CN1354078A - Parameter setting method of ejection forming machine - Google Patents
Parameter setting method of ejection forming machine Download PDFInfo
- Publication number
- CN1354078A CN1354078A CN 00132669 CN00132669A CN1354078A CN 1354078 A CN1354078 A CN 1354078A CN 00132669 CN00132669 CN 00132669 CN 00132669 A CN00132669 A CN 00132669A CN 1354078 A CN1354078 A CN 1354078A
- Authority
- CN
- China
- Prior art keywords
- forming machine
- jet forming
- neural network
- parameter
- ejection
- 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
Images
Landscapes
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
The present invention utilizes experimental desgin method, and combines it with modular flow analysis software to simulate the process of actual ejection forming and analyze simulation result so as to create the data base of qualitative relationship of ejection forming machine parameter and ejection formed product quality parameter, and utilizes said data base to train neural network, and create said neural network to predict quality of the ejection product formed by ejection forming machine. When it is used, firstly, the parameter to be tested of ejection forming machine can be inputted into said neural network. After said parameter is executed by said neural network, the predicted ejection formed product qualitity parameter can be outputted so as to help the operator to set the ejection forming machine parameter.
Description
The method that the present invention sets about a kind of Jet forming machine parameter, especially about with method of experimental design in conjunction with mold flow analysis software, simulate the process of actual injection molding, and analysis mode result, to set up the data bank of qualitative relationships between Jet forming machine parameter and the ejection formed piece Q factor, and utilize this data bank to train the class neural network, to train and to set up such neural network, in order to predict the ejection formed piece quality of this Jet forming machine.
Traditionally, the operator of Jet forming machine is when Jet forming machine parameter assignment procedure, the decision of all parameters is to need to lean on the operator according to the shape of die cavity, the characteristic of plastics, the performance of machine and the factors such as countermeasure of moulding product defective, adjusts suitable molding parameter through permanent experience of accumulating.
More systematic ejaculator parameter setting method is that utilization field mouth method or method of experimental design obtain enough information and set up an empirical mode, and pattern is done the setting of ejaculator parameter according to this again.But its shortcoming is need expend considerable time and material resources in setting up the empirical mode process.Also can utilize a series of experiment, the relation between moulding product quality and the molding parameter is represented to unify the sequence pattern of having the records of distance by the log.Utilize this statistical relationship in forming process, to do comparison on the line, to adjust best molding parameter with the molding parameter signal of feedbacking.This quality control technology has developed quite completely, but shortcoming is need expend considerable time and material resources in setting up the statistical model process, and between the molding parameter of moulding product and the quality parameter, and can't summarize clear and definite qualitative relationships.
In addition, have with expert system suggestion engineering staff's parameter and set, its principle is the scheme that expertise is dealt with problems with the suggestion of " if first " rule.But expert system has it restricted, as between moulding product molding parameter and quality parameter, there is no clear and definite qualitative relationships, and it can't provide the information that exceeds outside its rule.
The known patent documentation whole world patent relevant with ejection formation roughly surpasses thousand pieces nearly ten year every year, and rises year by year, shows that ejection forming technique is just flourish.The whole world patent record of setting about the ejection formation parameter has 20 pieces.No. the 5518687th, the United States Patent (USP) more relevant with the present invention imported known molding parameter, and pressure, speed and screw position in the contrast ejaculation manufacture process, so as to revising the molding parameter of input.Its shortcoming is to be not easy to find out suitable setting value and corresponding manufacture process parameter relation thereof.No. the 5997778th, United States Patent (USP) in addition, and its method is known issuing velocity curve of input and the dynamic response of trying to achieve ejaculator, controls (PID) feedback correction setup parameter with the Sustainable Control ejection formation with PID.Its shortcoming is for only controlling issuing velocity.
Therefore, for Jet forming machine parameter setting means, existing needs on the improved industry with urgent, in order to save manpower and material resources, promotes moulding product quality.
The object of the invention for provide can be instant quality prediction, find out the value range of setting the Jet forming machine parameter, help the operator to adjust suitable molding parameter fast immediately, save the parameter setting-up time and find out suitable molding parameter rapidly, make the ejaculation manufacture process stable.
For reaching purpose of the present invention, the invention provides the method that a kind of Jet forming machine parameter is set, comprise the following steps: to use a method of experimental design in conjunction with a mold flow analysis software, to simulate the ejection formation process of actual this Jet forming machine, analyze this analog result and set up the Jet forming machine parameter and the ejection formed piece Q factor between the data bank of qualitative relationships; According to this data bank, set up a class neural network, it is in order to predict the ejection formed piece quality of this Jet forming machine; Import Jet forming machine to be measured to such neural network of having set up; The ejection formed piece Q factor of such neural network prediction of output.
Fig. 1 is a flow chart of the present invention.
Fig. 2 penetrates the basic function neural network for the width of cloth that the present invention uses.
Fig. 3 is the embodiment of input Jet forming machine parameter of the present invention.
Fig. 4 is the embodiment of output ejection formed piece Q factor of the present invention.
Be illustrated in figure 1 as flow chart of the present invention, at first doing simulation according to method of experimental design on mold flow analysis software penetrates, the present invention is on concrete enforcement, method of experimental design can utilize existing field mouth formula method of experimental design (Taguchi Parameter Design Method), the C-MOLD mold flow analysis software that mold flow analysis software can utilize existing U.S. Cornell University to be developed.Jet forming machine parameter through design, comply with above-mentioned field mouth formula method of experimental design in conjunction with C-MOLD mold flow analysis software, to simulate the ejection formation process of actual this Jet forming machine, analyze this analog result, set up the data bank of qualitative relationships between Jet forming machine parameter and the ejection formed piece Q factor according to this.Wherein above-mentioned simulation be with this mold flow analysis software be in the higher limit and lower limit of this Jet forming machine parameter (or being called the parameter window) carry out according to this method of experimental design, wherein the higher limit of this Jet forming machine parameter and lower limit are provided by this mold flow analysis software.Storing this analysis of data then learns for the class neural network, wherein class neural network study, be mainly according to this data bank, set up a class neural network, it is in order to predict the ejection formed piece quality of this Jet forming machine, and above-mentioned Jet forming machine parameter comprises cool time, dwell time, dwell pressure, issuing velocity at least, melts glue temperature, mold temperature.Above-mentioned ejection formed piece Q factor comprises that at least output weight, maximum volume shrink, average external volume is shunk, maximum caves in, average depression.The present invention is on concrete enforcement, and the class neural network can utilize the width of cloth to penetrate the basic function neural network, and penetrating the basic function neural network about the width of cloth will be in hereinafter explanation.
Class neural network prediction moulding product quality pattern in the flow process shown in Figure 1, and input Jet forming machine parameter is to the class neural network, be for importing Jet forming machine parameter to be measured to such neural network of having set up, the above-mentioned parameter value of importing to be measured is all in parameter window value range.Through the execution of class neural network of the present invention, the ejection formed piece Q factor of the last prediction of output.
Fig. 2 penetrates the basic function neural network for the width of cloth that the present invention uses.Input layer parameter X in Fig. 2
1X
2X
iFor above-mentioned cool time, dwell time, dwell pressure, issuing velocity, melt the Jet forming machine parameter of glue temperature, mold temperature etc.Output layer parameter O
1O
2O
kFor above-mentioned output weight, maximum volume shrink, average external volume is shunk, the ejection formed piece Q factor of maximum depression, average depression etc.Hide a plurality of class neuron F in the layer
1F
2F
HExcitation function R
1R
2R
HCan adopt Gaussian function.W
11And W
HkIt is the expression weighted value.
Fig. 3 is the embodiment of input Jet forming machine parameter of the present invention.Above-mentioned through training and set up such neural network, in concrete enforcement of the present invention, can be written as a software, make it on an existing computer, to carry out.Fig. 3 demonstration is carried out on a computer by the class neural network that software is implemented, and the operator sets the Jet forming machine parameter with input equipment at the parameter window.
Fig. 4 is the embodiment of output ejection formed piece Q factor of the present invention.The operator carries out on a computer through the class neural network that existing software is implemented, at the shown output ejection formed piece Q factor of computer screen with the input of Fig. 3 display mode.
Though the present invention with a preferred embodiment openly as above; yet it is not in order to limit the present invention; anyly be familiar with this operator; without departing from the spirit and scope of the present invention; also can do various changes and modification, so protection scope of the present invention should be as the criterion with the protection domain that accompanying Claim was defined.
Claims (5)
1. the method set of a Jet forming machine parameter comprises with the following step:
Use a method of experimental design in conjunction with a mold flow analysis software, to simulate the actual ejection formation process of this Jet forming machine, analyze this analog result and set up the Jet forming machine parameter and the ejection formed piece Q factor between the data bank of qualitative relationships;
According to this data bank, set up a class neural network, in order to predict the ejection formed piece quality of this Jet forming machine;
Import Jet forming machine parameter to be measured to such neural network of having set up;
The ejection formed piece Q factor of such neural network prediction of output.
