机械科技股份有限公司 | Electrical discharge machining discharging headlamp circuit and method |
Non-Patent Citations (1)
| Title |
|---|
| 张玲瑄: "高效微细电火花加工若干关键技术研究", 《中国知网》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108213621A (en) * | 2017-12-13 | 2018-06-29 | 清华大学 | A kind of electrode gap control method for EDM Grinding |
| CN108213621B (en) * | 2017-12-13 | 2019-04-02 | 清华大学 | A kind of electrode gap control method for EDM Grinding |
| CN109202192A (en) * | 2018-10-31 | 2019-01-15 | 清华大学 | Fine electric spark servo scanning machined parameters preferred method and system |
| CN109202192B (en) * | 2018-10-31 | 2019-09-17 | 清华大学 | Fine electric spark servo scanning machined parameters preferred method and system |
| CN110328419A (en) * | 2019-07-22 | 2019-10-15 | 南京理工大学 | No resistive spark pulse power and its processing and clearance discharge condition recognition methods |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI422460B (en) | Tool nose detection method for cutting machine tool | |
| CN107398612A (en) | A kind of electric spark clearance discharge condition detection apparatus and method based on cluster analysis | |
| CN111354366B (en) | Abnormal sound detection method and abnormal sound detection device | |
| CN110287552A (en) | Based on the motor bearings fault diagnosis method and system for improving random forests algorithm | |
| CN1683936A (en) | On-line detecting method and its device for multiple path transient wave form over voltage of power system | |
| CN108693448B (en) | Partial discharge mode recognition system applied to power equipment | |
| CN110561195B (en) | Method for monitoring flutter in machining process | |
| EP4024063B1 (en) | Fault-arc identification method, device and apparatus, and storage medium | |
| CN110434676A (en) | A kind of boring monitoring chatter method of multisensor time-frequency characteristics fusion | |
| CN104297213A (en) | Blood cell analyzing instrument and recognizing method and system for abnormal cells thereof | |
| CN216848010U (en) | Cable partial discharge online monitoring device for edge calculation | |
| CN112418065A (en) | Equipment operation state identification method, device, equipment and storage medium | |
| CN101915727B (en) | FPGA-based particle detection statistical system and method | |
| CN201083800Y (en) | Transformer substation insulated live-wire detector based on dummy instrument technology | |
| CN111331211B (en) | On-line penetration detection method and system for electric spark small hole machining | |
| CN112952830A (en) | Method for predicting transient stability of power system after fault | |
| CN109558041B (en) | GPU acceleration-based blade tip clearance signal acquisition, processing and transmission method | |
| CN112207631A (en) | Method for generating tool detection model, method, system, device and medium for detecting tool detection model | |
| CN114227382A (en) | Cutter damage monitoring system and method based on novel capsule network | |
| CN112781721B (en) | Intelligent noise acquisition method and intelligent noise acquisition device of mobile inspection platform | |
| CN113311858A (en) | Flight control system and method based on unmanned aerial vehicle | |
| CN104706366A (en) | Distraction testing method, device and system | |
| CN105987801B (en) | The abnormal vibrations detection device and its method of a kind of numerically-controlled machine tool | |
| CN116338454A (en) | New energy automobile electrical property detection method and device | |
| CN106021852A (en) | Density clustering algorithm based blood glucose data abnormal value computing method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171128 |
|
| RJ01 | Rejection of invention patent application after publication |