CN109002862A - A kind of flexible measurement method neural network based and system towards copper ore floatation machine - Google Patents
A kind of flexible measurement method neural network based and system towards copper ore floatation machine Download PDFInfo
- Publication number
- CN109002862A CN109002862A CN201810960867.XA CN201810960867A CN109002862A CN 109002862 A CN109002862 A CN 109002862A CN 201810960867 A CN201810960867 A CN 201810960867A CN 109002862 A CN109002862 A CN 109002862A
- Authority
- CN
- China
- Prior art keywords
- neural network
- nng
- algorithm
- copper ore
- floatation machine
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F1/00—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow
- G01F1/05—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow by using mechanical effects
- G01F1/20—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow by using mechanical effects by detection of dynamic effects of the flow
- G01F1/32—Measuring the volume flow or mass flow of fluid or fluent solid material wherein the fluid passes through a meter in a continuous flow by using mechanical effects by detection of dynamic effects of the flow using swirl flowmeters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G19/00—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups
- G01G19/08—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for incorporation in vehicles
- G01G19/12—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for incorporation in vehicles having electrical weight-sensitive devices
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Fluid Mechanics (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to a kind of flexible measurement methods neural network based and system towards copper ore floatation machine, it is characterized in that, the following steps are included: S1:RBF Learning Algorithm step, Basis Function Center c is sought based on K- means clustering method, NNG algorithm carries out input variable compression to RBF neural；S2: the selection of optimal N NG algorithm parameter and error prediction；The modeling of S3:NNG-RBF algorithm.
Description
Technical field
The invention belongs to software field of measuring technique, it is related to a kind of flexible measurement method neural network based and system, especially
It is a kind of flexible measurement method neural network based and system towards copper ore floatation machine.
Background technique
In modern industrial production, it to obtain more qualified high quality of products, and then increases economic efficiency, it is necessary to right
Product quality or the significant process variable closely related with product quality carry out strict control.
Fig. 1 is copper ore floatation machine device mineralization process block plan, which has mixed zone, traffic zone, Disengagement zone and foam
Four, area region is enriched with mineral step by step for separating the sulphur and copper in ore pulp, and the mineral in guarantee froth bed are not
Cause falls off, and foam can be flowed into successfully in foam tank.
In order to guarantee product quality and prevent the loss of copper and sulphur in ore pulp, need to the pH value to the device examined in real time
It surveys, so that pH value is kept in a certain range.However, PH meter is easy precipitated ore pulp due to the high vicidity of copper mine ore pulp
It encases, agglomerate that accurate, real-time detection can not be carried out.
On-line equipment cannot reach requirement well, it is therefore desirable to a kind of flexible measurement method instead of on-line analysis instrument.
And when facing the input variable of many complexity, how fast and accurately to realize and Effective selection is carried out to multiple input variables and is
Number compression, and extremely difficult is become to the prediction of the flotation device device pH value.
This is in place of the deficiencies in the prior art.Therefore, in view of the above-mentioned drawbacks in the prior art, provide design it is a kind of towards
The flexible measurement method neural network based and system of copper ore floatation machine are very to solve drawbacks described above in the prior art
It is necessary to.
Summary of the invention
It is an object of the present invention to design one kind towards copper ore floatation in view of the above-mentioned drawbacks of the prior art, providing
The flexible measurement method neural network based and system of machine, to solve the above technical problems.
To achieve the above object, the present invention provides following technical scheme:
A kind of flexible measurement method neural network based towards copper ore floatation machine, which comprises the following steps:
S1:NNG-RBF Learning Algorithm step seeks Basis Function Center c based on K- means clustering method, specifically
Steps are as follows:
Netinit: h sample is randomly selected as cluster centre ci(i=1,2 ..., h)；
The training sample set of input is grouped by Nearest Neighbor Method: according to xpWith center ciBetween Euclidean distance by xp
It is assigned to each cluster set v of input samplepIn (p=1,2 ... P)；
xpFor p-th of input sample, p=1,2,3, P；P is total sample number, that is, nerve net
The training sample of network；
It readjusts cluster centre: calculating each cluster set vpThe average value of middle training sample, i.e., new cluster centre
ciIf new cluster centre is no longer changed, obtained ciAs in the final basic function of NNG-RBF neural network
Otherwise the heart carries out for the training sample set of input being grouped by Nearest Neighbor Method again, the center for carrying out next round solves；
Solve variances sigmai, the basic function of the function of the NNG-RBF neural network is Gaussian function, variances sigmaiIt can ask as follows
Solution:
In formula, cmaxIt is the maximum distance between selected center；
The weight between hidden layer and output layer is calculated, the connection weight of hidden layer to neuron between output layer can be used
Least square method is directly calculated:
Further, by the RBF neural, in conjunction with Nonnegative garrote (NNG) algorithm, by increasing newly
Factor beta (β1, β2..., βp) calculation formula is reformulated, utilize newly-increased factor beta (β1, β2..., βp) compression input variable:
This algorithm can solve the problem of quadratic nonlinearity constraint, can roll over cross validation with V for the selection of optimal s,
Data set L=X, Y is divided into V subset, in constraint conditionUnder to { βiMinimization:
And it willAs new input variable weight coefficient；βiValue depend on s, s is the NNG being additionally added
Algorithm parameter；βiSize reflect corresponding auxiliary variable to the importance of prediction model.For example, if βi=0, then explanation pair
The variable x answerediThere is no any influence to objective function, thus xiIt will be removed.If βi=1 unconverted to dependent variable
It remains.If 0 < βi< 1 then illustrates that corresponding variation coefficient is compressed, that is, effect of the variable for prediction model
It is compressed by.By reducing s, make more { βiBecome zero, to achieve the purpose that variable compression, this method is exactly NNG-
RBF algorithm.
S2: the selection of optimal N NG algorithm parameter and error prediction
The purpose of variables choice seeks to find the auxiliary variable for being affected to y, is likely to occur by auxiliary variable to y
Situation is predicted.Modeling accuracy evaluation index: mean square error (MSE) assessment models precision of prediction is used.Mathematical formulae indicates
Are as follows:
V-fold cross-validation method is data set to be equally divided into V parts first, takes out a number from V parts of data concentrations every time
According to collection as verifying collection, remaining V-1 parts of data set repeats V times as training set, finally averagely V times result conduct
The estimation of last extensive error.The value of usual V can obtain preferably when being 5 to 10 as a result, when V value is too big, and variance is just
It can increase with it；It will lead to the increase of prediction error due to participating in the sample data reduction of training when V value is smaller；Formula is such as
Under:
Optimal NNG algorithm parameter s is selected by the formula, and s value is substituted into formula (4) and is solved, and obtains system optimal
Compressed coefficient β*。
The modeling of S3:NNG-RBF algorithm
After being handled by v-fold cross-validation method data, obtained s is the parameter that training obtains, s band
Enter into formula, calculates βiValue.βiSize reflect corresponding auxiliary variable to the importance of prediction model, pass through βi's
Value rejecting does not have the variable of any influence to prediction model, chooses optimal variable, to play the purpose to variation coefficient compression.
Input variable is brought into modeling and forecasting in trained neural network.
A kind of hard measurement system neural network based towards copper ore floatation machine, which is characterized in that it includes power supply mould
Block, main control module and communication module, the main control module connection power module and communication module, the communication mould
Block is also connected with collection in worksite module and upper computer module；
The mathematical modeling module of above-mentioned NNG-RBF algorithm is integrated in the main control module.
The collection in worksite module includes:
SWINGWIRL II capacitor type vortex street flow sensor；SWINGWIRL II capacitor type vortex street flow sensor is to adopt
Differential switched capacitor (DSC) is used as detecting element to sense a kind of equipment of the vortex frequency of swirl generating body generation.Its is excellent
Point is that operating temperature range is very wide, and from -200 DEG C~+400 DEG C, resistance to shock is especially good.Also having the following characteristics that simultaneously can not
Moving part, up to 40:1, the pressure loss is small, and accuracy of measurement is higher etc. for measurement range.It can be used for measuring gas in closed conduct, steam
Vapour and fluid flow.Inside nominal diameter used in present apparatus SWINGWIRL II capacitor type vortex street flow sensor is 300mm, empty
Gas measurement range is 1655m3/h-19330m3/h.Setting value of the flow sensor for the air aeration quantity in measuring device
Deng the detection circuit based on SWINGWIRL II capacitor type vortex street flow sensor is as shown in Figure 2.
CLHGM-2 type spoke type weighing sensor；CLHGM-2 type spoke type weighing sensor is to utilize resistance-strain principle
It constitutes, elastomer uses more advanced spoke structure form.Resistance strain plate is attached on the neutral surface of spoke, composition
The measurement circuit of electric bridge.Under normal conditions, electric bridge is in equilibrium state, bridge without output when sensor is by external force,
Spoke generates corresponding deformation, and resistance strain gage resistance value changes, and makes bridge disequilibrium.It is acted in the external world for bridge voltage
Under, electric bridge exports unbalance voltage signal.The signal magnitude is directly proportional to external force.CLHGM-2 type spoke type weighing sensor is defeated
400 Ω of impedance out, 460 Ω of input impedance, -20 DEG C of the temperature range that can be worked~80 DEG C are made in various industrial and mining enterprises' systems
The measurement of power is analyzed.The spoke type weighing sensor in measuring device for, to mine total amount, lime total amount of adding etc., being based on
CLHGM-2 type spoke type weighing sensor detection circuit is as shown in Figure 3.
TCD128C-CCD imaging sensor；TCD128C-CCD imaging sensor is that one kind can be carried out photoelectric conversion storage letter
The device of breath and transitional information charge function.PN junction photodiode and CCD (charge-coupled device) constitute the one of several pixels
First photodiode arrangement, object form real image on this array by optical lens.Each light-sensitive element (pixel) is presented
The weak current of varying strength can get vision signal by handling by scanning circuit captured image signal.TCD128C-CCD
Imaging sensor advantage is self-scanning, highly sensitive, low noise, the long-life, low-power consumption, highly reliable.Its pixel dimension is small, geometry essence
Degree is high, configures optical system appropriate, can get very high spatial resolution, easy to use and flexible is adaptable, output signal
It is easy to digitized processing, is easy to connect composition automatic measurement and control with computer.Valid pixel number 1728, effectively reading length
210mm.Area, the area of middle bubble etc. of the CD128C-CCD imaging sensor for bulla in measuring device, are based on
TCD128C-CCD imaging sensor amplification imaging measuring circuit is as shown in Figure 4.
Power module can play the role of pressure stabilizing, protection chip while powering to measuring device.Main control module
Data are received, input the model built up then to export hard measurement result.Communication module is the data of reception collection in worksite, and
Hard measurement result is sent to host computer.
The main control module is the embedded system based on STM32F103, which can work in -40~105 DEG C
Temperature range can adapt to severe industrial production environment.MAX232 chip is used for the level translation of serial port, realizes controller
Communication between communication interface.STM32F103 main control chip is as shown in Figure 5.Primary control program process such as Fig. 6 institute of main control module
Show.
The beneficial effects of the present invention are the pH value measured is difficult in copper ore floatation machine device, by device 18
A screening for surveying input variable, then data are handled, then pH value is predicted using the variable modeling screened.
Selection to auxiliary variable is the Variable Selection based on NNG-RBF, then using NNG-RBF algorithm modeling and
The on line emendation of soft measuring instrument guarantees the precision of prediction of pH value；This method is rapid with response, investment is low, maintenance
Maintain the advantages that simple.
The selection of auxiliary variable will be primarily determined by Analysis on Mechanism to copper ore floatation machine device and process flow
The related auxiliary variable for influencing leading variable, including types of variables, variables number and the selection of monitoring point.It was verified that being based on
The variable selection algorithm of NNG-RBF can choose out optimal auxiliary variable, and then improves our precision of prediction and reduce meter
It is counted as this.
In addition, design principle of the present invention is reliable, structure is simple, has very extensive application prospect.
It can be seen that compared with prior art, the present invention have substantive distinguishing features outstanding and it is significant ground it is progressive, implementation
Beneficial effect be also obvious.
Detailed description of the invention
Table 1 is that 18 of copper ore floatation machine device can survey input variable.
Fig. 1 is copper ore floatation machine device mineralization process block plan.
Fig. 2 is the detection circuit based on SWINGWIRL II capacitor type vortex street flow sensor.
Fig. 3 is based on CLHGM-2 type spoke type weighing sensor detection circuit.
Fig. 4 is to amplify imaging measuring circuit based on TCD128C-CCD imaging sensor.
Fig. 5 is STM32F103 main control chip.
Fig. 6 is the primary control program process of main control module.
Specific embodiment
The present invention will be described in detail with reference to the accompanying drawing and by specific embodiment, and following embodiment is to the present invention
Explanation, and the invention is not limited to following implementation.
Embodiment 1:
Table 1 is that 18 of copper ore floatation machine device can survey input variable.
1 copper ore floatation machine device of table surveys input variable table
A kind of flexible measurement method neural network based towards copper ore floatation machine provided by the invention, which is characterized in that
The following steps are included:
S1:NNG-RBF Learning Algorithm step seeks Basis Function Center c based on K- means clustering method, specifically
Steps are as follows:
Netinit: h sample is randomly selected as cluster centre ci(i=1,2 ..., h)；
The training sample set of input is grouped by Nearest Neighbor Method: according to xpWith center ciBetween Euclidean distance by xp
It is assigned to each cluster set v of input samplepIn (p=1,2 ... P)；
It readjusts cluster centre: calculating each cluster set vpThe average value of middle training sample, i.e., new cluster centre
ciIf new cluster centre is no longer changed, obtained ciAs in the final basic function of NNG-RBF neural network
Otherwise the heart carries out for the training sample set of input being grouped by Nearest Neighbor Method again, the center for carrying out next round solves；
Solve variances sigmai, the basic function of the function of the NNG-RBF neural network is Gaussian function, variances sigmaiIt can ask as follows
Solution:
In formula, cmaxIt is the maximum distance between selected center；
The weight between hidden layer and output layer is calculated, the connection weight of hidden layer to neuron between output layer can be used
Least square method is directly calculated:
Further, by the RBF neural, in conjunction with Nonnegative garrote (NNG) algorithm, by increasing newly
Factor beta (β1, β2..., βp) calculation formula is reformulated, utilize newly-increased factor beta (β1, β2..., βp) compression input variable:
This algorithm can solve the problem of quadratic nonlinearity constraint, can roll over cross validation with V for the selection of optimal s,
Data set L=X, Y is divided into V subset, in constraint conditionUnder to { βiMinimization:
And it willAs new input variable weight coefficient；βiValue depend on s, s is the NNG being additionally added
Algorithm parameter；βiSize reflect corresponding auxiliary variable to the importance of prediction model.For example, if βi=0, then explanation pair
The variable x answerediThere is no any influence to objective function, thus xiIt will be removed.If βi=1 unconverted to dependent variable
It remains.If 0 < βi< 1 then illustrates that corresponding variation coefficient is compressed, that is, effect of the variable for prediction model
It is compressed by.By reducing s, make more { βiBecome zero, to achieve the purpose that variable compression, this method is exactly NNG-
RBF algorithm.
S2: the selection of optimal N NG algorithm parameter and error prediction
The purpose of variables choice seeks to find the auxiliary variable for being affected to y, is likely to occur by auxiliary variable to y
Situation is predicted.Modeling accuracy evaluation index: mean square error (MSE) assessment models precision of prediction is used.Mathematical formulae indicates
Are as follows:
V-fold cross-validation method is data set to be equally divided into V parts first, takes out a number from V parts of data concentrations every time
According to collection as verifying collection, remaining V-1 parts of data set repeats V times as training set, finally averagely V times result conduct
The estimation of last extensive error.The value of usual V can obtain preferably when being 5 to 10 as a result, when V value is too big, and variance is just
It can increase with it；It will lead to the increase of prediction error due to participating in the sample data reduction of training when V value is smaller；Formula is such as
Under:
Optimal NNG algorithm parameter s is selected by the formula, and s value is substituted into formula (4) and is solved, and obtains system optimal
Compressed coefficient β*。
The modeling of S3:NNG-RBF algorithm
After being handled by v-fold cross-validation method data, obtained s is the parameter that training obtains, s band
Enter into formula, calculates βiValue.βiSize reflect corresponding auxiliary variable to the importance of prediction model, pass through βi's
Value rejecting does not have the variable of any influence to prediction model, chooses optimal variable, to play the purpose to variation coefficient compression.
Input variable is brought into modeling and forecasting in trained neural network.
Embodiment 2:
As shown in figures 2-6, a kind of hard measurement system neural network based towards copper ore floatation machine provided by the embodiment,
It is characterized in that, it includes power module, main control module and communication module, the main control module connection power module
And communication module, the communication module are also connected with collection in worksite module and upper computer module；
The mathematical modeling module of above-mentioned NNG-RBF algorithm is integrated in the main control module.
The collection in worksite module includes:
SWINGWIRL II capacitor type vortex street flow sensor；SWINGWIRL II capacitor type vortex street flow sensor is to adopt
Differential switched capacitor (DSC) is used as detecting element to sense a kind of equipment of the vortex frequency of swirl generating body generation.Its is excellent
Point is that operating temperature range is very wide, and from -200 DEG C~+400 DEG C, resistance to shock is especially good.Also having the following characteristics that simultaneously can not
Moving part, up to 40: 1, the pressure loss is small, and accuracy of measurement is higher etc. for measurement range.It can be used for measuring gas in closed conduct, steam
Vapour and fluid flow.Inside nominal diameter used in present apparatus SWINGWIRL II capacitor type vortex street flow sensor is 300mm, empty
Gas measurement range is 1655m3/h-19330m3/h.Setting value of the flow sensor for the air aeration quantity in measuring device
Deng the detection circuit based on SWINGWIRL II capacitor type vortex street flow sensor is as shown in Figure 2.
CLHGM-2 type spoke type weighing sensor；CLHGM-2 type spoke type weighing sensor is to utilize resistance-strain principle
It constitutes, elastomer uses more advanced spoke structure form.Resistance strain plate is attached on the neutral surface of spoke, composition
The measurement circuit of electric bridge.Under normal conditions, electric bridge is in equilibrium state, bridge without output when sensor is by external force,
Spoke generates corresponding deformation, and resistance strain gage resistance value changes, and makes bridge disequilibrium.It is acted in the external world for bridge voltage
Under, electric bridge exports unbalance voltage signal.The signal magnitude is directly proportional to external force.CLHGM-2 type spoke type weighing sensor is defeated
400 Ω of impedance out, 460 Ω of input impedance, -20 DEG C of the temperature range that can be worked~80 DEG C are made in various industrial and mining enterprises' systems
The measurement of power is analyzed.The spoke type weighing sensor in measuring device for, to mine total amount, lime total amount of adding etc., being based on
CLHGM-2 type spoke type weighing sensor detection circuit is as shown in Figure 3.
TCD128C-CCD imaging sensor；TCD128C-CCD imaging sensor is that one kind can be carried out photoelectric conversion storage letter
The device of breath and transitional information charge function.PN junction photodiode and CCD (charge-coupled device) constitute the one of several pixels
First photodiode arrangement, object form real image on this array by optical lens.Each light-sensitive element (pixel) is presented
The weak current of varying strength can get vision signal by handling by scanning circuit captured image signal.TCD128C-CCD
Imaging sensor advantage is self-scanning, highly sensitive, low noise, the long-life, low-power consumption, highly reliable.Its pixel dimension is small, geometry essence
Degree is high, configures optical system appropriate, can get very high spatial resolution, easy to use and flexible is adaptable, output signal
It is easy to digitized processing, is easy to connect composition automatic measurement and control with computer.Valid pixel number 1728, effectively reading length
210mm.Area, the area of middle bubble etc. of the TCD128C-CCD imaging sensor for bulla in measuring device, are based on
TCD128C-CCD imaging sensor amplification imaging measuring circuit is as shown in Figure 4.
Power module can play the role of pressure stabilizing, protection chip while powering to measuring device.Main control module
Data are received, input the model built up then to export hard measurement result.Communication module is the data of reception collection in worksite, and
Hard measurement result is sent to host computer.
The main control module is the embedded system based on STM32F103, which can work in -40~105 DEG C
Temperature range can adapt to severe industrial production environment.MAX232 chip is used for the level translation of serial port, realizes controller
Communication between communication interface.STM32F103 main control chip is as shown in Figure 5.Primary control program process such as Fig. 6 institute of main control module
Show.
Disclosed above is only the preferred embodiment of the present invention, but the present invention is not limited to this, any this field
What technical staff can think does not have creative variation, and without departing from the principles of the present invention made by several improvement and
Retouching, should all be within the scope of the present invention.
Claims (7)
1. a kind of flexible measurement method neural network based towards copper ore floatation machine, which comprises the following steps:
S1:NNG-RBF Learning Algorithm step seeks Basis Function Center c based on K- means clustering method；
S2: the selection of optimal compression parameter and error prediction；
The modeling of S3:NNG-RBF algorithm.
2. a kind of flexible measurement method neural network based towards copper ore floatation machine according to claim 1, feature
Be, the step S1 specifically includes the following steps:
Netinit: h sample is randomly selected as cluster centre ci(i=1,2 ..., h)；
The training sample set of input is grouped by Nearest Neighbor Method: according to xpWith center ciBetween Euclidean distance by xpDistribution
To each cluster set v of input samplepIn (p=1,2 ... P)；
It readjusts cluster centre: calculating each cluster set vpThe average value of middle training sample, i.e., new cluster centre ci, such as
The new cluster centre of fruit is no longer changed, then obtained ciThe as final Basis Function Center of NNG-RBF neural network, it is no
It then carries out for the training sample set of input being grouped by Nearest Neighbor Method again, the center for carrying out next round solves；
Solve variances sigmai, the basic function of the function of the NNG-RBF neural network is Gaussian function, variances sigmaiIt can solve as follows:
In formula, cmaxIt is the maximum distance between selected center；
The weight between hidden layer and output layer is calculated, the connection weight of hidden layer to neuron between output layer can use minimum
Square law is directly calculated:
By the RBF neural, in conjunction with Nonnegative garrote (NNG) algorithm, by increasing new factor beta (β1,
β2..., βp) calculation formula is reformulated, utilize newly-increased factor beta (β1, β2..., βp) compression input variable:
Cross validation is rolled over for the selection V of optimal s, data set L=X, Y is divided into V subset, in constraint conditionUnder to { βiMinimization:
And it willAs new input variable weight coefficient；βiValue depend on s, s is the NNG algorithm being additionally added
Parameter.
3. a kind of flexible measurement method neural network based towards copper ore floatation machine according to claim 1 or 2, special
Sign is, the step S2 specifically includes the following steps:
The purpose of variables choice seeks to find the auxiliary variable for being affected to y, the case where being likely to occur by auxiliary variable to y
It is predicted；Modeling accuracy evaluation index: mean square error (MSE) assessment models precision of prediction is used；Mathematical formulae indicates are as follows:
V-fold cross-validation method is data set to be equally divided into V parts first, takes out a data set from V parts of data concentrations every time
Collect as verifying, remaining V-1 parts of data set repeats V times as training set, and finally averagely V times result is as last
The estimation of extensive error；It will lead to the increase of prediction error due to participating in the sample data reduction of training when V value is smaller；It is public
Formula is as follows:
Optimal NNG algorithm parameter s is selected by the formula, and s value is substituted into formula (4) and is solved, and obtains system optimal compression
Factor beta*。
4. a kind of flexible measurement method neural network based towards copper ore floatation machine according to claim 3, feature
Be, the step S3 specifically includes the following steps:
After being handled by v-fold cross-validation method data, obtained s is the parameter that training obtains, and passes through ciValue
Rejecting does not have the variable of any influence to prediction model, chooses optimal variable, input variable is brought into trained nerve
Modeling and forecasting in network.
5. a kind of hard measurement system neural network based towards copper ore floatation machine, which is characterized in that it include power module,
Main control module and communication module, the main control module connection power module and communication module, the communication module
It is also connected with collection in worksite module and upper computer module；
The mathematical modeling module of above-mentioned NNG-RBF algorithm is integrated in the main control module.
6. a kind of hard measurement system neural network based towards copper ore floatation machine according to claim 5, feature
It is, the collection in worksite module includes:
SWINGWIRL II capacitor type vortex street flow sensor；
CLHGM-2 type spoke type weighing sensor；
TCD128C-CCD imaging sensor.
7. a kind of hard measurement system neural network based towards copper ore floatation machine according to claim 6, feature
It is, the main control module is the embedded system based on STM32F103.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810960867.XA CN109002862B (en) | 2018-08-22 | 2018-08-22 | Neural network-based soft measurement method and system for copper ore flotation machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810960867.XA CN109002862B (en) | 2018-08-22 | 2018-08-22 | Neural network-based soft measurement method and system for copper ore flotation machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109002862A true CN109002862A (en) | 2018-12-14 |
| CN109002862B CN109002862B (en) | 2021-05-04 |
Family
ID=64594121
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810960867.XA Active CN109002862B (en) | 2018-08-22 | 2018-08-22 | Neural network-based soft measurement method and system for copper ore flotation machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109002862B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110672792A (en) * | 2019-10-09 | 2020-01-10 | 中南大学 | Soft measurement method and system for pH value in neutral leaching process of zinc hydrometallurgy |
| CN111141341A (en) * | 2019-12-31 | 2020-05-12 | 华东理工大学 | Compensation method of turbine flowmeter, system and storage medium thereof |
| CN111482280A (en) * | 2020-04-22 | 2020-08-04 | 齐鲁工业大学 | Intelligent soft measurement method and system for copper ore flotation based on wireless sensor network |
| CN111983140A (en) * | 2020-07-21 | 2020-11-24 | 齐鲁工业大学 | Carbon monoxide measuring system and method for dry quenching production |
| CN114111942A (en) * | 2021-11-25 | 2022-03-01 | 宁夏隆基宁光仪表股份有限公司 | Nonmagnetic intelligent water meter metering method and system based on nonmagnetic sampling and metering equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103344583A (en) * | 2013-07-09 | 2013-10-09 | 华东交通大学 | Praseodymium-neodymium (Pr/Nd) component content detection system and method based on machine vision |
-
2018
- 2018-08-22 CN CN201810960867.XA patent/CN109002862B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103344583A (en) * | 2013-07-09 | 2013-10-09 | 华东交通大学 | Praseodymium-neodymium (Pr/Nd) component content detection system and method based on machine vision |
Non-Patent Citations (2)
| Title |
|---|
| KAI SUN等: "Development of a variable selection method for soft sensor using artificial neural network and nonnegative garrote", 《JOURNAL OF PROCESS CONTROL》 * |
| 张勇等: "基于PCA-RBF神经网络的浮选过程软测量建模", 《南京航空航天大学学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110672792A (en) * | 2019-10-09 | 2020-01-10 | 中南大学 | Soft measurement method and system for pH value in neutral leaching process of zinc hydrometallurgy |
| CN110672792B (en) * | 2019-10-09 | 2020-08-28 | 中南大学 | Soft measurement method and system for pH value in neutral leaching process of zinc hydrometallurgy |
| CN111141341A (en) * | 2019-12-31 | 2020-05-12 | 华东理工大学 | Compensation method of turbine flowmeter, system and storage medium thereof |
| CN111482280A (en) * | 2020-04-22 | 2020-08-04 | 齐鲁工业大学 | Intelligent soft measurement method and system for copper ore flotation based on wireless sensor network |
| CN111983140A (en) * | 2020-07-21 | 2020-11-24 | 齐鲁工业大学 | Carbon monoxide measuring system and method for dry quenching production |
| CN111983140B (en) * | 2020-07-21 | 2022-05-10 | 齐鲁工业大学 | Carbon monoxide measuring system and method for dry quenching production |
| CN114111942A (en) * | 2021-11-25 | 2022-03-01 | 宁夏隆基宁光仪表股份有限公司 | Nonmagnetic intelligent water meter metering method and system based on nonmagnetic sampling and metering equipment |
| CN114111942B (en) * | 2021-11-25 | 2024-07-16 | 宁夏隆基宁光仪表股份有限公司 | Non-magnetic intelligent water meter metering method, system and metering equipment based on non-magnetic sampling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109002862B (en) | 2021-05-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109002862A (en) | A kind of flexible measurement method neural network based and system towards copper ore floatation machine | |
| CN109344894B (en) | Garbage classification and identification method and device based on multi-sensor fusion and deep learning | |
| CN111340791A (en) | Photovoltaic module unsupervised defect detection method based on GAN improved algorithm | |
| CN109472822A (en) | Dimension of object measurement method based on depth image processing | |
| CN105389814A (en) | Air bubble detection method for air tightness test | |
| CN106643965B (en) | Method for accurately identifying liquid level by utilizing template matching | |
| CN103344530B (en) | Soft measurement method for granularity of cement raw material grinded by ball mill | |
| CN115169733B (en) | Deep learning-based method for predicting resuspension amount of internal solitary waves on deep sea sediment | |
| CN101464172A (en) | Soft measuring method for power boiler breeze concentration mass flow | |
| CN109993154A (en) | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula | |
| CN110362044B (en) | Limestone addition amount prediction control system and method for copper ore flotation device | |
| CN111046761A (en) | Belt load distribution detection system and method based on multi-sensing information fusion | |
| CN104318241A (en) | Local density spectral clustering similarity measurement algorithm based on Self-tuning | |
| CN103413375A (en) | Discrimination system and method of old and new paper currency based on image statistical features | |
| CN116523913B (en) | Intelligent detection method for quality of screw rod | |
| CN108647722A (en) | A kind of zinc ore grade flexible measurement method of Kernel-based methods size characteristic | |
| CN103927516A (en) | Seawater pearl authentication system based on digital image processing | |
| CN103745238A (en) | Pantograph identification method based on AdaBoost and active shape model | |
| Xu et al. | Multi-model soft measurement method of the froth layer thickness based on visual features | |
| CN111402249A (en) | Image evolution analysis method based on deep learning | |
| CN108931621A (en) | A kind of zinc ore grade flexible measurement method of Kernel-based methods textural characteristics | |
| CN113828531B (en) | Multi-channel coal gangue identification method based on gray-thickness | |
| CN110059573A (en) | Wild ginseng based on image recognition is classified calibration method | |
| CN112819087B (en) | Method for detecting abnormality of BOD sensor of outlet water based on modularized neural network | |
| CN110120035A (en) | A kind of tire X-ray defect detection method differentiating defect grade |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220713 Address after: Room 409, Yimeng cloud Valley, No.19 Chang'an Road, Zhimadun street, economic development zone, Linyi City, Shandong Province 276000 Patentee after: SHANDONG TONGQI INTELLIGENT TECHNOLOGY CO.,LTD. Address before: 250353 University Road, Changqing District, Ji'nan, Shandong Province, No. 3501 Patentee before: Qilu University of Technology |