CN110532318A - A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks - Google Patents
A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks Download PDFInfo
- Publication number
- CN110532318A CN110532318A CN201910824125.9A CN201910824125A CN110532318A CN 110532318 A CN110532318 A CN 110532318A CN 201910824125 A CN201910824125 A CN 201910824125A CN 110532318 A CN110532318 A CN 110532318A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- hidden layer
- layer
- curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
The present invention relates to floor data analysis systems, more particularly to a kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks, including controller and nerve network system, controller is connected with the data reception module for receiving injection molding machine operation data, controller is connected with the curve graph production module for storing data in data memory module to be converted to curve graph, controller is connected with the data computation module that the curve graph for being converted into curve graph production module calculates the multistage slope of curve, controller with input the label input module of qualitative label for manually receiving data according to data reception module and be connected, controller is connected with the multistage slope of curve that computing module for storing data calculates and corresponding qualitative label and to the tranining database that nerve network system training provides training data；Technical solution provided by the invention can overcome linear model present in the prior art to be difficult to solve non-linear relation and strong coupling defect between multivariable factor.
Description
Technical field
The present invention relates to floor data analysis systems, and in particular to a kind of injection molding machine operation based on more hidden layer neural networks
Floor data analysis system.
Background technique
With being constantly progressive for industrial automation, the integrated level and complexity of injection molding machine are continuously increased.Plastics belong to non-ox
Pause fluid, has the characteristics that physico-chemical property is complicated, Variable Factors are mostly strong with coupling, causes injection molding machine production process controllability
It is low, it is easy to frequently occur exception.Therefore, timely and effectively monitoring injection molding machine operating condition can provide important production guarantor for plastic injection
Barrier.
Currently, injection molding machine operating condition is mainly that the linear model based on principal component analysis is utilized to monitor injection molding machine operating condition number
According to.However, linear model is difficult to ensure reasonability for non-linear relation between multivariable factor and strong coupling.
Summary of the invention
(1) the technical issues of solving
For disadvantages mentioned above present in the prior art, the present invention provides a kind of injection moldings based on more hidden layer neural networks
Machine operating condition data analysis system can effectively overcome linear model present in the prior art to be difficult to solve multivariable factor
Between non-linear relation and strong coupling defect.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs:
A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks, including controller and nerve net
Network system, the controller are connected with the data reception module for receiving injection molding machine operation data, the controller be used for
By storing data in the data memory module be converted to curve graph curve graph production module be connected, the controller be used for
The data computation module for calculating the multistage slope of curve to the curve graph that curve graph production module is converted into is connected, the control
Device with input the label input module of qualitative label for manually receiving data according to the data reception module and be connected, the control
Device processed with for storing the multistage slope of curve and corresponding qualitative label that the data computation module calculates and to the nerve
The tranining database that network system training provides training data is connected, the controller with for storing the data computation module
The multistage slope of curve of calculating is simultaneously connected to the data repository that the nerve network system provides data to be analyzed；
Training data, number to be analyzed are transmitted by data transmission module between the controller and the nerve network system
According to the nerve network system includes neural networks with single hidden layer module, for receiving the data transmission module transmission training number
According to neural metwork training module, more hidden layer neural network modules and for receiving the data transmission module transmit training number
According to the neural network correction module of, data to be analyzed；
The neural networks with single hidden layer module includes for receiving the neural metwork training module transfer training data
Neural network input layer, first with initialization weight for Connection Neural Network input layer, peripheral sensory neuron output layer
Hidden layer, and for determining result of the training data in the neural network input layer after first hidden layer with corresponding
Property label match and the peripheral sensory neuron output layer that is adjusted the weight in first hidden layer according to matching result；
More hidden layer neural network modules include for receiving the neural metwork training module transfer training number
According to the neural network input layer of, data to be analyzed, adjusted for Connection Neural Network input layer, having for N neuron output layer
The first hidden layer, the second hidden layer ... the N-1 hidden layer of weight after whole are exported for Connection Neural Network input layer, N neuron
The N hidden layer with initialization weight of layer, and for by training data, the number to be analyzed in the neural network input layer
Matching is carried out and according to matching result to described with corresponding qualitative label according to the result after first hidden layer to N hidden layer
The N neuron output layer that weight in N hidden layer is adjusted.
Preferably, the various dimensions floor data stored in data memory module described in the curve graph production module generates bent
Line chart, the data computation module are segmented the curve in curve graph according to plastic injection workshop section to obtain multistage curve, and
The slope of curve on every section of curve is calculated separately using scheduled slope calculation formula.
Preferably, the slope of curve set on every section of curve obtains training data, data to be analyzed.
Preferably, the qualitative label includes operating normally label, failure label, suggesting shutting down label.
Preferably, the training process of the neural networks with single hidden layer module is as follows: the neural metwork training module will instruct
Practice data and be input to the neural network input layer one by one, until each data in training data are entered at least once
More than, via error back propagation process adjusting and the associated weight of the first hidden layer, so that from the peripheral sensory neuron
The result qualitative tag match corresponding with training data that output layer generates；
Abandoning the peripheral sensory neuron output layer and adding on the basis of first hidden layer has initialization weight
New hidden layer and the new neuron output layer being connect with the new hidden layer, to generate new more hidden layer neural networks.
Preferably, the training process of more hidden layer neural network modules is as follows: the neural metwork training module will instruct
Practice data and be input to the neural network input layer one by one, until each data in training data are entered at least once
More than, via error back propagation process adjusting and the associated weight of N hidden layer, so that defeated from the N neuron
The result qualitative tag match corresponding with training data that layer generates out；
Above-mentioned movement is repeated, until the hidden layer of added specified quantity.
Preferably, the data reception module establishes data transfer communications by CAN bus and injection molding machine.
(3) beneficial effect
Compared with prior art, a kind of injection molding machine operating condition number based on more hidden layer neural networks provided by the present invention
It is had the advantages that according to analysis system
1, data reception module receives injection molding machine operation data, and curve graph production module will deposit in the data memory module
Storage data are converted to curve graph, and the curve graph that data computation module is converted into curve graph production module calculates multistage curve
Slope, so that operating condition data to be converted to the controllable slope of curve；
2, tranining database stores the multistage slope of curve and correspond to qualitative label simultaneously that the data computation module calculates
Training data is provided to nerve network system training, so that nerve network system is constantly improve by self deep learning,
Nerve network system precision of analysis is effectively promoted, and can effectively solve the problem that the nonlinear dependence between multivariable factor
It is and strong coupling.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described.It should be evident that the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is present system schematic diagram；
Fig. 2 is neural networks with single hidden layer module diagram in invention figure 1；
Fig. 3 is more hidden layer neural network module schematic diagrames in invention figure 1.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described.Obviously, described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks, as shown in Figure 1 to Figure 3, packet
Controller and nerve network system are included, controller is connected with the data reception module for receiving injection molding machine operation data, controls
Device with for by storing data in data memory module be converted to curve graph curve graph production module be connected, controller be used for
The data computation module for calculating the multistage slope of curve to the curve graph that is converted into of curve graph production module is connected, controller be used for
Data manually received according to data reception module input the label input module of qualitative label and be connected, controller with for storing number
According to the multistage slope of curve of computing module calculating and corresponding qualitative label and training data is provided to nerve network system training
Tranining database be connected, the multistage slope of curve that controller is calculated with computing module for storing data and to neural network system
The data repository that system provides data to be analyzed is connected；
Training data, data to be analyzed, nerve are transmitted by data transmission module between controller and nerve network system
Network system includes neural networks with single hidden layer module, transmission module transmits the neural metwork training of training data for receiving data
Module, more hidden layer neural network modules and for receiving data transmission module transmission training data, data to be analyzed nerve net
Network correction module；
Neural networks with single hidden layer module includes the neural network for receiving neural metwork training module transfer training data
Input layer, for first hidden layer with initialization weight of Connection Neural Network input layer, peripheral sensory neuron output layer, and
For result of the training data in neural network input layer after the first hidden layer to be matched simultaneously with corresponding qualitative label
The peripheral sensory neuron output layer that the weight in the first hidden layer is adjusted according to matching result；
More hidden layer neural network modules include for receiving neural metwork training module transfer training data, data to be analyzed
Neural network input layer, for Connection Neural Network input layer, N neuron output layer have adjustment after weight first
Hidden layer, the second hidden layer ... N-1 hidden layer are initialized for Connection Neural Network input layer, having for N neuron output layer
The N hidden layer of weight, and for by neural network input layer training data, data to be analyzed are by the first hidden layer to the
Result after N hidden layer with corresponding qualitative label carries out matching and is adjusted according to matching result to the weight in N hidden layer
N neuron output layer.
The various dimensions floor data formation curve figure stored in curve graph production module data memory module, data calculate mould
Root tuber is segmented the curve in curve graph according to plastic injection workshop section to obtain multistage curve, and calculates public affairs using scheduled slope
Formula calculates separately the slope of curve on every section of curve.
Slope of curve set on every section of curve obtains training data, data to be analyzed.
Qualitative label includes operating normally label, failure label, suggesting shutting down label.
The training process of neural networks with single hidden layer module is as follows: neural metwork training module is defeated one by one by training data
Enter to neural network input layer, until each data in training data are entered at least more than once, is reversely passed via error
Process adjusting and the associated weight of the first hidden layer are broadcast, so that the result and training data pair that generate from peripheral sensory neuron output layer
The qualitative tag match answered；
Abandon peripheral sensory neuron output layer and on the basis of the first hidden layer addition have initialization weight new hidden layer with
And the new neuron output layer being connect with new hidden layer, to generate new more hidden layer neural networks.
The training process of more hidden layer neural network modules is as follows: neural metwork training module is defeated one by one by training data
Enter to neural network input layer, until each data in training data are entered at least more than once, is reversely passed via error
Process adjusting and the associated weight of N hidden layer are broadcast, so that the result generated from N neuron output layer is corresponding with training data
Qualitative tag match；
Above-mentioned movement is repeated, until the hidden layer of added specified quantity.
Data reception module establishes data transfer communications by CAN bus and injection molding machine.
Data reception module receives injection molding machine operation data, and curve graph makes module for storing data in data memory module
Curve graph is converted to, the curve graph that data computation module is converted into curve graph production module calculates the multistage slope of curve, thus
Operating condition data are converted to the controllable slope of curve.
The various dimensions floor data formation curve figure stored in curve graph production module data memory module, data calculate mould
Root tuber is segmented the curve in curve graph according to plastic injection workshop section to obtain multistage curve, and calculates public affairs using scheduled slope
Formula calculates separately the slope of curve on every section of curve.
Slope of curve set on every section of curve obtains training data, data to be analyzed.
The multistage slope of curve and corresponding qualitative label that tranining database storing data computing module calculates and to nerve
Network system training provides training data, so that nerve network system is constantly improve by self deep learning, effectively promotes mind
Accuracy through Network System Analysis result.
Nerve network system includes neural networks with single hidden layer module, for receiving data transmission module transmission training data
Neural metwork training module, more hidden layer neural network modules and transmission module transmission training data, to be analyzed for receiving data
The neural network correction module of data.
Neural networks with single hidden layer module includes the neural network for receiving neural metwork training module transfer training data
Input layer, for first hidden layer with initialization weight of Connection Neural Network input layer, peripheral sensory neuron output layer, and
For result of the training data in neural network input layer after the first hidden layer to be matched simultaneously with corresponding qualitative label
The peripheral sensory neuron output layer that the weight in the first hidden layer is adjusted according to matching result.
The training process of neural networks with single hidden layer module is as follows: neural metwork training module is defeated one by one by training data
Enter to neural network input layer, until each data in training data are entered at least more than once, is reversely passed via error
Process adjusting and the associated weight of the first hidden layer are broadcast, so that the result and training data pair that generate from peripheral sensory neuron output layer
The qualitative tag match answered；
Abandon peripheral sensory neuron output layer and on the basis of the first hidden layer addition have initialization weight new hidden layer with
And the new neuron output layer being connect with new hidden layer, to generate new more hidden layer neural networks.
More hidden layer neural network modules include for receiving neural metwork training module transfer training data, data to be analyzed
Neural network input layer, for Connection Neural Network input layer, N neuron output layer have adjustment after weight first
Hidden layer, the second hidden layer ... N-1 hidden layer are initialized for Connection Neural Network input layer, having for N neuron output layer
The N hidden layer of weight, and for by neural network input layer training data, data to be analyzed are by the first hidden layer to the
Result after N hidden layer with corresponding qualitative label carries out matching and is adjusted according to matching result to the weight in N hidden layer
N neuron output layer.
The training process of more hidden layer neural network modules is as follows: neural metwork training module is defeated one by one by training data
Enter to neural network input layer, until each data in training data are entered at least more than once, is reversely passed via error
Process adjusting and the associated weight of N hidden layer are broadcast, so that the result generated from N neuron output layer is corresponding with training data
Qualitative tag match；
Above-mentioned movement is repeated, until the hidden layer of added specified quantity.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, can't be such that the essence of corresponding technical solution departs from the spirit and scope of the technical scheme of various embodiments of the present invention.
Claims (7)
1. a kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks, including controller and neural network
System, it is characterised in that: the controller is connected with the data reception module for receiving injection molding machine operation data, the control
Device is connected with the curve graph production module for storing data in the data memory module to be converted to curve graph, the control
Device is connected with the data computation module that the curve graph for being converted into curve graph production module calculates the multistage slope of curve,
The controller and the label input module phase for manually inputting qualitative label according to data reception module reception data
Even, the controller and the multistage slope of curve and corresponding qualitative label for storing data computation module calculating and to
The tranining database that nerve network system training provides training data is connected, the controller with for storing the data
The multistage slope of curve of computing module calculating is simultaneously connected to the data repository that the nerve network system provides data to be analyzed；
Training data, data to be analyzed are transmitted by data transmission module between the controller and the nerve network system,
The nerve network system includes neural networks with single hidden layer module, for receiving the data transmission module transmission training data
Neural metwork training module, more hidden layer neural network modules and for receive data transmission module transmission training data, to
Analyze the neural network correction module of data；
The neural networks with single hidden layer module includes the nerve for receiving the neural metwork training module transfer training data
Network input layer, for Connection Neural Network input layer, peripheral sensory neuron output layer have initialization weight the first hidden layer,
And for by result of the training data in the neural network input layer after first hidden layer and corresponding qualitative mark
Label carry out the peripheral sensory neuron output layer for matching and being adjusted according to matching result to the weight in first hidden layer；
More hidden layer neural network modules include for receive the neural metwork training module transfer training data, to
Analyze data neural network input layer, for Connection Neural Network input layer, N neuron output layer have adjustment after weigh
First hidden layer of weight, the second hidden layer ... N-1 hidden layer, for Connection Neural Network input layer, the tool of N neuron output layer
There is the N hidden layer of initialization weight, and for passing through the training data in the neural network input layer, data to be analyzed
Result after first hidden layer to N hidden layer and corresponding qualitative label carry out matching and hidden to the N according to matching result
The N neuron output layer that weight in layer is adjusted.
2. the injection molding machine operating condition data analysis system according to claim 1 based on more hidden layer neural networks, special
Sign is: the various dimensions floor data formation curve figure stored in data memory module described in the curve graph production module, institute
It states data computation module the curve in curve graph is segmented according to plastic injection workshop section to obtain multistage curve, and utilizes predetermined
Slope calculation formula calculate separately the slope of curve on every section of curve.
3. the injection molding machine operating condition data analysis system according to claim 2 based on more hidden layer neural networks, special
Sign is: the slope of curve set on every section of curve obtains training data, data to be analyzed.
4. the injection molding machine operating condition data analysis system according to claim 1 based on more hidden layer neural networks, special
Sign is: the qualitative label includes operating normally label, failure label, suggesting shutting down label.
5. the injection molding machine operating condition data analysis system according to claim 1 based on more hidden layer neural networks, special
Sign is: the training process of the neural networks with single hidden layer module is as follows: the neural metwork training module is by training data one
A one is input to the neural network input layer, until each data in training data be entered at least more than once, warp
By error back propagation process adjusting and the associated weight of the first hidden layer, so that raw from the peripheral sensory neuron output layer
At result qualitative tag match corresponding with training data；
Abandoning the peripheral sensory neuron output layer and adding on the basis of first hidden layer has the new hidden of initialization weight
Layer and the new neuron output layer being connect with the new hidden layer, to generate new more hidden layer neural networks.
6. the injection molding machine operating condition data analysis system according to claim 1 based on more hidden layer neural networks, special
Sign is: the training process of more hidden layer neural network modules is as follows: the neural metwork training module is by training data one
A one is input to the neural network input layer, until each data in training data be entered at least more than once, warp
By error back propagation process adjusting and the associated weight of N hidden layer, so that being generated from the N neuron output layer
Result qualitative tag match corresponding with training data；
Above-mentioned movement is repeated, until the hidden layer of added specified quantity.
7. the injection molding machine operating condition data analysis system according to claim 1 based on more hidden layer neural networks, special
Sign is: the data reception module establishes data transfer communications by CAN bus and injection molding machine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910824125.9A CN110532318A (en) | 2019-09-02 | 2019-09-02 | A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910824125.9A CN110532318A (en) | 2019-09-02 | 2019-09-02 | A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110532318A true CN110532318A (en) | 2019-12-03 |
Family
ID=68666262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910824125.9A Pending CN110532318A (en) | 2019-09-02 | 2019-09-02 | A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110532318A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113628457A (en) * | 2021-09-07 | 2021-11-09 | 重庆交通大学 | Intelligent control method and system for traffic signal lamp |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184454A (en) * | 2011-05-26 | 2011-09-14 | 浙江迦南科技股份有限公司 | Granulator formula generation method based on neural network system |
| CN103049792A (en) * | 2011-11-26 | 2013-04-17 | 微软公司 | Discriminative pretraining of Deep Neural Network |
| CN105574587A (en) * | 2016-01-21 | 2016-05-11 | 华中科技大学 | On-line condition process monitoring method for plastic injection moulding process |
| CN108171329A (en) * | 2017-12-13 | 2018-06-15 | 华南师范大学 | Deep learning neural network training method, number of plies adjusting apparatus and robot system |
| CN108229557A (en) * | 2017-12-29 | 2018-06-29 | 中国人民解放军陆军装甲兵学院 | The acceleration training method and system of a kind of neural network with label |
| CN109648816A (en) * | 2018-12-03 | 2019-04-19 | 武汉市衡德实业有限公司 | The method and system of injection molding machine operating condition are monitored in a kind of plastic injection molding process |
-
2019
- 2019-09-02 CN CN201910824125.9A patent/CN110532318A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184454A (en) * | 2011-05-26 | 2011-09-14 | 浙江迦南科技股份有限公司 | Granulator formula generation method based on neural network system |
| CN103049792A (en) * | 2011-11-26 | 2013-04-17 | 微软公司 | Discriminative pretraining of Deep Neural Network |
| CN105574587A (en) * | 2016-01-21 | 2016-05-11 | 华中科技大学 | On-line condition process monitoring method for plastic injection moulding process |
| CN108171329A (en) * | 2017-12-13 | 2018-06-15 | 华南师范大学 | Deep learning neural network training method, number of plies adjusting apparatus and robot system |
| CN108229557A (en) * | 2017-12-29 | 2018-06-29 | 中国人民解放军陆军装甲兵学院 | The acceleration training method and system of a kind of neural network with label |
| CN109648816A (en) * | 2018-12-03 | 2019-04-19 | 武汉市衡德实业有限公司 | The method and system of injection molding machine operating condition are monitored in a kind of plastic injection molding process |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113628457A (en) * | 2021-09-07 | 2021-11-09 | 重庆交通大学 | Intelligent control method and system for traffic signal lamp |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11214850B2 (en) | Prediction control method and system for component contents in rare earth extraction process | |
| US12259721B2 (en) | Industrial internet of things for intelligent control of filling production, control methods and media thereof | |
| CN1598720B (en) | Method for controlling process and process controller | |
| Gonzaga et al. | ANN-based soft-sensor for real-time process monitoring and control of an industrial polymerization process | |
| US20090319059A1 (en) | Apparatus and method for model predictive control (mpc) of a nonlinear process | |
| WO2020188328A1 (en) | Method of performing a process and optimizing control signals used in the process | |
| CN104657792A (en) | Early warning method based on rule engine and intelligent prediction | |
| CN120065935A (en) | Intelligent sensing and self-adaptive regulation-based dosing method, system, equipment and medium for tap water plant water manufacturing process unit | |
| CN110202768A (en) | A kind of barrel temperature control of injection mold method | |
| CN110221540B (en) | Control Method of Continuous Stirred Reactor System Based on Hammerstein Model | |
| CN104317269A (en) | A Comprehensive Predictive Iterative Learning Control Method Based on 2D Theory | |
| CN108008632A (en) | A kind of method for estimating state and system of the time lag Markov system based on agreement | |
| CN107512754A (en) | A kind of Powdered Activated Carbon Automatic Dosage Control of Additives system for water process | |
| CN110532318A (en) | A kind of injection molding machine operating condition data analysis system based on more hidden layer neural networks | |
| CN104330972A (en) | Comprehensive prediction iterative learning control method based on model adaptation | |
| CN108984851A (en) | A kind of Weighted Gauss model soft-measuring modeling method with time delay estimation | |
| US12058215B2 (en) | Methods, industrial internet of things systems, and storage mediums for controlling production line detection data | |
| CN103675010B (en) | The industrial melt index soft measurement instrument of support vector machine and method | |
| US11783194B1 (en) | Evolutionary deep learning with extended Kalman filter for modeling and data assimilation | |
| CN115485632B (en) | Contextual data modeling and dynamic process intervention for industrial plants | |
| CN207699239U (en) | A kind of Powdered Activated Carbon Automatic Dosage Control of Additives system for water process | |
| Krishnan et al. | Robust model predictive control of an industrial solid phase polymerizer | |
| CN113049793A (en) | Production detection verification method and device for Abs composite material | |
| CN114228008A (en) | Plastic molding method and system | |
| Lightbody et al. | Neural network modelling of a polymerisation reactor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191203 |