CN105095273A - Fuzzy BP neural network based glass tempering process parameter setting method - Google Patents
Fuzzy BP neural network based glass tempering process parameter setting method Download PDFInfo
- Publication number
- CN105095273A CN105095273A CN201410198246.4A CN201410198246A CN105095273A CN 105095273 A CN105095273 A CN 105095273A CN 201410198246 A CN201410198246 A CN 201410198246A CN 105095273 A CN105095273 A CN 105095273A
- Authority
- CN
- China
- Prior art keywords
- glass
- neural network
- tempering
- fuzzy
- technological parameter
- 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
-
- 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
- Y02P40/00—Technologies relating to the processing of minerals
- Y02P40/50—Glass production, e.g. reusing waste heat during processing or shaping
- Y02P40/57—Improving the yield, e-g- reduction of reject rates
Abstract
The invention discloses a fuzzy BP neural network based glass tempering process parameter setting method. The process parameter setting method comprises: according to an initial sample, training a BP neural network; dividing existing tempered glass into multiple categories; obtaining the optimal process parameters of each category according to the already trained BP neural network; establishing a process database according to the process parameters of all the categories; directly judging the category of to-be-tempered glass; directly choosing the process parameters corresponding to the category from the technology database; and setting the tempering process parameters of the to-be-tempered glass by utilizing the chosen process parameters. According to the fuzzy BP neural network based glass tempering process parameter setting method, the BP neural network is used to directly obtain the optimal process parameters of the to-be-tempered glass in the tempering process, so that the degree of dependence on artificial experience is reduced and the production cost is reduced; the process parameter setting method is realized through a computer, so that the setting precision is high, the error rate is low, the quality of the tempered glass can be improved, and the production efficiency is greatly improved.
Description
Technical field
The present invention relates to glass tempering technology field, be specifically related to a kind of technological parameter method to set up of the glass tempering based on fuzzy BP neural network.
Background technology
Along with the development of economic globalization, China proposes the requirement of dressing to advanced international standard to industry and the throughput requirements of civilian glassware and quality standard.Tempered glass is as safety glass, processed through tempering by former sheet glass and obtain, than the impact strength of simple glass exceed four times and more than, when running into extraneous factor and being broken, can not be formed sharp " knife face ", but become and be close to uniform fine granularity glass disintegrating slag, there is the advantage of not easily hurting sb.'s feelings, therefore in the safety glass of Global Auto industry, building industry, instrument and furniture industry uses, tempered glass occupation rate is more than 60%, and in continuous expansion, tempered glass market outlook are had an optimistic view of.Traditional glass tempering apparatus energy consumption is high, and glass self-explosion rate is high, and the tempering cycle is long, yields poorly.Along with the rising of tempered glass demand and the progress of science and technology, while the performance of glass fibre reinforced plastic equipment constantly promotes, how to overcome that the cost that the domestic traditional type tempered glass mode of production has is high, the recruitment of highly dependence technology, Operating Complexity is high, yield rate is low, the not retrospective shortcoming of quality, become tempered glass towards the future development of low energy consumption, high finished product rate, high yield further and produce the emphasis studied in chain.
Many developed countries are having relatively advanced yield-power and the relations of production decades ago, in order to pursue the high-quality of glass post-processing product and yield rate to obtain the larger surplus value, they focus on the links of production very much, while improving constantly the rigid technique of glass tempering furnace, also add the utility appliance having influence power, as electronic board etc.Technological parameter when glass tempering is directly set by electronic board, annealing furnace is controlled by electronic board, and this electronic board also detects the temperature of annealing furnace, pressure and other parameters showing in real time in real time, thus realize time detecting and the Automated condtrol of glass tempering process.
The technological parameter of tempering directly determines the quality of tempered glass, to its decisive role of quality of tempered glass.If technological parameter arranges improper, glass can be caused to occur crackle, the problems such as surface irregularity, affect product quality.In prior art, the technological parameter of glass tempering is many is rule of thumb arranged by technician, substantially increases production cost, and requires technician high, can not meet the needs of modern production scale.
Summary of the invention
For the deficiencies in the prior art, the present invention proposes a kind of technological parameter method to set up of the glass tempering based on fuzzy BP neural network.
(1) obtain several initial samples, each initial sample comprises a batch subvitreous glass kind, thickness of glass, the glass blocks number of a heat and the glass total area, and corresponding technological parameter;
(2) for any one initial sample, by the glass kind in this initial sample with thickness of glass is fuzzy turns to eigenwert, and using the sample after obfuscation as training sample;
(3) using the glass blocks number of the eigenwert of each training sample, a heat and the glass total area as the input of BP neural network, using technological parameter as output, training BP neural network obtains the BP neural network trained;
(4) according to the glass blocks number of the glass kind of tempered glass, thickness of glass, a heat and the glass total area, tempered glass is divided into several classifications, and determines each classification characteristic of correspondence sample;
(5) the BP neural network that the input of this feature samples trains is obtained the optimal procedure parameters of each classification, build according to optimal procedure parameters of all categories and obtain technological data bank;
(6) obtain the size and shape treating the glass kind of tempered glass, thickness of glass and glass, and treat tempered glass according to the size of upper slice platform of glass tempering furnace and carry out pre-scheduling, obtain glass blocks number and the glass total area of a heat;
(7) classification treated belonging to tempered glass is determined according to the glass blocks number and the glass total area for the treatment of the glass kind of tempered glass, thickness of glass, a heat, classification belonging to it selects corresponding optimal procedure parameters from technological data bank, and according to the technological parameter of optimal procedure parameters setting when the tempering of tempered glass selected.
Initial sample in the present invention extracts from the historical data that tempering different glass produces and obtains, according to initial sample, BP neural network is trained, and existing tempered glass is divided into some classifications, the BP neural network trained is utilized to obtain the optimal procedure parameters of each classification, and build technological data bank according to the technological parameter of all categories, the kind (namely belonging to which classification) of tempered glass is treated in direct judgement, directly choose technological parameter corresponding to this classification from technological data bank, utilize the processing parameter setting selected to treat the technological parameter of tempered glass tempering.
Because the glass kind of glass can not quantize usually, in order to can direct input neural network, BP neural network in the present invention is fuzzy BP neural network, fuzzy BP neural network uses one the most widely in fuzzy neural network, it is the combination of fuzzy system and BP neural network, be made up of neuron fuzzy in a large number, have independently network structure and algorithm, realize stronger non-linear mapping capability by the correlation parameter in adjustment fuzzy neural meta-model and network weight.The topological structure of existing fuzzy BP neural network is divided into five layers, is information input layer, fuzzy quantification input layer, BP network hidden layer (fuzzy reasoning layer), fuzzy quantification output layer and decision information output layer respectively.Application fuzzy BP neural network can reduce the requirement to input amendment, the glass kind after obfuscation directly can be inputted.Because thickness of glass is relevant to glass kind, therefore obfuscation is carried out to glass kind and thickness simultaneously.
According to the glass blocks number of glass kind, thickness of glass, a heat and the glass total area, existing tempered glass is divided into several classifications in step (4), concrete principle of classification is as follows:
Three classes can be divided according to glass kind, be respectively: white glass, gray glass and coloured glass (LOW-E glass) three classes, 3mm, 5mm and 6mm tri-kinds can be divided into according to thickness of glass, according to the glass blocks number of a heat and transparency area, tempered glass is divided into 2 classes, is respectively: block number is many, the total area large and block number is few, little two classes of the total area.Think in the present invention that the glass blocks number of a heat is glass blocks number many (not comprising 10) when being 10 ~ 100 pieces, 1 ~ 10 piece is glass blocks number few (comprising 10), with the glass total area of a heat for 7 ~ 10m
2time be that the total area (does not comprise 7m greatly
2), the glass total area of a heat is 5 ~ 7m
2time be that the total area is little.According to permutation and combination principle, existing tempered glass can be divided into 36 classifications altogether according to the glass blocks number of glass kind, thickness of glass, a heat and the glass total area.
In the present invention, transparency area being turned in a large number transparency area is 8.5m
2, transparency area being turned in a large number transparency area is 6m
2.It is 40 that glass blocks number volume is turned to glass blocks number, and glass blocks number being turned on a small quantity glass blocks number is 10.At the feature samples by obfuscation being obtained each class after quantification.And then obtain all kinds of optimal procedure parameters.
In the present invention, if described step (7) if in treat that tempered glass does not belong to any classification, then illustrate that this treats that tempered glass is not conventional tempering glass, need to carry out Process Exploration in addition.
Described technological parameter comprises interval time, preheating time, heat time, enhanced time and cool time.
Technological parameter during glass tempering is a lot, normal adjusting process parameter comprises strengthening blower fan strengthening blast, strengthening blower fan cooling wind pressure, cooling blower cooling wind pressure, flat windward, flat leeward, top heating target temperature, bottom heating target temperature, interval time, preheating time, heat time, enhanced time, cool time etc., but topmost or to adjusting frequency the highest interval time, preheating time, heat time, enhanced time, cool time.Preferably these 5 times, conventional need of production can be met, and can training effectiveness be ensured.
Glass kind and thickness of glass is fuzzy turns to an eigenwert in described step (2).
Utilize fuzzyTECH software by glass kind in described step (2) and thickness of glass is fuzzy turns to eigenwert.
By glass kind with thickness of glass is fuzzy turns to same value in the present invention, can related coefficient be improved, and then improve fuzzy precision.Due to the numerical value that glass kind is not concrete, and the thermal conductivity coefficient that different glass kinds is corresponding different, therefore usually adopt corresponding thermal conductivity coefficient to represent corresponding glass kind.Therefore time fuzzy, the glass kind of input is actual is thermal conductivity coefficient corresponding to this kind.It is 1,0.1,0.01 that the thermal conductivity coefficient of usual white glass, gray glass and coloured glass is divided into.
As preferably, the hidden layer number of described BP neural network is 5, and learning rate is 0.1, and error range is-3 ~+3, and maximum cycle is 100000, and input number of nodes is 3, and output node number is 5.
Be directly connected to the accuracy of the BP neural network after final training by specification error scope and hidden layer number, thus improve the degree of accuracy of the optimal procedure parameters obtained, for treating that tempered glass tempering provides better technique to instruct.Owing to being an eigenwert by glass kind and thickness mode gelatinization, therefore in fact input number of nodes is 4.In practical application, input number of nodes regulates according to applicable cases.
Described step (7) is using the optimal procedure parameters selected as the technological parameter when the tempered glass tempering.
As preferably, in described step (7), the corresponding optimal procedure parameters of selection is manually revised, and using artificial revised optimal procedure parameters as the technological parameter when the tempered glass tempering.
In actual toughening process, owing to being subject to the impact of tempering environment (comprising the ruuning situation etc. of physical environment, annealing furnace and other equipment), by revising optimal procedure parameters, improve the quality of glass.This correction needs to carry out according to practical experience usually.
Also comprise the glass quality after according to tempering in described step (5) to optimal procedure parameters of all categories scoring, and this scoring is recorded in technological data bank.
Detect using revised optimal procedure parameters as the technological parameter when the tempered glass tempering, quality according to glass after tempering is marked to revised optimal procedure parameters, if this scoring is greater than the scoring of corresponding optimal procedure parameters in technological data bank, then replace corresponding optimal procedure parameters in technological data bank with revised optimal procedure parameters.
The scoring of the technological parameter in the present invention is actually the quality score of the tempered glass adopting this technological parameter tempering.By constantly updating technological data bank in the present invention, can glass tempering process be improved constantly, improving glass quality.
According to historical data in the technological parameter method to set up of the glass tempering based on fuzzy BP neural network in the present invention, utilize BP neural network, the technological parameter of direct acquisition when tempered glass tempering, reduce the degree of dependence to artificial experience, reduce production cost, and realized by computing machine, it is high that precision is set, error rate is low, is conducive to the quality improving tempered glass, substantially increases production efficiency.
Embodiment
Below in conjunction with specific embodiment, the present invention is described in detail.
Based on a technological parameter method to set up for the glass tempering of fuzzy BP neural network, comprising:
(1) obtain several initial samples, each initial sample comprises a batch subvitreous glass kind, thickness of glass, the glass blocks number of a heat and the glass total area, and corresponding technological parameter.
In the present embodiment, initial sample extracts and obtains from a large amount of historical data.In the present embodiment, the number of initial sample is 180.Historical data is with the input of txt form, and for guaranteeing to be transfused to and to extract initial sample, this historical data must meet the restriction of specific form.In the present embodiment, all historical datas are inputted with txt form, then directly extract initial sample from the historical data of this txt form by the corresponding code that extracts.
Technological parameter in initial sample comprises interval time, preheating time, heat time, enhanced time and cool time.
(2) for any one initial sample, by the glass kind in this initial sample with thickness of glass is fuzzy turns to eigenwert, and using the sample after obfuscation as training sample.
Utilize fuzzyTECH software by glass kind in the present embodiment and thickness of glass is fuzzy turns to an eigenwert.
(3) using the glass blocks number of the eigenwert of each training sample, a heat and the glass total area as the input of BP neural network, using technological parameter as output, training BP neural network obtains the BP neural network trained.
BP neural network in the present embodiment is fuzzy BP neural network, and its hidden layer number is 5, and learning rate is 0.1, and error range is-3 ~+3, and maximum cycle is 100000, and input number of nodes is 3, and output node number is 5.
(4) according to the glass blocks number of the glass kind of tempered glass, thickness of glass, a heat and the glass total area, tempered glass is divided into several classifications, and determines each classification characteristic of correspondence sample.
In the present embodiment, tempered glass is divided into 36 classes.
(5) the BP neural network that the input of this feature samples trains is obtained the optimal procedure parameters of each classification, build according to optimal procedure parameters of all categories and obtain technological data bank.
(6) obtain the size and shape treating the glass kind of tempered glass, thickness of glass and glass, and treat tempered glass according to the size of upper slice platform of glass tempering furnace and carry out pre-scheduling, obtain glass blocks number and the glass total area of a heat.
Treat the glass kind of tempered glass, thickness of glass and glass size and shape can according to contain have above information order in extract.This order is with the input of xls or txt form, and for guaranteeing input, this order must have corresponding form to limit.In the present embodiment, all historical datas are inputted with xls form, then extract by corresponding the size and shape that code directly extracts glass kind, thickness of glass and the glass for the treatment of tempered glass from the order of this xls form.
For enhancing productivity, during pre-scheduling, needing the size considering interval, upper slice platform, ensureing the glass total area large (namely the occupation rate of glass to upper slice platform is high) as far as possible, and corresponding interval must be left between adjacent glass.
(7) classification treated belonging to tempered glass is determined according to the glass blocks number and the glass total area for the treatment of the glass kind of tempered glass, thickness of glass, a heat, classification belonging to it selects corresponding optimal procedure parameters from technological data bank, and according to the technological parameter of optimal procedure parameters setting when the tempering of tempered glass selected.
Wherein, step (7) can directly using the optimal procedure parameters selected as the technological parameter when the tempered glass tempering.Also manually can revise the corresponding optimal procedure parameters of selection, and using artificial revised optimal procedure parameters as the technological parameter when the tempered glass tempering.
Using artificial revised optimal procedure parameters as the technological parameter when the tempered glass tempering in the present embodiment.
In the present embodiment, also comprise the glass quality after according to tempering in step (5) to optimal procedure parameters of all categories scoring, and this scoring is recorded in technological data bank.
Detect using revised optimal procedure parameters as the technological parameter when the tempered glass tempering, quality according to glass after tempering is marked to revised optimal procedure parameters, if this scoring is greater than the scoring of corresponding optimal procedure parameters in technological data bank, then replace corresponding optimal procedure parameters in technological data bank with revised optimal procedure parameters.
The foregoing is only the preferred embodiment of the present invention, protection scope of the present invention is not limited in above-mentioned embodiment, and every technical scheme belonging to the principle of the invention all belongs to protection scope of the present invention.For a person skilled in the art, some improvements and modifications of carrying out under the prerequisite not departing from principle of the present invention, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (9)
1., based on a technological parameter method to set up for the glass tempering of fuzzy BP neural network, it is characterized in that, comprising:
(1) obtain several initial samples, each initial sample comprises a batch subvitreous glass kind, thickness of glass, the glass blocks number of a heat and the glass total area, and corresponding technological parameter;
(2) for any one initial sample, by the glass kind in this initial sample with thickness of glass is fuzzy turns to eigenwert, and using the sample after obfuscation as training sample;
(3) using the glass blocks number of the eigenwert of each training sample, a heat and the glass total area as the input of BP neural network, using technological parameter as output, training BP neural network obtains the BP neural network trained;
(4) according to the glass blocks number of the glass kind of tempered glass, thickness of glass, a heat and the glass total area, tempered glass is divided into several classifications, and determines each classification characteristic of correspondence sample;
(5) the BP neural network that the input of this feature samples trains is obtained the optimal procedure parameters of each classification, build according to optimal procedure parameters of all categories and obtain technological data bank;
(6) obtain the size and shape treating the glass kind of tempered glass, thickness of glass and glass, and treat tempered glass according to the size of upper slice platform of glass tempering furnace and carry out pre-scheduling, obtain glass blocks number and the glass total area of a heat;
(7) classification treated belonging to tempered glass is determined according to the glass blocks number and the glass total area for the treatment of the glass kind of tempered glass, thickness of glass, a heat, classification belonging to it selects corresponding optimal procedure parameters from technological data bank, and according to the technological parameter of optimal procedure parameters setting when the tempering of tempered glass selected.
2., as claimed in claim 1 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, described technological parameter comprises interval time, preheating time, heat time, enhanced time and cool time.
3. as claimed in claim 2 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, glass kind and thickness of glass is fuzzy turns to an eigenwert in described step (2).
4. as claimed in claim 3 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, utilize fuzzyTECH software by glass kind in described step (2) and thickness of glass is fuzzy turns to eigenwert.
5. the technological parameter method to set up of the glass tempering based on fuzzy BP neural network as described in Claims 1 to 4, it is characterized in that, the hidden layer number of described BP neural network is 5, learning rate is 0.1, error range is-3 ~+3, maximum cycle is 100000, and input number of nodes is 3, and output node number is 5.
6. as claimed in claim 5 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, described step (7) is using the optimal procedure parameters selected as the technological parameter when the tempered glass tempering.
7. as claimed in claim 5 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, in described step (7), the corresponding optimal procedure parameters of selection is manually revised, and using artificial revised optimal procedure parameters as the technological parameter when the tempered glass tempering.
8. the technological parameter method to set up of the glass tempering based on fuzzy BP neural network as claimed in claims 6 or 7, it is characterized in that, also comprise the glass quality after according to tempering in described step (5) to optimal procedure parameters of all categories scoring, and this scoring is recorded in technological data bank.
9. as claimed in claim 8 based on the technological parameter method to set up of the glass tempering of fuzzy BP neural network, it is characterized in that, detect using revised optimal procedure parameters as the technological parameter when the tempered glass tempering, quality according to glass after tempering is marked to revised optimal procedure parameters, if this scoring is greater than the scoring of corresponding optimal procedure parameters in technological data bank, then replace corresponding optimal procedure parameters in technological data bank with revised optimal procedure parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410198246.4A CN105095273B (en) | 2014-05-12 | 2014-05-12 | A kind of technological parameter setting method of the glass tempering based on fuzzy BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410198246.4A CN105095273B (en) | 2014-05-12 | 2014-05-12 | A kind of technological parameter setting method of the glass tempering based on fuzzy BP neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105095273A true CN105095273A (en) | 2015-11-25 |
| CN105095273B CN105095273B (en) | 2018-07-06 |
Family
ID=54575727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410198246.4A Active CN105095273B (en) | 2014-05-12 | 2014-05-12 | A kind of technological parameter setting method of the glass tempering based on fuzzy BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105095273B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528637A (en) * | 2015-11-26 | 2016-04-27 | 江南大学 | Diagnosis method based on linear interpolation type fuzzy neural network |
| CN110648727A (en) * | 2019-10-30 | 2020-01-03 | 华南理工大学 | Preparation method of glass material with specific physical properties |
| CN111222623A (en) * | 2018-11-26 | 2020-06-02 | 沈阳高精数控智能技术股份有限公司 | Ceramic glaze spraying robot glaze spraying technological parameter debugging method |
| CN113420771A (en) * | 2021-06-30 | 2021-09-21 | 扬州明晟新能源科技有限公司 | Colored glass detection method based on feature fusion |
| CN116589171A (en) * | 2023-07-14 | 2023-08-15 | 江西省博信玻璃有限公司 | Intelligent tempering method and system with automatic glass detection function |
| CN113420771B (en) * | 2021-06-30 | 2024-04-19 | 扬州明晟新能源科技有限公司 | Colored glass detection method based on feature fusion |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101042357A (en) * | 2007-04-11 | 2007-09-26 | 华中科技大学 | On-line detection device of defects in float glass based on machine vision |
-
2014
- 2014-05-12 CN CN201410198246.4A patent/CN105095273B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101042357A (en) * | 2007-04-11 | 2007-09-26 | 华中科技大学 | On-line detection device of defects in float glass based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 黄静 等: "玻璃钢化炉可视化看板系统的设计", 《仪表技术与传感器》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528637A (en) * | 2015-11-26 | 2016-04-27 | 江南大学 | Diagnosis method based on linear interpolation type fuzzy neural network |
| CN105528637B (en) * | 2015-11-26 | 2018-06-22 | 江南大学 | Diagnostic method based on linear interpolation type fuzzy neural network |
| CN111222623A (en) * | 2018-11-26 | 2020-06-02 | 沈阳高精数控智能技术股份有限公司 | Ceramic glaze spraying robot glaze spraying technological parameter debugging method |
| CN110648727A (en) * | 2019-10-30 | 2020-01-03 | 华南理工大学 | Preparation method of glass material with specific physical properties |
| CN110648727B (en) * | 2019-10-30 | 2021-09-21 | 华南理工大学 | Preparation method of glass material with specific physical properties |
| CN113420771A (en) * | 2021-06-30 | 2021-09-21 | 扬州明晟新能源科技有限公司 | Colored glass detection method based on feature fusion |
| CN113420771B (en) * | 2021-06-30 | 2024-04-19 | 扬州明晟新能源科技有限公司 | Colored glass detection method based on feature fusion |
| CN116589171A (en) * | 2023-07-14 | 2023-08-15 | 江西省博信玻璃有限公司 | Intelligent tempering method and system with automatic glass detection function |
| CN116589171B (en) * | 2023-07-14 | 2024-01-09 | 江西省博信玻璃有限公司 | Intelligent tempering method and system with automatic glass detection function |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105095273B (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105095273A (en) | Fuzzy BP neural network based glass tempering process parameter setting method | |
| CN106773682A (en) | Based on the glass furnace bottom of pond temperature intelligent forecast Control Algorithm that time lag is dynamically determined | |
| CN105740467B (en) | A kind of method for digging of C-Mn steel industry big data | |
| CN108311544B (en) | A kind of roll-force Parameter Self-learning method and device | |
| CN101391268B (en) | Reverse optimization method of steel plate rolling and cooling controlling-process temperature institution | |
| Tian et al. | TS fuzzy neural network predictive control for burning zone temperature in rotary kiln with improved hierarchical genetic algorithm | |
| CN103745273A (en) | Semiconductor fabrication process multi-performance prediction method | |
| CN108038797B (en) | Two-term coefficient method and multi-target planning hybrid evaluation method for power distribution network scheduling control level | |
| CN109858780B (en) | Steelmaking-continuous casting production scheduling optimization method | |
| CN107807518A (en) | A kind of Multiobjective Optimal Operation method of Chemical Manufacture raw material still-process | |
| US20190185960A1 (en) | Forging Method | |
| CN109492839A (en) | A kind of mineral hot furnace operating condition prediction technique based on RNN-LSTM network | |
| CN105084729A (en) | Technological parameter setting method of glass toughening technology | |
| CN106503389A (en) | A kind of numerical method of quick determination drop stamping production line process parameter | |
| CN108062583A (en) | A kind of heating furnace technique parameter optimization method towards energy-saving and emission-reduction | |
| Xi-jin et al. | Coal blending optimization of coal preparation production process based on improved GA | |
| CN104298214A (en) | Comprehensive optimization and control method for blast furnace molten iron productive process | |
| CN106048460A (en) | Longquan Tianfu sword and manufacturing process thereof | |
| CN112053037A (en) | Flexible PCB workshop scheduling optimization method and system | |
| CN109399959B (en) | Method for reducing stress of tempered glass and method for recycling defective glass | |
| CN107858492B (en) | Change the alloy carburizing steel preprocess method of netted acicular ferrite in microscopic structure | |
| CN104182853B (en) | A kind of pilot's Task Assignment Model is established and its method for solving | |
| CN106771095A (en) | The relation of glass batch and quality is based on the quantitative analysis method of rough set and ANFIS | |
| CN110526561A (en) | A kind of production technology of tempered glass | |
| CN110301664A (en) | A kind of quantitatively characterizing method of tobacco processing line processing intensity |
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 |