CN106862284A - A kind of cold rolled sheet signal mode knows method for distinguishing - Google Patents
A kind of cold rolled sheet signal mode knows method for distinguishing Download PDFInfo
- Publication number
- CN106862284A CN106862284A CN201710181131.8A CN201710181131A CN106862284A CN 106862284 A CN106862284 A CN 106862284A CN 201710181131 A CN201710181131 A CN 201710181131A CN 106862284 A CN106862284 A CN 106862284A
- Authority
- CN
- China
- Prior art keywords
- network
- plate shape
- layer
- quantum
- training
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B38/00—Methods or devices for measuring, detecting or monitoring specially adapted for metal-rolling mills, e.g. position detection, inspection of the product
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B37/00—Control devices or methods specially adapted for metal-rolling mills or the work produced thereby
- B21B37/28—Control of flatness or profile during rolling of strip, sheets or plates
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B21—MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVING MATERIAL; PUNCHING METAL
- B21B—ROLLING OF METAL
- B21B38/00—Methods or devices for measuring, detecting or monitoring specially adapted for metal-rolling mills, e.g. position detection, inspection of the product
- B21B38/04—Methods or devices for measuring, detecting or monitoring specially adapted for metal-rolling mills, e.g. position detection, inspection of the product for measuring thickness, width, diameter or other transverse dimensions of the product
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Mechanical Engineering (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Control Of Metal Rolling (AREA)
Abstract
A kind of cold rolled sheet signal mode knows method for distinguishing, and the method content is comprised the following steps:Each measuring section plate shape measurement value of cold-strip steel width of plate profile instrument on-line measurement is gathered, each measuring section plate shape value is obtained;The initial data that plate profile instrument is exported is input to a neutral net for n-layer as feature extraction layer, it is main to allow network to automatically extract feature to exclude artificial vestige by training;Plate shape identification is carried out using the modified quantum nerve network based on genetic algorithm.Be applied to the modified quantum nerve network of the multilayer excitation function of genetic algorithm optimization in Flatness Pattern identification technology by the present invention, the training effectiveness of network is significantly improved, the precision that is run into traditional plate shape recognition methods is efficiently solved and real-time is not ideal enough, the complicated network structure and training time is long and the problems such as stability and poor robustness.
Description
Technical field
The invention belongs to cold rolled sheet field, it is related to a kind of cold rolled sheet signal mode and knows method for distinguishing.
Background technology
In cold rolled sheet production process, the pattern-recognition of cold rolled sheet shape is the important composition portion of cold rolled sheet shape control system
Point, the accuracy of identification to flatness defect state in this link has vital effect to follow-up plate shape control effect.It is logical
The stress signal data crossed in plate profile instrument collection production scene sheet material are recognized, and judge the class of current flatness defect state
Type, and the controlling organization of plate shape is fed back to, by the degree of imperfection for controlling the adjustment of executing agency to reduce plate shape, finally make plate shape
Output meet the production standard of technique, so the recognition methods of high accuracy plate shape defect mode should be sought and be studied.
Traditional Shape signal mode identification method is based on the Factoring Polynomials method of least square method and improved
Orthogonal polynomial regression decomposition method, these method interference rejection abilities are poor, in theory existing defects, do not meet flatness defect distribution
Essence, it is difficult to meet high-precision shape control demand.Plate shape recognition methods based on fuzzy theory is easy and effective, but precision
It is not ideal enough with real-time, practical application it is little.In recent years, experts and scholars in the art have studied based on neutral net
The technology of the aspect such as Shape signal identification and application, achieves extraordinary effect.But the complexity due to neutral net and
The deficiency of existing optimisation technique, causes to debate knowledge model structure complexity based on neural network filter system, and have net
The network training time is long, stability difference the problems such as.
Plate shape refers to distribution situation of the Strip internal residual stress along plate width direction, and the task of plate shape identification is exactly that handle exists
One group of tension distribution centrifugal pump of line detection is mapped as less several characteristic parameters by certain treatment, and can be preferably anti-
Reflect the classification situation of flatness defect.The quantum nerve network of multilayer excitation function uses the superposition of multiple Sigmoid functions, by spy
Levy space and be divided into more multiple level, to represent more states, there are multiple different amounts by the hidden layer excitation function being superimposed
Son interval, and the different quantum level of each quantum interval correspondence or ladder, you can to correspond to different space structures respectively.
In the training process of model, when sample data enters feature space, it is not necessary in whole feature space be to be gone in the range of [0,1]
Search for corresponding space structure, but the removal search in the space structure being layered, so that sample information is quickly reflected
It is mapped on corresponding magnitude or ladder;Meanwhile, if the reality output of model is larger with desired output error, by appropriate
Learning algorithm adjustment quantum interval, the quantum level that has divided of change or ladder width, to adapt to the knot of available data sample
Structure characteristic, more preferably, to be quickly mapped on corresponding space structure, makes the uncertainty in sample data by quantum neural
Network Capture simultaneously quantifies, and so as to greatly shorten the training time of model, improves the accuracy and convergence rate of model prediction result.
On the other hand, intelligent optimization algorithm has obtained quick development in nearest decades, by intelligent optimization method
The optimal solution of many nonlinear optimal problems can be obtained.Intelligent optimization algorithm technology is applied to quantum nerve network training
In, modeling accuracy and efficiency are improve, it is the research direction for being rich in prospect and application value in Nonlinear Modeling field.This is also
High accuracy and efficient cold rolled sheet signal line model identification technology problem provide technical support and theoretical foundation.
In sum, research and development, with high accuracy and efficient cold rolled sheet shape signal line model recognition methods, are control
System processed provides reliable control foundation, is further to improve current cold so as to produce high-quality cold-rolled plate and strip product
Roll a key technical problem urgently to be resolved hurrily of belt plate shape controlled level.
The content of the invention
Know method for distinguishing it is an object of the invention to provide a kind of cold rolled sheet signal mode, the method can be solved effectively
The precision and real-time run into using traditional plate shape mode identification method not enough, debate knowledge model structure complexity and net training time
Long, the technical problem such as stability difference can provide reliable control foundation, to improve the plate shape of cold-strip steel for control system
Control quality provides strong guarantee.
In order to solve the above problems, the technical scheme that the present invention is provided is:
A kind of cold rolled sheet signal mode knows method for distinguishing, and the method content is comprised the following steps:
Step 1 gathers each measuring section plate shape measurement value of cold-strip steel width of plate profile instrument on-line measurement, obtains each survey
Amount section plate shape value;The number of measuring section is made for m, theiIndividual measuring section plate shape measurement value is Fi;
The initial data that plate profile instrument is exported is input to step 2 neutral net for n-layer as feature extraction layer, mainly
Network is allowed to automatically extract feature to exclude artificial vestige by training;Mainly have because of plate shape left side wave, the right wave, middle wave,
Unrestrained eight kinds of defect modes in bilateral wave, right three points of waves, left three points of waves, four points of waves and sides, so making network be output as ai, i=1,
2,3...m, wherein m are output number, and m=8 is taken here, then the computing formula of each layer of network is as follows:
al i=f (Wlxl+bl), l=1,2 ... n
Wherein l is l layer of network;WlIt is l layer network weights;xlIt is l layers of input;
Step 3 carries out plate shape identification using the modified quantum nerve network based on genetic algorithm, by data processing module
Output aiAs the input of quantum nerve network, the output u of quantum nerve network1,u2,u3,u4It is the degree of membership of plate shape;
If u1>0, represent that plate shape has left side wave, if u1<0, represent that plate shape has the right wave;
If u2>0, represent that plate shape has middle wave, if u2<0, represent that plate shape has bilateral wave;
If u3>0, represent that plate shape has left three points of waves, if u3<0, represent that plate shape has right three points of waves;
If u4>0, represent that plate shape has four points of waves, if u4<0, represent that plate shape has wave in side.
In step 3, the modified quantum nerve network based on genetic algorithm, it sets up process includes following step
Suddenly:
1) control parameter of the quantum nerve network training study for Flatness Pattern identification is determined:
Specific control parameter includes:For the initial population number NP of network training genetic algorithm, the maximum of genetic algorithm
Study algebraically N, the desired value e of network training effect, and training sample number M;
2) treatment of training sample data:
The data that plate profile instrument is exported are directly inputted to network, allow the multilayer neural network of network front end to carry out automated characterization
Extract, be then input in the quantum nerve network of multilayer excitation function;
3) setting of multilayer excitation function modified quantum nerve network:
Multilayer excitation function modified quantum nerve network model uses fully connected topology, is by feature extraction layer, hidden
Constituted containing layer and output layer, feature extraction layer is also referred to as input layer;Model hidden layer neuron excitation function uses Multiple-quantum energy level
Transforming function transformation function, each multiple level transforming function transformation function is a series of n with quantum spacing biassThe linear of individual Sigmoid functions is folded
Plus, i.e. multilayer excitation function;Therefore, the output y of hidden layer neuronhIt is represented by:
Wherein V is the hidden layer weights of multilayer excitation function quantum nerve network, and a is characterized the output of extract layer, θrFor
Quantum is spaced, nsIt is quantum skip number;
4) by the weights of feature extraction layer network and biasing, the hidden layer weights of multilayer excitation function quantum nerve network, amount
Son interval and output layer weights are defined as the individuality vector of genetic algorithm, and NP individual vector is determined according to equiprobability random distribution
Population initial value;
5) Optimization Learning of individual vector is carried out according to the mutation operation of genetic algorithm, crossover operation and selection operation;Make
The network parameter obtained with Optimization Learning configures the modified quantum nerve network parameter of multilayer excitation function, by M instruction
Practice sample data to bring the modified quantum nerve network of multilayer excitation function into calculate output valve, then by calculating above-mentioned M
The quadratic sum of the arithmetical difference between the individual corresponding 4M network output valve of training sample data and its actual value;The arithmetical difference it is flat
Side and be defined as error criterion functional value;Mean net if desired value g of the error criterion functional value less than network training effect
Network training success, otherwise, should continue training network, and g should choose according to the scale of the quantity of training sample and network;
6) network parameter of record training study, obtains the one 15 modified amount of the multilayer excitation function of the output of input 4
Sub- nerve net;
After establishing network according to above-mentioned steps, then carry out plate shape identification.
The present invention has the beneficial effect that:
1) data that plate profile instrument is input into are carried out automatically extracting feature by a neutral net for multilayer, are automatically processed,
Human disturbance factor is excluded, the precision of Network Recognition plate shape can be significantly improved.
2) the modified quantum nerve network of the multilayer excitation function of genetic algorithm optimization is applied to Flatness Pattern identification
In technology, the training effectiveness of network is significantly improved, efficiently solve the precision that is run into traditional plate shape recognition methods and in real time
The not ideal enough, complicated network structure of the property and training time is long and the problems such as stability and poor robustness, for Strip Shape Control is provided
Reliable foundation.
Brief description of the drawings
Fig. 1 is implementing procedure figure of the present invention;
Fig. 2 is the modified quantum nerve network structural topology figure of type multilayer excitation function;
Fig. 3 is the modified quantum nerve network instruction of the multilayer excitation function based on genetic algorithm in one embodiment of the invention
Practice learning process figure;
Fig. 4 is to carry out resulting recognition effect figure after plate shape identification using the present invention.
Specific embodiment
With reference to instantiation and accompanying drawing, the present invention will be further described.
A kind of cold rolled sheet signal mode of the invention knows method for distinguishing, and Fig. 1 is the flow chart of one embodiment of the invention, it
Comprise the following steps:
Step 1 gathers each measuring section plate shape measurement value of cold-strip steel width of plate profile instrument on-line measurement, obtains each survey
Amount section plate shape value;The number for making measuring section is m, m=15, theiIndividual measuring section plate shape measurement value is Fi;
The initial data that plate profile instrument is exported is input to step 2 neutral net for n-layer as feature extraction layer, mainly
Network is allowed to automatically extract feature to exclude artificial vestige by training;Mainly there are left side wave, the right wave, centre due to plate shape
Unrestrained eight kinds of defect modes in unrestrained, bilateral wave, right three points of waves, left three points of waves, four points of waves and sides, so making network be output as ai, i=
1,2,3...m, wherein m are output number, and m=8 is taken here;Then the computing formula of each layer of network is as follows:
al i=f (Wlxl+bl), l=1,2 ... n
Wherein l is l layer of network;WlIt is l layer network weights;xlIt is l layers of input;
Step 3 carries out plate shape identification using the modified quantum nerve network based on genetic algorithm, by data processing module
Output aiAs the input of quantum nerve network;The output u of quantum nerve network1,u2,u3,u4It is the degree of membership of plate shape;
If u1>0, represent that plate shape has left side wave, if u1<0, represent that plate shape has the right wave;
If u2>0, represent that plate shape has middle wave, if u2<0, represent that plate shape has bilateral wave;
If u3>0, represent that plate shape has left three points of waves, if u3<0, represent that plate shape has right three points of waves;
If u4>0, represent that plate shape has four points of waves, if u4<0, represent that plate shape has wave in side.
According to such scheme, the process of setting up of the modified quantum nerve network based on genetic algorithm is:
1) control parameter of the quantum nerve network training study for Flatness Pattern identification is determined:
Specific control parameter includes:For the initial population number NP of network training genetic algorithm, the maximum of genetic algorithm
Study algebraically N, the desired value e of network training effect, and training sample number M;
2) treatment of training sample data:
The data that plate profile instrument is exported are directly inputted to network, allow the multilayer neural network of network front end to carry out automated characterization
Extract, be then input in the quantum nerve network of multilayer excitation function.
3) setting of multilayer excitation function modified quantum nerve network:
Multilayer excitation function modified quantum nerve network model uses fully connected topology as shown in Figure 2, is by spy
Levy extract layer, hidden layer and output layer to constitute, feature extraction layer is also referred to as input layer;Model hidden layer neuron excitation function is used
Multiple-quantum energy level transforming function transformation function, each multiple level transforming function transformation function is a series of n with quantum spacing biassIndividual Sigmoid letters
Several linear superpositions, i.e. multilayer excitation function;Therefore, the output y of hidden layer neuronhIt is represented by:
Wherein V is the hidden layer weights of multilayer excitation function modified quantum nerve network, and a is characterized the defeated of extract layer
Go out, θrFor quantum is spaced, nsIt is quantum skip number;
4) by the weights of feature extraction layer network and biasing, the hidden layer weights of multilayer excitation function quantum nerve network, amount
Son interval and output layer weights are defined as the individuality vector of genetic algorithm, and NP individual vector is determined according to equiprobability random distribution
Population initial value;
5) Optimization Learning of individual vector is carried out according to the mutation operation of genetic algorithm, crossover operation and selection operation;Make
The network parameter obtained with Optimization Learning configures the modified quantum nerve network parameter of multilayer excitation function, by M instruction
Practice sample data to bring the modified quantum nerve network of multilayer excitation function into calculate output valve, then by calculating above-mentioned M
The quadratic sum of the arithmetical difference between the individual corresponding 4M network output valve of training sample data and its actual value;The arithmetical difference it is flat
Side and be defined as error criterion functional value;Mean net if desired value g of the error criterion functional value less than network training effect
Network training success, otherwise, should continue training network, and g should choose according to the scale of the quantity of training sample and network;
6) network parameter of record training study, obtains the one 15 modified amount of the multilayer excitation function of the output of input 4
Sub- nerve net;After establishing network according to above-mentioned steps, then carry out plate shape identification.
The inventive method can be used for four-roller, six roller single chassis or multi-frame tandem mills.With a rolling of single chassis six
As a example by machine, the product of six-high cluster mill section rolling includes common plate, high-strength steel, part stainless steel and silicon steel etc..This example is used
Type be 900HC milling trains, the unit main technical performance index and device parameter are:
Mill speed:Max 300m/min, draught pressure:Max 8000KN, maximum rolling force square:60KN × m, curling
Power:4~80KN;Supplied materials thickness range:2.0~4.0mm, plate wide scope:460~780mm, finished product thickness:0.24~1.5mm;;
Working roll size:245 × the 900mm of φ of φ 270/, intermediate calender rolls size:320 × the 920mm of φ of φ 340/, support roller chi
It is very little:The φ 790 of φ 850/~× 850;
The maximum traversing amount of intermediate calender rolls:200mm, the positive/negative bending roller force of working roll (one side):400/254.5KN.
Plate Profile Measuring System uses the self-control plate profile instrument of 15 passages.
The specific calculation process that this example carries out the Shape signal of dead channels is:
(1) collection receives each measuring section plate shape measurement value in cold-strip steel direction of plate profile instrument on-line measurement.Plate profile instrument is cold
Roll and several are configured with strip width direction have the measured zone for determining width, each measured zone provides corresponding region
Plate shape value.Side is eaten from plate profile instrument fore side to transmission and has 15 effective measuring areas, its measuring section hop count m-15, i.e. Fi, i=
1,2 ... 15 units are plate shape international unit I.The data of plate profile instrument measurement are introduced directly into network and do not do artificial treatment.
(2) initial data that plate profile instrument is exported is input to a neutral net for n-layer as feature extraction layer, its is main
It is to allow network to automatically extract feature by training to exclude artificial vestige.Mainly there are left side wave, the right wave, centre due to plate shape
Unrestrained eight kinds of defect modes in unrestrained, bilateral wave, right three points of waves, left three points of waves, four points of waves and sides, so making network be output as ai, i=
1,2,3...m, wherein m are output number, and m=8 is taken here.Then the computing formula of each layer of network is as follows:
al i=f (Wlxl+bl), l=1,2 ... n
Wherein l is l layer of network;WlIt is l layer network weights;xlIt is l layers of input.
(3) plate shape identification is carried out using the modified quantum nerve network based on genetic algorithm, by data processing module
Output aiAs the input of quantum nerve network;The output u of quantum nerve network1,u2,u3,u4It is the degree of membership of plate shape;
If u1>0, represent that plate shape has left side wave, if u1<0, represent that plate shape has the right wave;
If u2>0, represent that plate shape has middle wave, if u2<0, represent that plate shape has bilateral wave;
If u3>0, represent that plate shape has left three points of waves, if u3<0, represent that plate shape has right three points of waves;
If u4>0, represent that plate shape has four points of waves, if u4<0, represent that plate shape has wave in side.
According to such scheme, as shown in figure 3, the process of setting up of the new quantum nerve network based on genetic algorithm is:
1. the control parameter of the quantum nerve network training study for Flatness Pattern identification is determined:
Design parameter includes:For the initial population number NP of the genetic algorithm of network training, the most university of genetic algorithm
Algebraically N is practised, the desired value of network training effect is e, and training sample number is M;
2. the treatment of training sample data:
The data that plate profile instrument is exported are directly inputted to network, allow the multilayer neural network of network front end to carry out automated characterization
Interest is extracted to be then input in the quantum nerve network of multilayer excitation function.
3. the setting of the modified quantum nerve network of multilayer excitation function:
The modified quantum nerve network model of multilayer excitation function employs fully connected topology as shown in Figure 2, is
It is made up of feature extraction layer (input layer), hidden layer and output layer.Model hidden layer neuron excitation function uses Multiple-quantum energy
Level transforming function transformation function, each multiple level transforming function transformation function is a series of n with quantum spacing biassIndividual Sigmoid functions it is linear
Superposition, i.e. multilayer excitation function;Therefore, the output y of hidden layer neuronhIt is represented by:
Wherein V is the hidden layer weights of multilayer excitation function modified quantum nerve network, and a is characterized the defeated of extract layer
Go out, θrFor quantum is spaced, nsIt is quantum skip number.
4. by the weights of feature extraction layer network and biasing, the hidden layer power of multilayer excitation function modified quantum nerve network
Value, quantum interval and output layer weights are defined as the individuality vector of genetic algorithm, and NP is determined according to equiprobability random distribution
The population initial value of body vector;
5., according to the mutation operation of genetic algorithm, crossover operation and selection operation carry out individual vectorial Optimization Learning;Make
The network parameter obtained with Optimization Learning configures the improvement quantum nerve network parameter of multilayer excitation function, by M training
Sample data brings the modified quantum nerve network of multilayer excitation function into calculate output valve, then by calculating above-mentioned M
The quadratic sum of the arithmetical difference between the corresponding 4M network output valve of training sample data and its actual value.The arithmetical difference square
Be defined as error criterion functional value;Mean network if desired value g of the error criterion functional value less than network training effect
Train successfully, otherwise, training network should be continued, g should choose according to the scale of the quantity of training sample and network;
6. the network parameter of record training study, obtains the one 15 modified amount of the multilayer excitation function of the output of input 4
Sub- nerve net;
After establishing network according to above-mentioned steps, then carry out plate shape identification.
Fig. 4 gives this example and carries out resulting recognition effect figure after line model identification.Fig. 4-a are Network Recognitions
Flatness defect curve distribution figure;Fig. 4-b are the left side wave components after identification;Fig. 4-c are the bilateral unrestrained components after identification;Fig. 4-d
It is the left three points of waves after identification;Fig. 4-e be identification after side in wave.As can be seen from the figure technical scheme energy proposed by the present invention
Enough identification missions for completing flatness defect well.Solve that the conventional method precision and real-time that run into are not ideal enough, debate knowledge mould
Type complex structure is cut the technology that net training time is long, data characteristics processes human factor too big, stability and poor robustness and is asked
Topic, can put forward reliable control foundation for control system, and strong guarantee is provided to improve cold rolled sheet steel quality.
Above example is merely to illustrate Computation schema of the invention and feature, its object is to make technology in the art
Personnel will appreciate that present disclosure and according to implement, and protection scope of the present invention is not limited to examples detailed above.So it is all according to
The equivalent variations made according to disclosed principle, mentality of designing or modification, within protection scope of the present invention.
Claims (2)
1. a kind of cold rolled sheet signal mode knows method for distinguishing, it is characterised in that:The method content is comprised the following steps:
Step 1 gathers each measuring section plate shape measurement value of cold-strip steel width of plate profile instrument on-line measurement, obtains each measuring section
Plate shape value;The number of measuring section is made for m, theiIndividual measuring section plate shape measurement value is Fi;
The initial data that plate profile instrument is exported is input to a neutral net for n-layer as feature extraction layer for step 2, is mainly passed through
Training allows network to automatically extract feature to exclude artificial vestige;Mainly there is left side wave, the right wave, middle wave, bilateral because of plate shape
Unrestrained eight kinds of defect modes in wave, right three points of waves, left three points of waves, four points of waves and sides, so making network be output as ai, i=1,2,
3...m, wherein m is output number, and m=8 is taken here, then the computing formula of each layer of network is as follows:
al i=f (Wlxl+bl), l=1,2 ... n
Wherein l is l layer of network;WlIt is l layer network weights;xlIt is l layers of input;
Step 3 carries out plate shape identification using the modified quantum nerve network based on genetic algorithm, by the defeated of data processing module
Go out aiAs the input of quantum nerve network, the output u of quantum nerve network1,u2,u3,u4It is the degree of membership of plate shape;
If u1>0, represent that plate shape has left side wave, if u1<0, represent that plate shape has the right wave;
If u2>0, represent that plate shape has middle wave, if u2<0, represent that plate shape has bilateral wave;
If u3>0, represent that plate shape has left three points of waves, if u3<0, represent that plate shape has right three points of waves;
If u4>0, represent that plate shape has four points of waves, if u4<0, represent that plate shape has wave in side.
2. a kind of cold rolled sheet signal mode according to claim 1 knows method for distinguishing, it is characterised in that:It is described based on something lost
The modified quantum nerve network of propagation algorithm, it is set up process and comprises the following steps:
1) control parameter of the quantum nerve network training study for Flatness Pattern identification is determined:
Specific control parameter includes:For the initial population number NP of network training genetic algorithm, the maximum study of genetic algorithm
Algebraically N, the desired value e of network training effect, and training sample number M;
2) treatment of training sample data:
The data that plate profile instrument is exported are directly inputted to network, are allowed the multilayer neural network of network front end to carry out automated characterization and is carried
Take, be then input in the quantum nerve network of multilayer excitation function;
3) setting of multilayer excitation function modified quantum nerve network:
Multilayer excitation function modified quantum nerve network model uses fully connected topology, is by feature extraction layer, hidden layer
Constituted with output layer, feature extraction layer is also referred to as input layer;Model hidden layer neuron excitation function is converted using Multiple-quantum energy level
Function, each multiple level transforming function transformation function is a series of n with quantum spacing biassThe linear superposition of individual Sigmoid functions, i.e.,
Multilayer excitation function;Therefore, the output y of hidden layer neuronhIt is represented by:
Wherein V is the hidden layer weights of multilayer excitation function quantum nerve network, and a is characterized the output of extract layer, θrFor between quantum
Every nsIt is quantum skip number;
4) by between the weights of feature extraction layer network and biasing, hidden layer weights, the quantum of multilayer excitation function quantum nerve network
Every the individuality vector that genetic algorithm is defined as with output layer weights, the NP kind of individual vector is determined according to equiprobability random distribution
Group's initial value;
5) Optimization Learning of individual vector is carried out according to the mutation operation of genetic algorithm, crossover operation and selection operation;Using excellent
Chemical acquistion to network parameter configure the modified quantum nerve network parameter of multilayer excitation function, by M training sample
Notebook data brings the modified quantum nerve network of multilayer excitation function into calculate output valve, then by calculating above-mentioned M instruction
Practice the quadratic sum of the arithmetical difference between the corresponding 4M network output valve of sample data and its actual value;The quadratic sum of the arithmetical difference
It is defined as error criterion functional value;Mean that network is instructed if desired value g of the error criterion functional value less than network training effect
Practice successfully, otherwise, training network should be continued, g should choose according to the scale of the quantity of training sample and network;
6) network parameter of record training study, obtains the one 15 modified quantum god of the multilayer excitation function of the output of input 4
Through net;
After establishing network according to above-mentioned steps, then carry out plate shape identification.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710181131.8A CN106862284B (en) | 2017-03-24 | 2017-03-24 | A kind of cold rolled sheet signal mode knowledge method for distinguishing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710181131.8A CN106862284B (en) | 2017-03-24 | 2017-03-24 | A kind of cold rolled sheet signal mode knowledge method for distinguishing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106862284A true CN106862284A (en) | 2017-06-20 |
CN106862284B CN106862284B (en) | 2018-09-04 |
Family
ID=59173517
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710181131.8A Active CN106862284B (en) | 2017-03-24 | 2017-03-24 | A kind of cold rolled sheet signal mode knowledge method for distinguishing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106862284B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109063728A (en) * | 2018-06-20 | 2018-12-21 | 燕山大学 | A kind of fire image deep learning mode identification method |
CN109668842A (en) * | 2019-02-21 | 2019-04-23 | 岭南师范学院 | A kind of shellfish heavy metal pollution rapid detection method |
CN110348131A (en) * | 2019-07-15 | 2019-10-18 | 燕山大学 | A kind of FPGA implementation method of RBF plate shape identification model |
CN112270649A (en) * | 2020-09-30 | 2021-01-26 | 燕山大学 | Improved deep learning-based plate shape identification method |
CN113172097A (en) * | 2021-05-27 | 2021-07-27 | 燕山大学 | Cold-rolled strip shape pattern recognition method and system based on depth self-coding |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06142725A (en) * | 1992-10-29 | 1994-05-24 | Kobe Steel Ltd | Blanking method |
JPH07290125A (en) * | 1994-04-22 | 1995-11-07 | Nippon Steel Corp | Method for scheduling physical distribution in rolling mill |
JP2004148377A (en) * | 2002-10-31 | 2004-05-27 | Jfe Steel Kk | Finish rolling method |
CN101097581A (en) * | 2006-06-27 | 2008-01-02 | 宝山钢铁股份有限公司 | Processing parameter setting method of tension straightening machine set correct roller |
CN101846969A (en) * | 2009-03-23 | 2010-09-29 | 上海都峰智能科技有限公司 | Temper mill |
CN103341505A (en) * | 2013-06-25 | 2013-10-09 | 中冶南方工程技术有限公司 | Cold-rolled sheet shape signal online mode recognition method |
CN103745101A (en) * | 2013-12-30 | 2014-04-23 | 中冶南方(武汉)自动化有限公司 | Improved neural network algorithm based forecasting method of set value of rolling force of medium plate |
-
2017
- 2017-03-24 CN CN201710181131.8A patent/CN106862284B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06142725A (en) * | 1992-10-29 | 1994-05-24 | Kobe Steel Ltd | Blanking method |
JPH07290125A (en) * | 1994-04-22 | 1995-11-07 | Nippon Steel Corp | Method for scheduling physical distribution in rolling mill |
JP2004148377A (en) * | 2002-10-31 | 2004-05-27 | Jfe Steel Kk | Finish rolling method |
CN101097581A (en) * | 2006-06-27 | 2008-01-02 | 宝山钢铁股份有限公司 | Processing parameter setting method of tension straightening machine set correct roller |
CN101846969A (en) * | 2009-03-23 | 2010-09-29 | 上海都峰智能科技有限公司 | Temper mill |
CN103341505A (en) * | 2013-06-25 | 2013-10-09 | 中冶南方工程技术有限公司 | Cold-rolled sheet shape signal online mode recognition method |
CN103745101A (en) * | 2013-12-30 | 2014-04-23 | 中冶南方(武汉)自动化有限公司 | Improved neural network algorithm based forecasting method of set value of rolling force of medium plate |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109063728A (en) * | 2018-06-20 | 2018-12-21 | 燕山大学 | A kind of fire image deep learning mode identification method |
CN109668842A (en) * | 2019-02-21 | 2019-04-23 | 岭南师范学院 | A kind of shellfish heavy metal pollution rapid detection method |
CN110348131A (en) * | 2019-07-15 | 2019-10-18 | 燕山大学 | A kind of FPGA implementation method of RBF plate shape identification model |
CN112270649A (en) * | 2020-09-30 | 2021-01-26 | 燕山大学 | Improved deep learning-based plate shape identification method |
CN113172097A (en) * | 2021-05-27 | 2021-07-27 | 燕山大学 | Cold-rolled strip shape pattern recognition method and system based on depth self-coding |
Also Published As
Publication number | Publication date |
---|---|
CN106862284B (en) | 2018-09-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106862284B (en) | A kind of cold rolled sheet signal mode knowledge method for distinguishing | |
CN105510546B (en) | A kind of biochemical oxygen demand (BOD) BOD intelligent detecting methods based on self-organizing Recurrent RBF Neural Networks | |
CN104018831B (en) | A kind of fractured well reservoir evaluation methods | |
CN110033021A (en) | A kind of Fault Classification based on one-dimensional multichannel convolutional neural networks | |
CN107377634B (en) | A kind of hot-strip outlet Crown Prediction of Media method | |
CN106529818B (en) | Water quality assessment Forecasting Methodology based on Fuzzy Wavelet Network | |
CN108655186B (en) | Roll-force presetting method based on artificial neural network and mathematical model | |
CN108898215A (en) | A kind of sludge bulking INTELLIGENT IDENTIFICATION method based on two type fuzzy neural networks | |
CN106815971A (en) | A kind of landslide disaster forecasting procedure based on RBF neural | |
WO2023130666A1 (en) | Strip steel plate convexity prediction method based on data-driving and mechanism model fusion | |
CN110322014A (en) | A kind of finished cement specific surface area prediction technique based on BP neural network | |
CN109828089A (en) | A kind of on-line prediction method of the water quality parameter cultured water based on DBN-BP | |
CN104634265B (en) | A kind of mineral floating froth bed soft measurement method of thickness based on multiplex images Fusion Features | |
CN106446478A (en) | System and method for optimizing cutting process | |
Wang et al. | Application of mind evolutionary algorithm and artificial neural networks for prediction of profile and flatness in hot strip rolling process | |
CN107122851A (en) | A kind of lake water systems connects engineering proposal optimization model Sensitivity Analysis Method | |
CN106056235A (en) | Power transmission grid efficiency and benefit detection method based on Klee method and matter element extension model | |
CN106841545A (en) | Based on the tealeaves organoleptic quality evaluations method for improving fuzzy comprehensive evaluation method | |
CN107451594A (en) | A kind of various visual angles Approach for Gait Classification based on multiple regression | |
CN107274016A (en) | The strip exit thickness Forecasting Methodology of the random symmetrical extreme learning machine of algorithm optimization that leapfrogs | |
CN110479770A (en) | A kind of optimal control method of mill plate shape | |
CN108053066A (en) | The Forecasting Methodology of adjustable electric load in industrial production user's energy management system | |
CN107578100A (en) | The result feedback method and device of a kind of Kernel-based methods parameter | |
CN101546389A (en) | Primary direction neural network system | |
CN103316928B (en) | On-line cold-rolled flatness signal pattern recognition system |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |