CN102779287B - Ink key opening forecasting method having increment type learning capacity - Google Patents
Ink key opening forecasting method having increment type learning capacity Download PDFInfo
- Publication number
- CN102779287B CN102779287B CN201210164591.7A CN201210164591A CN102779287B CN 102779287 B CN102779287 B CN 102779287B CN 201210164591 A CN201210164591 A CN 201210164591A CN 102779287 B CN102779287 B CN 102779287B
- Authority
- CN
- China
- Prior art keywords
- neural network
- network
- ink
- fuzzy art
- printing
- 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.)
- Expired - Fee Related
Links
Landscapes
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for presetting digital printing ink and provides an ink key opening forecasting method based on a Fuzzy ART-BP mixed neural network algorithm of a Fuzzy ART neural network and a BP neural network. An input vector is stably classified by fully utilizing the self-learning, self-organizing and information fuzzy processing capacities of the Fuzzy ART neural network by the network; for each classification, nonlinear mapping is performed on the input and output vectors of a training sample by utilizing the BP neural network, namely, a mapping relation between image-text digital information and ink key control parameter of the training sample is established by taking printing on-site temperature, humidity, a printer rotating speed and a website area rate corresponding to an ink area as input vectors and an ink key opening value as an output vector; and a converged network is used for forecasting the ink key opening value of a new sample. The network learning is higher in pertinence, the iteration times of the BP network is reduced, the network has the increment type learning capacity and the generalization of the network is increased.
Description
Technical field
The invention belongs to digital printing art, be specifically related to a kind of black key aperture Forecasting Methodology with incremental learning ability.
Background technology
What use along with digital document in the front field of print increases, and digital flow process is applied to typography more and more, and digital flow process is in CTP(Computer To Plate) effect in technology is more and more important.Simultaneously printing enterprise face short slab, complexity and fast print production live and get more and more, so also printing enterprise is had higher requirement.Concerning printing enterprise, the effective way shortening printing setup time pre-sets ink exactly.Pre-estimate the best amount out of ink of black groove and to carry out Mo Jian preset, the start setup time of a lot of printing machine can be saved, reduce production cost, improve printing quality and efficiency, significantly can also reduce the waste of paper.
The method of printing machine control ink usage is actually and part printable on forme is divided into a lot of long and narrow region-ink-covered areas or ink channel in the direction of vertical long side, the ink amount of each ink-covered area can carry out fine adjustment according to the number of the area percentage-dot area percentage in this ink-covered area area on forme shared by areas, the area percentage of areas is higher, then the ink amount needed is more.Ink pre-setting is exactly obtain the information such as the dot area percentage in each ink-covered area according to film, forme or other carriers before starting printing, and sets up the funtcional relationship between dot area percentage and black key aperture, and then the inking amount of each ink-covered area on initial setting printing machine.Ink pre-setting technology is the representative art that digitizing technique enters print production link, is one of gordian technique important in Digital Printing Workflow, plays conclusive effect to printing quality and printing efficiency.
Traditional ink pre-setting technology based on BP neural network algorithm does not possess the incremental learning ability (on-line study) to training sample data, and generalization ability is weak, to the pattern can destroyed network during new samples data training study and remembered, the black key aperture of network is caused to predict the outcome not accurate enough.Fuzzy ART(fuzzy self-adaption is used to resonate neural network herein for solving this problem) and BP(BackPropagation) neural network, two kinds of neural networks are carried out integrated application and form a kind of Fuzzy ART-BP hybrid neural networks with incremental learning ability.But, ink pre-setting technology based on FuzzyART-BP hybrid neural networks algorithm can save the regulation time of printing machine start effectively, reduce the paper of start preparation, ink waste, reduce the labour intensity of press operator simultaneously, overcome press operator rule of thumb to regulate and control ink and measure the drawback brought and BP neural network to the drawback that can not realize incremental learning during new samples data learning training, improve the precision of prediction of ink pre-setting technology.
Summary of the invention
The present invention relates to a kind of black key aperture Forecasting Methodology with incremental learning ability.Domestic press is in order to improve printing quality and production efficiency further, some enterprises have introduced external various ink pre-setting system in succession, the black key of printing machine is adjusted in advance before start, but ink pre-setting system is unsatisfactory in actual applications, do not consider influence each other between black key, the impact of printing condition, incremental learning can not be realized to the learning training of new samples data simultaneously, cause the result of use not reaching expection.
Method of the present invention meets the printed sheet of GB printing standard for training sample with field density (evenly and without the surface color density that prints of blank ground), use Fuzzy ART-BP hybrid neural networks to have tutor to train to training sample and carry out the prediction of black key aperture to the sample of not training and training, wherein BP neural network is selected 3 layers (input layer, hidden layer and output layers).First by the original copy digitizing of client, namely complete layout data information is obtained, then by RIP(raster image processor) produce lattice information after rasterization process, and the lattice information produced is transformed the layout information producing low resolution, i.e. dot area percentage by software.(scene temperature is comprised to print field condition, on-the-spot humidity, printing machine rotating speed) dot area percentage corresponding with ink-covered area be as the input raw data of Fuzzy ART-BP hybrid neural networks, and [0 is carried out to input raw data, 1] input layer of Fuzzy ART-BP hybrid neural networks is delivered to after normalization, using black key aperture corresponding to training sample as the output layer raw data of Fuzzy ART-BP hybrid neural networks, also [0 is carried out to output raw data simultaneously, 1] output layer of Fuzzy ART-BP hybrid neural networks is delivered to after normalized, wherein the node in hidden layer of BP neural network is set in 21-35, optimal correction is carried out according to training result, finally choose node in hidden layer 23.
Call Fuzzy ART-BP algorithm routine to qualified actual printed sheet training study, thus establish printed sheet picture and text numerical information and printing condition (scene temperature, on-the-spot humidity, printing machine rotating speed) with the Nonlinear Mapping relation of black key aperture, first Fuzzy ART neural network carries out self-adaption cluster operation to printed sheet picture and text numerical information and printing condition, the study of BP neural metwork training is carried out for sorted data, when BP neural network convergence error is less than 10e-4, BP neural network finally restrains, preserve the weights and threshold of Nonlinear Mapping, and the weights of Fuzzy ART neural network are to database.If when the convergence error of BP neural network is not less than 10e-4, continued to training sample to iterative computation, until convergence error is less than 10e-4.Predict non-training sample with the FuzzyART-BP hybrid neural networks trained, black key aperture predicted value is sent in the real-time data base of printing machine by network or storage medium, and then controls press printing by control desk.The method effectively can shorten start setup time, improves printing efficiency and quality, realizes study new samples data being carried out to increment type.
Fuzzy ART-BP hybrid neural networks algorithm is the supervised learning mode adopted when training, its main thought is: the sample vector of input study, first self-adaptation Competition Clustering is carried out through Fuzzy ART neural network, the weights of amendment input vector generic, then back-propagation algorithm BP is used to carry out adjusting training repeatedly to the weights and threshold of BP neural network with regard to input vector generic, make the vector of actual output and desired output vector close as much as possible, train when the error sum of squares of network output layer is less than the error of specifying, preserve the weights and threshold of now network, Fuzzy ART-BP hybrid neural networks training study terminates.
In order to use the digitizing ink pre-setting technology based on Fuzzy ART-BP hybrid neural networks algorithm better, there is shown herein the step that Fuzzy ART-BP hybrid neural networks algorithm is detailed, being specifically described as follows:
(1) parameter introduction
Input vector P=(a of Fuzzy ART-BP hybrid neural networks
1, a
2... a
n)
t, 20≤n≤30; The i.e. dot area percentage of ink-covered area, scene temperature, on-the-spot humidity and the normalized data of printing machine rotating speed.
Fuzzy ART-BP hybrid neural networks desired output vector T=(s
1, s
2... s
q)
t, q=n-3; The normalized data of Jimo key aperture;
Hidden layer unit input vector S=(s
1, s
2... s
p)
t, p gets 21-35; Output vector B=(b
1, b
2... b
p)
t, p gets 21-35.
Output layer unit input vector L=(l
1, l
2... l
q)
t, q=n-3; Actual output vector C=(c
1, c
2, c
q)
t, q=n-3.
Input layer is to the connection weight W of hidden layer
i1j1,
i1=1,2 ..., p, j1=1,2 ..., n.
Hidden layer is to the connection weight V of output layer
ti1,
i1=1,2 ..., p, t=1,2 ..., q.
The output threshold value θ of each unit of hidden layer
i1,
i1=1,2,...,p。
The output threshold value y of each unit of output layer
t,
t=1,2,...,q。
α is the factor of momentum of BP neural network, 0 < α < 1.
β is the learning rate of BP neural network, 0 < β < 1.
(2) the concrete learning process of Fuzzy ART-BP hybrid neural networks
1) one group of input, the actual sample exported is chosen to P=(a
1, a
2... a
n)
t, T=(s
1, s
2... s
q)
t, and to sample to normalized, be then supplied to input layer and the output layer of Fuzzy ART-BP hybrid neural networks respectively.
2) to the input vector P=(a of the input layer of Fuzzy ART-BP hybrid neural networks
1, a
2... a
n)
tcarry out the input vector I of complementary operation as Fuzzy ART neural network, effectively can suppress the hyperplasia of Fuzzy ART neural network classification like this, concrete operations are such as formula shown in (1-1).
I=(a
1,a
2,a
3,...a
n,1-a
1,1-a
2,1-a
3,...,1-a
n) (1-1)
2) then by the principle of " the victor is a king " of Fuzzy ART neural network, input vector I is classified.
3) the BP neural network of respective classes numbering is set up with regard to input vector I generic numbering J, and the BP neural network initialization to classification J, namely give each connection weights W
i1j1and V
ti1, threshold value θ
i1with y
tgive the random number in (-1,1), finally set the convergence error ε of BP neural network algorithm;
4) by the sample of the original input vector P corresponding with the input vector I obtained after complementary operation and its desired output to P=(a
1, a
2... a
n)
t, T=(s
1, s
2... s
q)
t, after normalization, be input to the BP neural network of classification J.
5) with input layer sample data P=(a
1, a
2... a
n)
t, connect weights W
i1j1with threshold value θ
i1calculate the input s of each unit of hidden layer
i1, then use s
i1the output b of each unit of hidden layer is calculated by transport function
i1, transport function selects sigmoid function, and its form is:
, s
i1computing formula such as formula shown in (1-2), b
i1computing formula is such as formula shown in (1-3).
6) the output b of hidden layer is utilized
i1, connection weight V
ti1threshold value y
t, calculate the output l of output layer unit
t, shown in (1-4), then utilize transport function to calculate the response c of output layer unit
t, transport function selects sigmoid function, and its form is:
, shown in (1-5).
7) the actual output C=(c of Fuzzy ART-BP hybrid neural networks is utilized
1, c
2... c
q)
twith the desired output T=(s of network
1, s
2... s
q)
t, error of calculation E, shown in (1-6), if E is less than the convergence error ε of setting, then Fuzzy ART-BP hybrid neural networks convergence, finishing iteration also preserves weights W
i1j1, V
ti1with threshold value θ
i1, y
t; Otherwise continuation step 8), continues the convergence error whether error in judgement E is less than setting after revising weights and threshold matrix.
8) the actual output C=(c of Fuzzy ART-BP hybrid neural networks is utilized
1, c
2... c
q)
t, the desired output T=(s of network
1, s
2... s
q)
t, calculate the vague generalization error d of each unit of output layer
t, shown in (1-7).
d
t=(s
t-c
t)·c
t·(1-c
t) (1-7)
9) the vague generalization error d of each unit of output layer of the BP network of Fuzzy ART-BP hybrid neural networks is utilized
t, connection weight V
ti1with the output b of hidden layer
i1calculate the vague generalization error e of each unit of hidden layer
i1, shown in (1-8).
10) the vague generalization error d of each unit of output layer is utilized
twith the output b of each unit of hidden layer
i1revise connection weight V
ti1, shown in (1-9), threshold value y
tamendment is such as formula shown in (1-10).
11) the vague generalization error e of each unit of hidden layer is utilized
i1, the input P=(a of each unit of input layer
1, a
2... a
n) revise connection weights W
i1j1, shown in (1-11), threshold value θ
i1amendment is such as formula shown in (1-12).
12) with input layer sample data P=(a
1, a
2... a
n)
t, connect weights
and threshold value
calculate the input of each unit of hidden layer
, such as formula (1-13), then use
the output of each unit of hidden layer is calculated by transport function
, such as formula (1-14), transport function selects sigmoid function, and its form is:
13) output of hidden layer is utilized
, connection weight
threshold value
.Calculate the output of each unit of output layer
, such as formula (1-15), then utilize transport function to calculate the response of each unit of output layer
, transport function selects sigmoid function, and its form is:
such as formula (1-16).
14) the actual output of network is utilized
, the desired output T=(s of network
1, s
2... s
q)
t, calculating target function (error) E ', such as formula (1-17), if E ' is less than the convergence error ε of setting, then network convergence, finishing iteration also preserves weights and threshold; Otherwise return step 8), amendment weights and threshold matrix, continue iterative computation.
15) choosing next training sample to being supplied to Fuzzy ART-BP hybrid neural networks, continuing step 1)-15), until all training samples are complete to training.
Said process can represent with the learning training process flow diagram of figure (2) Fuzzy ART-BP hybrid neural networks.
Accompanying drawing explanation
Fig. 1 ink pre-setting technology schematic diagram.
The learning training process flow diagram of Fig. 2 Fuzzy ART-BP hybrid neural networks.
In the prediction of Fig. 3 ink key aperture, cyan predicts the outcome figure.
In the checking of Fig. 4 ink key aperture, cyan predicts the outcome figure.
Fig. 5 trains the visual representation figure of the neural network forecast difference of the network of 100 groups of data each ink-covered area corresponding.Fig. 6 trains the visual representation figure of the neural network forecast difference of the network of 150 groups of data each ink-covered area corresponding.
Embodiment
First by the digitizing of training sample original copy, complete layout data information is namely obtained, then by RIP(raster image processor) produce lattice information after rasterization process, and the lattice information produced is transformed by software produce layout information-dot area percentage.The relative humidity at printing scene is 30%, printing machine rotating speed is 4000/hour, temperature is 25 DEG C, because each data unit gathered is inconsistent, in order to accelerate the convergence of training network, thus [0 must be carried out to data, 1] normalized, do normalized to three conditions to be above respectively: 0.3,0.4,0.25.The full lattice of ink fountain used are 100, and when being normalized, the actual black key aperture of standby is divided by 100, and the value of dot area percentage, between [0-100%], also directly gets fractional value without normalized.Testing ground temperature, testing ground humidity, printing machine rotating speed are the principal elements affecting black key aperture, also have between adjacent black key and influence each other, so the input using certain any dot area percentage as BP neural network that can not be single, so the neuron number determining input layer is 23, comprise the dot area percentage of scene temperature, on-the-spot humidity, printing machine rotating speed and 20 ink-covered areas; The determination of output layer nodes: output layer node is followed successively by black key aperture corresponding to 20 ink-covered area dot area percentages; Node in hidden layer is set in 23.Learning rate β=0.4, factor of momentum α=0.9, the error ε expected=10e-4; Learning rate η=0.15 of Fuzzy ART neural network, alarm threshold ρ=0.95 of Fuzzy ART neural network.
Fuzzy ART-BP hybrid neural networks algorithm and BP neural network algorithm have obvious advantage can carry out incremental learning to new samples data exactly, carry out the test of precision of prediction below for this function.Because Fuzzy ART-BP hybrid neural networks algorithm has very strong analysis ability to more data, it predicts the outcome and BP neural network prediction result is significantly increased.Therefore, existing data basis utilizes Monte Carlo simulation to simulate more data, be supplied to Fuzzy ART-BP hybrid neural networks and carry out training study.Wherein as shown in the table after cyan existing partial data normalized:
Network input data:
Network output data:
Under identical printing environment, namely, under printing machine rotating speed, temperature and air humidity same case, (CMYK is the initial of 4 kinds of printing-ink titles: cyan Cyan, carmetta Magenta, yellow Yellow, K get is black the last letter for four looks that the dot area percentage that same printed sheet is identical is corresponding.) black key aperture not identical yet, so neural network will be set up to existing data and emulated data respectively according to four looks, and carry out training study and the prediction of respective neural network.
Adopt, from low capacity to the training method of Large Copacity training sample transition, network supervised learning is carried out to the ink pre-setting technology based on Fuzzy ART-BP hybrid neural networks, namely carry out there is tutor's training study according to the detailed step of Fuzzy ART-BP hybrid neural networks algorithm in summary of the invention, verify the superiority of the ink pre-setting technology of itself and BP neural network.First choose 100 groups of samples and to based on BP neural network ink pre-setting technology with based on Fuzzy ART-BP hybrid neural networks ink pre-setting technology, e-learning is carried out respectively to data, after the convergence error of its network reaches the convergence error ε=10e-4 of setting, preserve the weights and threshold etc. of each automatic network; Then get one group of data from sample to data and be input to the two kinds of ink pre-setting systems trained, carry out the prediction of black key aperture, wherein cyan predicts the outcome as shown in Figure 3.Finally, again choose other 50 groups of samples and learn the ink pre-setting technology of data to two neural networks, get the predictive ability that same group of data verifies network again, cyan prediction effect as shown in Figure 4.
In order to more intuitively find out the precision of prediction of Fuzzy ART-BP neural network algorithm and BP neural network algorithm, the actual value of black key aperture time qualified for actual printing printing and two kinds of neural network prediction values are subtracted each other the difference obtained herein and be called neural network forecast difference, shown in Fig. 5, be the visual representation of the neural network forecast difference of network each ink-covered area corresponding of training 100 groups of data.The visual representation of the neural network forecast difference of network each ink-covered area corresponding of training 150 groups of data is shown in Fig. 6.
As can be seen from Fig. 3 to Fig. 6, the prediction effect based on the ink pre-setting technology of Fuzzy ART-BP hybrid neural networks is considerable, and prediction difference all controls substantially about ± 2%.Along with training sample data amount increases, the prediction difference variation range of BP neural network strengthens, and part reaches ± 3%, and the pattern that the sample newly learnt has destroyed network and remembered is described, do not possess the ability of incremental learning, cause the error that predicts the outcome amid a sharp increase.Compared with BP neural network, Fuzzy ART-BP hybrid neural networks can carry out adaptive classification study for input data, then BP neural network learning is carried out with regard to sorted pattern, make network more targeted to training mode, reduce as much as possible and destroy network memory pattern degree to after new model study, predict the outcome and substantially also control about ± 2%, improve neural network forecast accuracy, network the is possessed ability of incremental learning.
Claims (1)
1. there is a black key aperture Forecasting Methodology for incremental learning ability, comprise the following steps:
1) with field density namely evenly and without blank the surface color density that prints meet the printed sheet of GB printing standard for training sample, for four look CMYK, i.e. cyan Cyan, magenta Magenta, yellow Yellow and black Black, sets up corresponding Fuzzy ART-BP hybrid neural networks respectively;
2) by printing condition, i.e. as Fuzzy ART-BP hybrid neural networks input layer input data after the dot area percentage normalized of scene temperature, on-the-spot humidity, printing machine rotating speed and 20-30 ink-covered area; As Fuzzy ART-BP hybrid neural networks output layer input data after ink key aperture normalized; Input layer input data are through Fuzzy ART self-adaption cluster, and the weights preserving class are to database, then set up BP neural network for certain the class data after cluster, regulate the node in hidden layer of suitable BP neural network, be set as 21-35; Input layer, hidden layer and output layer 3 layers of BP neural network algorithm procedural training module are used to have tutor's training study to training sample, when the convergence error of BP neural network is less than 10e-4, BP neural network finally restrains, and preserves the weights and threshold of BP neural network algorithm Nonlinear Mapping to database;
3) can by the dot area percentage of printed sheet and printing condition for non-training sample, i.e. scene temperature, on-the-spot humidity and printing machine rotating speed, be defeated by the prediction module of Fuzzy ART-BP hybrid neural networks algorithm, Fuzzy ART-BP hybrid neural networks algorithm routine utilizes the weights and threshold stored to carry out prediction and calculation to the scene temperature of input, on-the-spot humidity, printing machine rotating speed and dot area percentage, thus the black key aperture that the dot area percentage doping non-training sample is corresponding; Ink key aperture predicted value is sent to printing machine control desk by network or storage medium, and control desk receives data and automatically controls corresponding black key and is inked to printing machine and completes printing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210164591.7A CN102779287B (en) | 2012-05-24 | 2012-05-24 | Ink key opening forecasting method having increment type learning capacity |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210164591.7A CN102779287B (en) | 2012-05-24 | 2012-05-24 | Ink key opening forecasting method having increment type learning capacity |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102779287A CN102779287A (en) | 2012-11-14 |
CN102779287B true CN102779287B (en) | 2015-05-06 |
Family
ID=47124197
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210164591.7A Expired - Fee Related CN102779287B (en) | 2012-05-24 | 2012-05-24 | Ink key opening forecasting method having increment type learning capacity |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102779287B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106274051B (en) * | 2015-06-05 | 2018-03-27 | 靳鹏 | A kind of ink pre-setting method |
US10135146B2 (en) | 2016-10-18 | 2018-11-20 | At&T Intellectual Property I, L.P. | Apparatus and methods for launching guided waves via circuits |
CN109525956B (en) * | 2019-01-02 | 2020-06-12 | 吉林大学 | Energy-saving data collection method based on data-driven clustering in wireless sensor network |
CN110533167B (en) * | 2019-08-23 | 2022-02-08 | 中国石油化工股份有限公司 | Fault diagnosis method and system for electric valve actuating mechanism |
CN113094804A (en) * | 2020-01-08 | 2021-07-09 | 星河动力(北京)空间科技有限公司 | Method for predicting specific impulse performance of solid rocket engine with incremental learning capability |
-
2012
- 2012-05-24 CN CN201210164591.7A patent/CN102779287B/en not_active Expired - Fee Related
Non-Patent Citations (4)
Title |
---|
"Fuzzy ART神经网络在统计质量控制中的应用";高彦章等;《仪器仪表与检测》;20050531;全文 * |
"基于神经网络的智能油墨预置技术";咎涛等;《北京工业大学学报》;20110531;第37卷(第5期);657-660 * |
"基于色墨控制模型的印刷色彩质量控制研究";康建山;《工程科技I辑》;20111015;全文 * |
"小波包分解与Fuzzy ART神经网络在磨削振动监测中的应用";咎涛等;《北京工业大学学报》;20080731;第34卷(第7期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN102779287A (en) | 2012-11-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102779287B (en) | Ink key opening forecasting method having increment type learning capacity | |
CN105824586B (en) | A kind of more order layout print publishing optimization methods of combined type | |
JP6927344B2 (en) | Ink deposition uniformity compensation mechanism | |
CN106067042B (en) | Polarization SAR classification method based on semi-supervised depth sparseness filtering network | |
CN103258214A (en) | Remote sensing image classification method based on image block active learning | |
CN105676649A (en) | Control method for sewage treatment process based on self-organizing neural network | |
CN111553114A (en) | Intelligent color matching method for textile printing and dyeing based on data driving | |
CN107451596A (en) | A kind of classified nodes method and device | |
CN113469266A (en) | Electricity stealing behavior detection method based on improved deep convolutional neural network | |
CN103823943A (en) | Multicolor printing system splitting modeling method for maximizing color gamut | |
CN102529388B (en) | Total ink amount measuring method and device for ink jet printing equipment | |
RU2359308C2 (en) | Neural network regulator for controlling ship heading | |
TW202223560A (en) | Manufacturing parameter of manufacturing equipment adjustment control system and method thereof | |
CN103862858A (en) | Color separation method of spectrum of printing system with multicolor printing | |
CN102582242A (en) | Ink key opening prediction method in digital printing working process | |
CN100402708C (en) | Real-time control method in metal electrodeposition process based on neuron networks | |
CN110083317A (en) | It is a kind of that the printed data system that intelligently arrangement services is provided | |
CN113156832B (en) | Method and system for rapidly analyzing performance parameters of ink path system of offset press | |
WO2024204254A1 (en) | Planning assistance device, information processing method, and program | |
Yu et al. | The ink preset algorithm based on the model optimized by chaotic bee colony | |
Yu-zhao et al. | Price forecasting algorithm for coal and electricity based on PSO and RBF neural network | |
Wei et al. | Deep Neural Networks for New Product Form Design. | |
Yu et al. | Study on the ink flow model of Lithographic Printing | |
Repeta et al. | Analysis of factors and construction of prognostic quality models of flexographic printing process of packaging with solvent based inks | |
JPH06304724A (en) | Method and device for quality control |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150506 Termination date: 20180524 |
|
CF01 | Termination of patent right due to non-payment of annual fee |