CN105150482B - It is a kind of to cross cushion rubber failure on-line detecting method - Google Patents
It is a kind of to cross cushion rubber failure on-line detecting method Download PDFInfo
- Publication number
- CN105150482B CN105150482B CN201510511272.2A CN201510511272A CN105150482B CN 105150482 B CN105150482 B CN 105150482B CN 201510511272 A CN201510511272 A CN 201510511272A CN 105150482 B CN105150482 B CN 105150482B
- Authority
- CN
- China
- Prior art keywords
- cushion rubber
- sampling instant
- data
- pivot
- mentioned steps
- 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.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
- B29C45/17—Component parts, details or accessories; Auxiliary operations
- B29C45/76—Measuring, controlling or regulating
Abstract
Cushion rubber failure on-line detecting method is crossed the invention discloses a kind of, includes following steps：It gathers the colloidal sol pressure data of injection stage and is represented with two-dimensional matrix；Two-dimensional matrix standardization；PCA decomposition is carried out to standardization two-dimensional matrix；Calculate the trail change of the first pivot；Calculate the sampling instant corresponding to the maximum variation turning point of the trail change of the first pivot；Data before sampling instant in above-mentioned steps are integrated, and calculate average and variance；The colloidal sol pressure data of injection stage is obtained, and the data before sampling instant in above-mentioned steps are integrated, judged whether cushion rubber breaks down.The present invention establishes characteristic model by the pressure change to injection molding machine Collosol Motor hydraulic circuit entrance and monitors in real time, can detect failure rapidly, solves the problems, such as that cushion rubber operating status was sightless, avoids the loss in production.The present invention can be widely applied to injection molding machine field of fault detection as a kind of cushion rubber failure on-line detecting method of crossing.
Description
Technical field
It is especially a kind of to cross cushion rubber failure on-line detecting method the present invention relates to injection molding machine field of fault detection.
Background technology
With reference to Fig. 1, injection molding machine check valve structure component is made of three parts, is crossed rubber head, is crossed cushion rubber and washer.Cross cushion rubber again
It claims check ring, is flowed in itself and automatic open and close flap by medium, for preventing the valve of medium refluence.Cross the work of cushion rubber
With being exactly non-return, it is a part for preventing plastic melt from being leaked backward in injection.At work, cushion rubber is crossed with crossing glue to be situated between
Son (crossing rubber gasket) contact forms the structure of a closing, crosses cushion rubber and moves back and forth crossing between rubber head and washer and turn around axle center
It is dynamic, plastic melt is prevented to leak.During Forward, the plastics of melting state are spurted by cushion rubber in cavity, when moving afterwards, cross cushion rubber resistance
Block material stream enters, and closes cavity.
The cooperation surface damage between cushion rubber and washer is served as, cushion rubber mating surface substantially collapses excessively or washer face occurs
It bursts apart.It crosses cushion rubber and damages back-flow of molten material when can cause injection, injecting products weight weight is inconsistent, and product quality can not protect
Card.
Whether previous enterprise can only lean on periodic production inspection normal to test cushion rubber work, but the general hair of non-return failure
Life is when high-volume order produces, and check (non-return) valve can not be seen inside gun barrel, and worker can not generally have found in time, so
Once failure will result in the generation of waste product and the waste of raw material, enterprises production efficiency and economic benefit are very influenced.It is existing
There is generally use in technology separately to set sensor for detecting the operation conditions of machine, such as injection molding machine self-diagnosable system and method
(201210592344.7) need to calculate pumping transmission efficiency in technical solution, not only the realization of technical solution is very complicated, and
Cost of implementation compared with it is high, later maintenance difficulty is big.
The content of the invention
In order to solve the above-mentioned technical problem, the purpose of the present invention is：One kind is changed by monitoring hydraulic circuit inlet pressure
What feature was realized crosses cushion rubber failure on-line detecting method.
Injection molding is typical multistage process, mainly comprising injection, pressurize, plasticizing, cooling four-stage, ginseng
According to Fig. 2, in injection stage, the pushed at high pressure screw rod of hydraulic cylinder forward, crosses cushion rubber from rocket head end motion to washer end, it has one
It is a from the process for being opened to closure, within this of short duration time melting sizing material is pushed to spray nozzle front end, while Collosol Motor liquid
Pressure oil road inlet pressure (hereinafter referred to as colloidal sol pressure) be subject to melting sizing material movement because being influenced to have small variation.But work as
When not tight failure occurs being closed in valve, the time that opposition maintains is long, in the case of no relief valve, colloidal sol
The turning point time that pressure reaches maximum and remains unchanged is many compared to that can lag under normal circumstances, therefore the method for the present invention master
It to be realized by the correlation computations to turning point.