2. the method for claim 1, wherein above-mentioned simulation is to be carried out according to this method of experimental design in the higher limit of this Jet forming machine parameter and lower limit by this mold flow analysis software, and wherein the higher limit of this Jet forming machine parameter and lower limit are provided by this mold flow analysis software.
3. the method for claim 1, wherein above-mentioned Jet forming machine parameter comprises cool time, dwell time, dwell pressure, issuing velocity at least, melts glue temperature, mold temperature.
4. the method for claim 1, wherein above-mentioned Jet forming machine parameter comprise at least that output weight, maximum volume shrink, average external volume is shunk, maximum depression, average depression.
5. the method for claim 1, wherein such neural network is that a width of cloth is penetrated the basic function neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00132669 CN1354078A (en) | 2000-11-20 | 2000-11-20 | Parameter setting method of ejection forming machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00132669 CN1354078A (en) | 2000-11-20 | 2000-11-20 | Parameter setting method of ejection forming machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1354078A true CN1354078A (en) | 2002-06-19 |
Family
ID=4595308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 00132669 Pending CN1354078A (en) | 2000-11-20 | 2000-11-20 | Parameter setting method of ejection forming machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1354078A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430720B (en) * | 2007-11-09 | 2011-02-09 | 邱维铭 | Analysis method used for in-mold decoration injection molding |
| CN103085231A (en) * | 2012-06-05 | 2013-05-08 | 桥弘数控科技（上海）有限公司 | Debugging method and device for injection molding machine |
| CN111497163A (en) * | 2018-12-28 | 2020-08-07 | 株式会社捷太格特 | Quality prediction system and molding machine |
-
2000
- 2000-11-20 CN CN 00132669 patent/CN1354078A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101430720B (en) * | 2007-11-09 | 2011-02-09 | 邱维铭 | Analysis method used for in-mold decoration injection molding |
| CN103085231A (en) * | 2012-06-05 | 2013-05-08 | 桥弘数控科技（上海）有限公司 | Debugging method and device for injection molding machine |
| CN103085231B (en) * | 2012-06-05 | 2015-03-18 | 桥弘数控科技（上海）有限公司 | Debugging method and device for injection molding machine |
| CN111497163A (en) * | 2018-12-28 | 2020-08-07 | 株式会社捷太格特 | Quality prediction system and molding machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107229789B (en) | Hybrid system modeling simulation platform and simulation method based on HSTPN model | |
| CN107123415A (en) | A kind of automatic music method and system | |
| CN110531722B (en) | Process parameter recommendation system and method based on data acquisition | |
| EP1376415A3 (en) | Method and apparatus for modeling injection of a fluid in a mold cavity | |
| CN108920147B (en) | Web page construction method and device, computing equipment and storage medium | |
| CN107516161A (en) | A kind of project progress management-control method based on BIM | |
| CN101934578A (en) | Method for unifying operation interfaces of various injection molding machines and injection molding system | |
| CN108766463B (en) | Electronic device, the music playing style recognition methods based on deep learning and storage medium | |
| CN101605246A (en) | The management system of plans of monitoring apparatus and method | |
| CN109948804A (en) | Cross-platform towed deep learning modeling and training method and device | |
| WO2004070491B1 (en) | Method and system for organizing and retrieving energy information | |
| CN109272800A (en) | A kind of Power Plant Simulation Training Methodology based on Unity3D | |
| CN1354078A (en) | Parameter setting method of ejection forming machine | |
| CN115292951A (en) | Intelligent planning design auxiliary system for urban and rural updating planning design | |
| CN109514774A (en) | A method of preventing air conditioning for automobiles grid backplate moulding buckling deformation | |
| CN111008706B (en) | Processing method for automatically labeling, training and predicting mass data | |
| CN106446338A (en) | Calculation quantity processing method and device based on Revit platform | |
| CN106295006A (en) | A kind of product design system | |
| EP2642359A1 (en) | Device for developing and method for creating a programm for an electronical control unit | |
| CN113128568B (en) | Excavator activity recognition method, system, device and storage medium | |
| CN1353039A (en) | Intelligent control method of ejection former | |
| CN113034422A (en) | Method and device for detecting yield of injection molding product and electronic equipment | |
| CN112365609B (en) | BIM technology-based hardware equipment operating software visual representation method | |
| CN117574695B (en) | Injection mold simulation pouring method, system and medium | |
| EP4190462A1 (en) | Method and evaluation component for monitoring a die casting or injection molding production process of a mechanical component with an according production machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |