CN110044767A - Based on LSSVM lime stone slurry density flexible measurement method - Google Patents
Based on LSSVM lime stone slurry density flexible measurement method Download PDFInfo
- Publication number
- CN110044767A CN110044767A CN201910331808.0A CN201910331808A CN110044767A CN 110044767 A CN110044767 A CN 110044767A CN 201910331808 A CN201910331808 A CN 201910331808A CN 110044767 A CN110044767 A CN 110044767A
- Authority
- CN
- China
- Prior art keywords
- lssvm
- lime stone
- stone slurry
- data
- measurement method
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N9/00—Investigating density or specific gravity of materials; Analysing materials by determining density or specific gravity
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Liquid Carbonaceous Fuels (AREA)
Abstract
The invention discloses one kind be based on LSSVM lime stone slurry density flexible measurement method, this method comprises: step 1, choose limestone slurry liquid preparing system in 18 monitoring points data;Step 2 determines 9 auxiliary variables by Analysis on Mechanism from 18 data of monitoring point that the step 1 is chosen;Step 3 is standardized 9 auxiliary variables in the step 2 and PCA dimensionality reduction;Contribution rate to 88% first three number of principal components finally enters variable according to as LSSVM model in step 4, the selection step 3;Step 5 calculates three number of principal components in the step 4 according to input LSSVM model, obtains lime stone slurry density.The invention has the benefit that on the one hand avoiding the susceptible to plugging problem of densitometer in conventional limestone serosity density measurement;It on the other hand can quick, economy, the density of accurately measure lime stone slurry.
Description
Technical field
The present invention relates to lime stone slurry density measure technical fields, are based on LSSVM lime stone in particular to one kind
Serum density flexible measurement method.
Background technique
With the attention of environmental issue, the standard limit of smog release of power plant is also paid attention to further, lime stone slurry is most of
The good absorption agent of factory smoke discharge.Therefore, lime stone slurry is prepared into a part important for factory's desulphurization system,
The density of lime stone slurry decides the absorptivity to flue gas.Most factories uses densimeter measurement limestone slurry now
Liquid density, since lime stone slurry density is larger, is easy to cause densitometer although this measurement method is more intuitive simple
Blocking, so that measurement is not prompt enough accurate.The side of lime stone slurry density specially can be accurately and fast measured currently without one
Method.
Summary of the invention
To solve the above problems, the purpose of the present invention is to provide one kind to be based on LSSVM lime stone slurry density hard measurement
Method realizes purpose quick to lime stone slurry density, economic and precise measurement.
The present invention provides one kind to be based on LSSVM lime stone slurry density flexible measurement method, this method comprises:
Step 1, the data for choosing 18 monitoring points in limestone slurry liquid preparing system;
Step 2 determines 9 auxiliary variables by Analysis on Mechanism from 18 data of monitoring point that the step 1 is chosen;
Step 3 is standardized 9 auxiliary variables in the step 2 and PCA dimensionality reduction;
Step 4 chooses first three number of principal components of contribution rate in the step 3 to 88% according to as LSSVM model
Finally enter variable;
Step 5 calculates three number of principal components in the step 4 according to input LSSVM model, obtains limestone slurry
Liquid density.
As a further improvement of the present invention, during choosing 18 data of monitoring point in the step 1, when lacking
Data then pass through interpolation method completion missing data, rejecting abnormalities data and unify timing to all data later.
As a further improvement of the present invention, the foundation that 9 auxiliary variables are determined in the step 2 is by similarity degree
Amount screens 18 data of monitoring point.
As a further improvement of the present invention, 9 auxiliary variables determined in the step 2 are as follows: feed frequency is instantaneously given
Doses, recycling tank liquid level, grinding water flow, dilution water flow, dilution water valve opening, recirculation pump electric current, recycling tank
Blender electric current and ball mill electric current.
As a further improvement of the present invention, the linear independence between of three number of principal components in the step 4.
As a further improvement of the present invention, the LSSVM model is programmed by MATLAB and is established.
As a further improvement of the present invention, the LSSVM model is the measurement model of multiple input single output.
The invention has the benefit that it is susceptible to plugging on the one hand to avoid densitometer in conventional limestone serosity density measurement
Problem;It on the other hand can quick, economy, accurately measure lime stone slurry density.
Detailed description of the invention
Fig. 1 shows for a kind of process based on LSSVM lime stone slurry density flexible measurement method described in the embodiment of the present invention
It is intended to;
Fig. 2 is that one kind described in the embodiment of the present invention is based on LSSVM mould in LSSVM lime stone slurry density flexible measurement method
Type structural schematic diagram.
Specific embodiment
The present invention is described in further detail below by specific embodiment and in conjunction with attached drawing.
As shown in Figure 1, be that one kind is based on LSSVM lime stone slurry density flexible measurement method described in the embodiment of the present invention,
This method comprises:
Step 1, the data for choosing 18 monitoring points in limestone slurry liquid preparing system;
Step 2 determines 9 auxiliary variables by Analysis on Mechanism from 18 data of monitoring point that the step 1 is chosen;
Step 3 is standardized 9 auxiliary variables in the step 2 and PCA dimensionality reduction;
Step 4 chooses first three number of principal components of contribution rate in the step 3 to 88% according to as LSSVM model
Finally enter variable;
Step 5 calculates three number of principal components in the step 4 according to input LSSVM model, obtains limestone slurry
Liquid density.
The principle that the present invention measures lime stone slurry density is: being set with the hardware in computer linguistic substitution actual production
It is standby, the auxiliary variable of LSSVM model is served as using the variable easily obtained, and mould is established by the auxiliary variable that these are easily obtained
Type, so that indirect predictions leading variable, realizes the ability of the variable measurement to hardly possible measurement.The present embodiment is first to obtain from DSC
The data of 18 monitoring points in limestone slurry liquid preparing system, by the basic principle and shadow that combine lime stone slurry density measure
The factor for ringing recycling tank serosity density measurement determines 9 data as auxiliary variable, to 9 from 18 data of monitoring point
Auxiliary variable is standardized, be used as LSSVM model after PCA dimensionality reduction finally enter variable, reach simplified data, subtract
The purpose of few redundancy.It is online to LSSVM that limestone slurry liquid preparing system data set is divided into training dataset and test data set
Prediction model is trained and emulates, and the density of lime stone slurry is calculated eventually by LSSVM model.
Further, during choosing 18 data of monitoring point in the step 1, then pass through interpolation when there is missing data
Method completion missing data rejecting abnormalities data and unifies timing to all data later.
Further, the foundation that 9 auxiliary variables are determined in the step 2 is by measuring similarity to 18 monitoring points
Data are screened.18 data of monitoring point are screened by measuring similarity, it is maximum while reducing data group number
Guarantee sample data covers operating condition quantity.
Further, 9 auxiliary variables determined in the step 2 are as follows: feed frequency, instantaneous feeding coal, recycling tank
Liquid level, grinding water flow, dilution water flow, dilution water valve opening, recirculation pump electric current, recycling tank blender electric current and ball
Grinding machine electric current.
In wet process slurrying, limestone block is sent after weighed belt feeder after the submitting of lime stone feed bin to wet type ball
Grinding machine, and the technique grinding water that certain proper ratio is added is made lime stone slurry and enters slurry circulation case, lime stone slurry is logical
It crosses slurries recirculation pump and enters lime stone cyclone, by water flow cyclone, underproof lime stone slurry is delivered to wet type
Ball mill is re-grind, and qualified lime stone slurry is sent to limestone slurry liquid case storage, is needed to use lime stone according to absorption tower
Slurries are pumped to absorption tower.The data of 18 monitoring points are chosen in the process and by interpolation method by missing data completion, pick
Except timing unified after exceptional value;Sample data is screened by measuring similarity again, while reducing data group number most
Big guarantee sample covers operating condition quantity, and 9 auxiliary variables have been determined by Analysis on Mechanism, is respectively as follows: feed frequency, instantaneous
Feeding coal, recycling tank liquid level, grinding water flow, dilution water flow, dilution water valve opening, recirculation pump electric current, recycling
Box mixer electric current, wet ball mill electric current.
Further, the linear independence between of three number of principal components in the step 4.It can be less in lime stone slurry
Expense during density measure over time and space.
Further, the LSSVM model is programmed by MATLAB and is established.
Further, the LSSVM model is the measurement model of multiple input single output.
It is theoretical based on LSSVM model are as follows:
There are training setsWherein xk∈Rd, yk∈Rd, d is the quantity of auxiliary variable.The base of support vector machines
Present principles are will to input airborne R by nonlinear mapping function φ (°)dIn input sample be mapped to feature space φ (x)=
(φ1(x),φ2(x),…φn(x)).Here is the estimation pattern function of an introducing error:
Wherein, ω ∈ Rdn, b ∈ R, e ∈ R, k=1,2 ... n.
According to result principle of minimization risk, and least square method is introduced, formula (1) must satisfy:
Wherein, e is error, and γ is regularization parameter.
Using Lagrangian method, formula (2) be can be rewritten as:
Wherein, αk(k=1,2 ..., n) it is Lagrange factor.
Searching meet formula (3) α and b be LSSVM model target.Define kernel function K (Xk,Xi), which should be one
A any symmetric function for meeting Mercer constraint condition.Then soft-sensing model are as follows:
Wherein, k=1,2 ..., n.
The selection of kernel function has certain influence to the regression analysis of support vector machines, but how to select kernel function currently without
Mature theory.Usually used kernel function has radial basis function (RBF), polynomial function, S function and linear function.
Partial derivative by 4 parameters in calculating formula (3) is 0, after eliminating parameter ω and e, obtains following linear equation
Group:
Wherein,
Y=[y1,y2,…yn];
1=[1,1 ..., 1];
α=[α1,α2,…,αn];
Ω=φ (xk)Tφ(xl), k=1,2 ..., n
In formula (5), the item number of model is that training sample sum adds 1, if number of training is more, scale of model is huge,
Influence the application of model.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (7)
1. one kind is based on LSSVM lime stone slurry density flexible measurement method characterized by comprising
Step 1, the data for choosing 18 monitoring points in limestone slurry liquid preparing system;
Step 2 determines 9 auxiliary variables by Analysis on Mechanism from 18 data of monitoring point that the step 1 is chosen;
Step 3 is standardized 9 auxiliary variables in the step 2 and PCA dimensionality reduction;
Step 4 chooses first three number of principal components of contribution rate in the step 3 to 88% according to as the final of LSSVM model
Input variable;
Step 5 calculates three number of principal components in the step 4 according to input LSSVM model, show that lime stone slurry is close
Degree.
2. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that the step
During choosing 18 data of monitoring point in rapid 1, then passes through interpolation method completion missing data when there is missing data, reject later
Abnormal data simultaneously unifies timing to all data.
3. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that the step
The foundation that 9 auxiliary variables are determined in rapid 2 is to be screened by measuring similarity to 18 data of monitoring point.
4. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that the step
9 auxiliary variables determined in rapid 2 are as follows: feed frequency, instantaneous feeding coal, recycling tank liquid level, grinding water flow, dilution water flow
Amount, dilution water valve opening, recirculation pump electric current, recycling tank blender electric current and ball mill electric current.
5. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that the step
Linear independence between three number of principal components evidence in rapid 4.
6. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that described
LSSVM model is programmed by MATLAB and is established.
7. according to claim 1 be based on LSSVM lime stone slurry density flexible measurement method, which is characterized in that described
LSSVM model is the measurement model of multiple input single output.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910331808.0A CN110044767A (en) | 2019-04-23 | 2019-04-23 | Based on LSSVM lime stone slurry density flexible measurement method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910331808.0A CN110044767A (en) | 2019-04-23 | 2019-04-23 | Based on LSSVM lime stone slurry density flexible measurement method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110044767A true CN110044767A (en) | 2019-07-23 |
Family
ID=67278873
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910331808.0A Pending CN110044767A (en) | 2019-04-23 | 2019-04-23 | Based on LSSVM lime stone slurry density flexible measurement method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110044767A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111238997A (en) * | 2020-02-12 | 2020-06-05 | 江南大学 | On-line measurement method for feed density in crude oil desalting and dewatering process |
CN111766179A (en) * | 2020-07-08 | 2020-10-13 | 大唐环境产业集团股份有限公司 | Limestone slurry density measurement method, system and equipment based on LSSVM |
CN111871589A (en) * | 2020-08-06 | 2020-11-03 | 保定正德电力技术有限公司 | Intelligent control method for wet ball mill pulping system |
CN113203656A (en) * | 2020-03-03 | 2021-08-03 | 大唐环境产业集团股份有限公司 | Limestone slurry density measuring system |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109634247A (en) * | 2019-01-23 | 2019-04-16 | 大唐环境产业集团股份有限公司 | A kind of lime stone slurry density hard measurement system and method |
-
2019
- 2019-04-23 CN CN201910331808.0A patent/CN110044767A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109634247A (en) * | 2019-01-23 | 2019-04-16 | 大唐环境产业集团股份有限公司 | A kind of lime stone slurry density hard measurement system and method |
Non-Patent Citations (4)
Title |
---|
伍跃辉 等: "《基于水生态功能分区的流域水环境监测体系构建与应用》", 31 August 2018, 中国环境出版集团 * |
俞佩菲 等: "基于PCA和LS-SVM的软测量建模与应用", 《江南大学学报(自然科学版)》 * |
刘定平 等: "湿法脱硫石灰石浆液密度的敏感性分析", 《环境工程学报》 * |
孟海东 等: "《大数据挖掘技术与应用》", 31 December 2014, 冶金工业出版社 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111238997A (en) * | 2020-02-12 | 2020-06-05 | 江南大学 | On-line measurement method for feed density in crude oil desalting and dewatering process |
CN111238997B (en) * | 2020-02-12 | 2021-07-27 | 江南大学 | On-line measurement method for feed density in crude oil desalting and dewatering process |
CN113203656A (en) * | 2020-03-03 | 2021-08-03 | 大唐环境产业集团股份有限公司 | Limestone slurry density measuring system |
CN111766179A (en) * | 2020-07-08 | 2020-10-13 | 大唐环境产业集团股份有限公司 | Limestone slurry density measurement method, system and equipment based on LSSVM |
CN111871589A (en) * | 2020-08-06 | 2020-11-03 | 保定正德电力技术有限公司 | Intelligent control method for wet ball mill pulping system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110044767A (en) | Based on LSSVM lime stone slurry density flexible measurement method | |
CN101907867B (en) | Self-tuning expert control method of burning trough point parameter based on working condition recognition | |
CN103695639B (en) | Sinter basicity inflation method | |
CN102151605A (en) | Advanced control method and system for vertical mill based on model identification and predictive control | |
CN107366560B (en) | A kind of exhaust steam pressure optimization method of wet type cooling unit steam turbine | |
CN109634247A (en) | A kind of lime stone slurry density hard measurement system and method | |
CN103034525A (en) | Virtual machine power consumption measuring method based on performance counter in cloud computation environment | |
CN106676934B (en) | A kind of high consistency refining systematic economy optimization progress control method and system | |
CN206451018U (en) | A kind of lithium battery mix dosing system | |
CN106050637A (en) | Online monitoring method for operation state of large-sized variable-speed water-feeding pump | |
CN203080038U (en) | Automatic batching system for alloy | |
CN108333943A (en) | Crude oil based on incremental mode concocts method for on-line optimization | |
CN113181799A (en) | Limestone slurry preparation system and control method | |
CN112631121A (en) | Automatic monitoring control method and system for cement self-supporting type rolling mill | |
CN116440670B (en) | Limestone slurry density stability control method | |
CN109733891B (en) | Energy-saving detection control instrument for storehouse pump and working method thereof | |
CN108414061A (en) | A kind of virtual belt conveyer scale implementation method based on multi-material automatic identification | |
CN101995845A (en) | Field programmable gate array (FPGA)-based auto-leveling control system and method | |
CN106599563B (en) | A kind of powder manufacturing apparatus method of calibration and device | |
CN110703713A (en) | Method for improving switching efficiency of single-device multi-product processing scheme | |
CN111399382B (en) | Control method based on full-automatic down filling machine | |
CN207362160U (en) | A kind of device for preventing gasification furnace twin furnace from continuously jumping car | |
Chen et al. | Discrete-time scheduling model of entire refinery with multiscale operation time | |
CN220194241U (en) | Pressure filter doublestage pan feeding adulterated thick concentration control system | |
CN206273352U (en) | A kind of intelligent mixer |
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 |
Application publication date: 20190723 |
|
RJ01 | Rejection of invention patent application after publication |