The technical solution adopted in the present invention is：It is a kind of to cross cushion rubber failure on-line detecting method, include following steps：
A, the colloidal sol pressure data of multiple batch injection stages is gathered in injection moulding process, and by the data collected with two
Matrix is tieed up to represent；
B, above-mentioned two-dimensional matrix is standardized；
C, PCA decomposition is carried out to the two-dimensional matrix after standardization, obtains score matrix and load matrix；
D, trail change of first pivot in time orientation is calculated；
E, the sampling instant corresponding to the maximum variation turning point of the trail change of the first pivot is calculated；
F, the data before sampling instant in above-mentioned steps are integrated, and calculates average and variance；
G, the colloidal sol pressure data of injection stage is obtained, and the data before sampling instant in above-mentioned steps are integrated,
Judged whether cushion rubber breaks down using Shewhart control chart method.
Further, the standardization in the step B is：Each variable in two-dimensional matrix is subtracted into its column
Average, then divided by the variance of its column.
Further, the differential that the trail change for calculating the first pivot is first passed through in the step E obtains maximum variation turnover
Point obtains corresponding sampling instant further according to maximum variation turning point.
Further, in the step G, the value that the data in above-mentioned steps before sampling instant are integrated exceeds
Statisti-cal control scope then judged that cushion rubber broke down；The lower limit of the Statisti-cal control scope is equal to be calculated in step F
Value subtracts 3 times of variances, and the upper limit of the Statisti-cal control scope adds 3 times of variances for the average being calculated in step F.
Further, in the step G, point for the value that the data in above-mentioned steps before sampling instant are integrated
Cloth is apparent Non-Gaussian Distribution, then judges that cushion rubber broke down.
The beneficial effects of the invention are as follows：The present invention is built by the pressure change to injection molding machine Collosol Motor hydraulic circuit entrance
Vertical characteristic model, monitors the pressure of hydraulic circuit entrance in real time, has served as to quickly detect when cushion rubber breaks down, solve
It has determined and has crossed the sightless problem of cushion rubber operating status, avoided the loss in production.
Description of the drawings
Fig. 1 is injection molding machine check valve structure sectional view；
Fig. 2 is check (non-return) valve operation principle schematic diagram；
Fig. 3 is the step flow chart of the method for the present invention；
Fig. 4 is the sampled point schematic diagram in the specific embodiment of the invention；
Fig. 5 is the Shewhart control chart that the present invention uses.
In Fig. 1：1st, rubber head is crossed；2nd, cushion rubber is crossed；3rd, washer；
In Fig. 2：21st, cushion rubber is crossed；22nd, washer；23rd, machine barrel；24th, high molecular material is melted.
Specific embodiment
The specific embodiment of the present invention is described further below in conjunction with the accompanying drawings：
It is a kind of to cross cushion rubber failure on-line detecting method with reference to Fig. 3, include following steps：
A, the colloidal sol pressure data of multiple batch injection stages is gathered in injection moulding process, and by the data collected with two
Matrix is tieed up to represent；
Each batch of injection phase at least K is using point, and using K as standard, such I duplication of production operates, data
It can be expressed as a two-dimensional matrix
B, above-mentioned two-dimensional matrix is standardized；
The standardization is：Each variable in two-dimensional matrix is subtracted to the average of its column, then divided by its
The variance of column.For two-dimensional matrixThe variable at any pointThe calculation formula of its standardization is such as
Under：
Wherein：It isThe average of matrix either rank, skIt isThe variance of matrix either rank,With skFormula it is as follows：
C, PCA decomposition is carried out to the two-dimensional matrix after standardization, obtains score matrix and load matrix；
PCA (Principal component are performed to two-dimensional matrix X (I*K) of the previous step after normalized
Analysis) decompose, wherein PCA decomposable processes are as follows：
S=trace (TTT/(I-1)) (5)
Wherein：tiFor orthogonal principal component vector, piFor the load vector of orthonomalization, S is the covariance matrix of pivot
Mark represents explanation degree size of each pivot for process.
Formula (4) X decomposes to obtain score matrix T (I × K) and load matrix P (K × K).
PCA is the modeling method for handling multivariable earliest, although the present invention only make use of this change of colloidal sol pressure
Amount, but pass through the matrix arrangement method of step A, the K an original variable sample being converted into one batch
A sampling.
D, trail change of first pivot in time orientation is calculated；
It is each in the load matrix P (K*K) that pivot is decomposed to arrange the weight for corresponding to a pivot, if the first pivot pair
The load vector answered is p1, p1It is the first row of load matrix P (K*K), i.e. p1=P (K × 1).
The contribution e_t (k) at the first pivot each moment=X (k) ' X (k) p1(k)·p1(k)'。
First pivot is in contribution E_T (K*1) calculation formula of time orientation：
E_T=X'*X*P1*P1' (6)
E_T i.e. the first pivot is in the trail change of time orientation.
E, the sampling instant corresponding to the maximum variation turning point of the trail change of the first pivot is calculated；
The differential for specially first passing through the trail change for calculating the first pivot obtains maximum variation turning point, further according to maximum
Variation turning point obtains corresponding sampling instant.
Differentiate to E_T diff (E_T), seeks the difference between two neighboring sampled point：
Diff (E_T)=E_T (k+1)-E_T (k) k=1,2......K-1 (7)
If udAnd kdThe corresponding maximum turning point of maximum variation turning point and sampling instant；
ud=max (abs (diff (E_T))) (8)
U at this timedCorresponding sampled point is kmax, then
kd=kmax+ 11 ＜ kmax＜ K (9)
F, the data before sampling instant in above-mentioned steps are integrated, and calculates average and variance；
To kdBefore pointData integrate and seek the average and variance of model data；
With reference to formula (2) and formula (3), to xis(I*1) averageAnd variance
G, the colloidal sol pressure data of injection stage is obtained, and the data before sampling instant in above-mentioned steps are integrated,
Judged whether cushion rubber breaks down using Shewhart control chart method.
Injection stage colloidal sol pressure data x is obtained firstk, and with reference to figure formula (10) to kdData before sampled point carry out
Integration obtainsWhen due to failure, crossing cushion rubber can not normally play a role, and cause the melting sizing material adverse current before rocket head,
In the case of no relief valve, reaching maximum pressure, that is, melting charge and sealing the time that can not flow back can postpone very compared to normal condition
It is more, therefore Shewhart control chart method is utilized, whenDo not existThis Statisti-cal control scope orDistribution it is apparent
Non-Gaussian Distribution, then it is assumed that cross cushion rubber and break down.
With reference to Fig. 4, as the specific embodiment of the method for the present invention, in shake hero MJ55 injection production processes, gathered with high density
Ethylene (High Density Polyethylene, HDPE) is used as raw material, three sections of temperature of barrel be respectively set as (210,
210,160), injection speed 35mm/s, the pressure of packing stage are set as 30 Pascals, and the time of packing stage is 5s, cold
But the time in stage be 15s, sampling period 50ms, historical dataModeling has selected I=29 normal operation batches, often
A batch injection stage remains K=31 colloidal sol force samples point.
The corresponding turning point k of diff (E_T) maximum is calculated according to above-mentioned stepsd=14, so to turning point before
Model data integrates, and the pressure integral mean value of 31 batches is asked to be obtained with varianceOnline
During monitoring, when each injection stage is completed, the sum of 14 sampled points before melten gel pressure is calculated, is measured(such as Fig. 5
It is shown), much smaller than Xiu Hate statistical valuesTherefore it is judged as non-return defective valve.
The above are implementing to be illustrated to the preferable of the present invention, but the invention is not limited to the implementation
Example, those skilled in the art can also make a variety of equivalents on the premise of without prejudice to spirit of the invention or replace
It changes, these equivalent deformations or replacement are all contained in the application claim limited range.
Claims (2)
1. a kind of cross cushion rubber failure on-line detecting method, it is characterised in that：Include following steps：
A, the colloidal sol pressure data of multiple batch injection stages, and the data Two-Dimensional Moment that will be collected are gathered in injection moulding process
Matrix representation；
B, above-mentioned two-dimensional matrix is standardized, each variable in two-dimensional matrix is subtracted to the average of its column,
Divided by the variance of its column then；
C, PCA decomposition is carried out to the two-dimensional matrix after standardization, obtains score matrix and load matrix；
D, trail change of first pivot in time orientation is calculated；
E, the sampling instant corresponding to the maximum variation turning point of the trail change of the first pivot is calculated；Specially first pass through calculating
The differential of the trail change of first pivot obtains maximum variation turning point, and corresponding sampling is obtained further according to maximum variation turning point
Moment；
F, the data before sampling instant in above-mentioned steps are integrated, and calculates average and variance；
G, the colloidal sol pressure data of injection stage is obtained, and the data before sampling instant in above-mentioned steps are integrated, is utilized
Shewhart control chart method judged whether cushion rubber breaks down；Data in above-mentioned steps before sampling instant integrate
To value beyond Statisti-cal control scope then judged cushion rubber break down；The lower limit of the Statisti-cal control scope is fallen into a trap for step F
Obtained average subtracts 3 times of variances, and the upper limit of the Statisti-cal control scope adds 3 times for the average being calculated in step F
Variance.
2. a kind of cushion rubber failure on-line detecting method excessively according to claim 1, it is characterised in that：In the step G, when
The value integrated to the data before sampling instant in above-mentioned steps is distributed as apparent Non-Gaussian Distribution, then judges glue
Circle breaks down.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510511272.2A CN105150482B (en) | 2015-08-19 | 2015-08-19 | It is a kind of to cross cushion rubber failure on-line detecting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510511272.2A CN105150482B (en) | 2015-08-19 | 2015-08-19 | It is a kind of to cross cushion rubber failure on-line detecting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105150482A CN105150482A (en) | 2015-12-16 |
| CN105150482B true CN105150482B (en) | 2018-05-18 |
Family
ID=54791669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510511272.2A Active CN105150482B (en) | 2015-08-19 | 2015-08-19 | It is a kind of to cross cushion rubber failure on-line detecting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105150482B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109932908B (en) * | 2019-03-20 | 2022-03-01 | 杭州电子科技大学 | Multi-directional principal component analysis process monitoring method based on alarm reliability fusion |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102313578A (en) * | 2011-08-04 | 2012-01-11 | 广州市香港科大霍英东研究院 | Mechanical seal online monitoring system |
| CN102431136B (en) * | 2011-09-16 | 2014-11-05 | 广州市香港科大霍英东研究院 | Multi-phase batch process phase dividing method based on multiway principal component analysis method |
| CN102495961B (en) * | 2011-12-05 | 2014-08-27 | 山东电力研究院 | Simplified statistical control method of measurement process |
-
2015
- 2015-08-19 CN CN201510511272.2A patent/CN105150482B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105150482A (en) | 2015-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8060240B2 (en) | Injection molding control method | |
| EP1091842B1 (en) | Automated molding technology for thermoplastic injection molding | |
| CN107831662B (en) | Design method of random 2D controller for intermittent process with actuator fault | |
| CN105150482B (en) | It is a kind of to cross cushion rubber failure on-line detecting method | |
| KR102538009B1 (en) | Method for determining the actual volume of an injection-mouldable compound in an injection-molding process | |
| CN105909595B (en) | A kind of method for judging hydraulic station oil leak and pressure oscillation | |
| CN103901773B (en) | Method for designing 2D hybrid controller according to input delay | |
| CN103640194A (en) | Intelligent injection molding machine and injection molding method thereof | |
| CN114770891B (en) | Injection molding machine abnormity analysis method and system | |
| KR20160132773A (en) | Method for determining a value for the description of the compression of a mouldable material | |
| CN113843993A (en) | Production control method, device and equipment of intelligent mold and storage medium | |
| CN111932080A (en) | Early warning protection system and method applied to water service pipe network | |
| CN112507487A (en) | Reliability evaluation method and system for hydraulic turbine speed governor servomotor | |
| DE3608973C2 (en) | Method and device for controlling the switchover from the holding pressure phase to the isochoric cooling phase during injection molding of thermoplastics | |
| US6309571B2 (en) | Method and apparatus for the control of injection molding | |
| Wang et al. | A fast monitoring method for multiple operating batch processes with incomplete modeling data types | |
| CN112879388A (en) | Abnormal leakage detection method for hydraulic system | |
| CN109932908B (en) | Multi-directional principal component analysis process monitoring method based on alarm reliability fusion | |
| CN106270437A (en) | The Auto-Test System of a kind of Metal in Tundish flow control device and method | |
| US20010019183A1 (en) | Injection molding apparatus and a method therefor | |
| CN107562007A (en) | Control method and system based on fan-shaped section position sensor malfunction | |
| JPH08281756A (en) | Operation control method for injection molding machine | |
| Meng et al. | A survey of machine vision-based monitoring methods for abnormalities in molds | |
| CN113533135B (en) | Flow control method based on non-Newtonian fluid rheological property test system | |
| CN115302727A (en) | Abnormity alarm analysis method in injection molding process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |