CN108171330A - A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake - Google Patents

A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake Download PDF

Info

Publication number
CN108171330A
CN108171330A CN201711421894.1A CN201711421894A CN108171330A CN 108171330 A CN108171330 A CN 108171330A CN 201711421894 A CN201711421894 A CN 201711421894A CN 108171330 A CN108171330 A CN 108171330A
Authority
CN
China
Prior art keywords
bittern
hollow tubular
magnesium carbonate
basic magnesium
pso
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.)
Withdrawn
Application number
CN201711421894.1A
Other languages
Chinese (zh)
Inventor
吴健松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Lingnan Normal University
Original Assignee
Lingnan Normal University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Lingnan Normal University filed Critical Lingnan Normal University
Priority to CN201711421894.1A priority Critical patent/CN108171330A/en
Publication of CN108171330A publication Critical patent/CN108171330A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • C—CHEMISTRY; METALLURGY
    • C30—CRYSTAL GROWTH
    • C30B—SINGLE-CRYSTAL GROWTH; UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL; PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; APPARATUS THEREFOR
    • C30B29/00—Single crystals or homogeneous polycrystalline material with defined structure characterised by the material or by their shape
    • C30B29/10—Inorganic compounds or compositions
    • C—CHEMISTRY; METALLURGY
    • C30—CRYSTAL GROWTH
    • C30B—SINGLE-CRYSTAL GROWTH; UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL; PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; APPARATUS THEREFOR
    • C30B29/00—Single crystals or homogeneous polycrystalline material with defined structure characterised by the material or by their shape
    • C30B29/60—Single crystals or homogeneous polycrystalline material with defined structure characterised by the material or by their shape characterised by shape
    • C30B29/62—Whiskers or needles
    • C—CHEMISTRY; METALLURGY
    • C30—CRYSTAL GROWTH
    • C30B—SINGLE-CRYSTAL GROWTH; UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL; PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; APPARATUS THEREFOR
    • C30B7/00—Single-crystal growth from solutions using solvents which are liquid at normal temperature, e.g. aqueous solutions
    • C30B7/14—Single-crystal growth from solutions using solvents which are liquid at normal temperature, e.g. aqueous solutions the crystallising materials being formed by chemical reactions in the solution

Landscapes

  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Organic Chemistry (AREA)
  • Metallurgy (AREA)
  • Materials Engineering (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Inorganic Chemistry (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • General Chemical & Material Sciences (AREA)
  • Compounds Of Alkaline-Earth Elements, Aluminum Or Rare-Earth Metals (AREA)

Abstract

The invention discloses a kind of intelligent methods for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake.Described method includes following steps：S1., multigroup experimental data that hollow tubular basic magnesium carbonate whisker is prepared using bittern in salt lake is used for the BP neural network that PSO is trained to optimize, the experimental data to include preparation parameter and experimental result, and the group number of experimental data is more than or equal to 100 groups；S2. the BP neural network of the PSO optimizations after training in step sl can provide optimization preparation parameter, prediction whisker ratio with the physical and chemical properties change of bittern；Using the optimization preparation parameter, hollow tubular basic magnesium carbonate whisker is prepared.Experimental data is realized into digital intellectualization using PSO Optimized BP Neural Networks, hollow tubular basic magnesium carbonate whisker can be produced for bittern at any time and optimal experimental program and Accurate Prediction experimental result are provided, greatly reduce and a large amount of human and material resources and time and the wasting of resources are brought due to groping property is tested.

Description

A kind of intelligence prepares hollow tubular basic magnesium carbonate whisker using bittern in salt lake Method
Technical field
The present invention relates to technical field of preparation for inorganic material, and bittern in salt lake system is utilized more particularly, to a kind of intelligence The method of standby hollow tubular basic magnesium carbonate whisker.
Background technology
Bittern, the by-product of preparing salt by working up seawater industry, wherein the magnesium ion containing high concentration, and China's by-product bittern is annual Total amount be up to 18,000,000 m3, therefore, hollow tubular basic magnesium carbonate whisker is prepared using bittern, not only increases the attached of bittern It is value added, the manufacture cost of hollow tubular basic magnesium carbonate whisker can also be reduced.But because of geographical and natural environment, sea salt extraction work The difference of skill etc., the pH value of bittern, the inorganic matter ionic species contained and concentration are different, therefore, per the new bittern of a batch When hollow tubular basic magnesium carbonate whisker is prepared, it is required for groping again the condition of its generation whisker, needs consumption big The time of amount can not carry out continual batch production.
Therefore it provides a kind of method for being quickly obtained new bittern and preparing the condition of hollow tubular basic magnesium carbonate whisker, Improve and the efficiency of hollow tubular basic magnesium carbonate whisker prepared using bittern, reduce its time cost, man power and material method, It is worth with important Economic Application.
Invention content
It is an object of the invention to overcome above-mentioned shortcoming and defect of the prior art, a kind of intelligence is provided and utilizes salt lake The method that bittern prepares hollow tubular basic magnesium carbonate whisker.
The present invention introduces BP-PSO (nerves during hollow tubular basic magnesium carbonate whisker is prepared using bittern Network-particle cluster algorithm, Back Propagation-Particle Swarm Optimization), using its adaptivity, Learning ability and Large-scale parallel computing ability, and can fast search to optimal value advantage, by bittern sample data into Row acquisition and analysis, can quickly learn that new bittern prepares the condition of hollow tubular basic magnesium carbonate whisker, save and manually touch one by one The process of rope optimum condition saves human and material resources, and improve and prepare hollow tubular basic magnesium carbonate crystalline substance using bittern in salt lake The efficiency of palpus.
The above-mentioned purpose of the present invention is achieved by the following method：
A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake, includes the following steps：
S1. the multigroup experimental data for being prepared hollow tubular basic magnesium carbonate whisker using bittern in salt lake is used to train PSO The BP neural network of optimization, the experimental data include preparation parameter and experimental result, and the group number of experimental data is more than or equal to 100 Group；
The mathematical model of the PSO is：
vij(t+1)=vij(t)+c1r1j(t)[pij(t)-xij(t)]+c2r2j(t)[gj(t)-xij(t)] (1)
xij(t+1)=xij(t)+vij(t+1) (2)
In formula (1) and formula (2), j is that the jth of particle is tieed up, and i is i-th of particle；T represents current evolutionary generation；vij(t) it is hidden Containing i-th of node of layer to the weights between j-th of node；pij(t) particle i passes through the position of j；gij(t) population i is passed through Cross the position of j；xij(t) current particle i is in the position of j；c1And c2All it is the restriction factor of change in displacement, is preset value；R be with Machine function；
PSO Optimized BP Neural Networks are to make (3) formula J values minimum, and J is mean square deviation index；
In formula (3), N is training sample sum,It is the target output value of j-th of neurode of i-th of sample, yj,i It is the real output value of j-th of neurode of i-th of sample, M is the number of neuron；
S2. the BP neural network of the PSO optimizations in S1 after training can become with the physicochemical property of bittern in salt lake Change and provide optimization preparation parameter, prognostic experiment result；Using the optimization preparation parameter, hollow tubular alkali formula carbon is prepared Sour magnesium whisker.
Preferably, c1And c2All it is the restriction factor of change in displacement, is preset value, usual value is 2.
BP (Back Propagation) neural network is a kind of multilayer feedforward god trained by error backpropagation algorithm It is one of current most widely used neural network model through network.It can learn and store a large amount of input-output pattern and reflect Penetrate relationship, the math equation of this mapping relations described without disclosing in advance, have stronger adaptivity, learning ability and Large-scale parallel computing ability.It is therefore possible to use BP neural network is acquired and analyzes to bittern sample data, quick Know that new bittern prepares the condition of hollow tubular basic magnesium carbonate whisker, have the advantages that fast and accurately.But BP neural network list Solely in use, there are the deficiencies of sample requirement amount is big, convergence rate is slow, generalization ability is weak.
PSO is a kind of new evolution algorithm developed in recent years, is the quality that solution is evaluated by fitness, passes through Follow current search to optimal value find global optimum, have many advantages, such as to realize easily, precision is high, convergence is fast.
For this purpose, experimental data is realized into digital intellectualization using PSO Optimized BP Neural Networks, can be " from complexity at any time Alkali magnesium sulfate crystal whisker is produced in changeable bittern " problem solving is carried out, after experimental data intelligence, make up BP neural network Deficiency, allows error smaller, the accuracy higher of BP-PSO.Bittern complicated and changeable is analyzed using BP-PSO, can quickly, Whisker is produced for bittern at any time and optimal experimental program and Accurate Prediction experimental result are provided, without expending a large amount of manpower, object Power and time cost.
The process of the BP neural network of PSO optimizations is as shown in Figure 3.
Preferably, the preparation parameter includes magnesium ion concentration, bittern volume, Na2CO3-NaHCO3Buffer solution volume, carbon The volume and concentration of sour hydrogen salt, reaction temperature, reaction time and system pH.
Preferably, the bicarbonate is sodium bicarbonate, saleratus and ammonium hydrogen carbonate.
It is highly preferred that the bicarbonate is sodium bicarbonate.
With bittern, Na2CO3-NaHCO3Buffer solution and bicarbonate solution are raw material, prepare hollow tubular basic magnesium carbonate The specific experiment operation of whisker method and preparation parameter can refer to the prior art and obtain.
Preferably, the experimental result includes whisker ratio.
Preferably, the algorithm flow of the BP neural network of the PSO optimizations is：Determine network topology structure, then initially Then BP neural network weight threshold length obtains optimal power threshold value, then calculates error, then weight threshold updates, if Meet termination condition, if then simulation and prediction is obtained as a result, termination condition is unsatisfactory for, back to error is calculated, until meeting Termination condition, final simulation and prediction obtain result；The acquisition best initial weights threshold value is obtained using cluster ion algorithm, the ion The flow of group's algorithm is that ion and speed initialization, then particle fitness value calculation, then looks for individual extreme value and group pole It is worth, then speed update and location updating, then ion fitness value calculation, then individual extreme value and group's extreme value, if full Sufficient condition then obtains optimal power threshold value, if being unsatisfactory for condition, back to speed update and location updating, until meeting item Part obtains best initial weights threshold value.
The data group number is more, and the prognostic experiment result of the BP neural network of PSO optimizations is more accurate.
Compared with prior art, it is of the invention to have the following advantages and beneficial effect：
Experimental data is realized digital intellectualization by the present invention using PSO Optimized BP Neural Networks, can be produced at any time for bittern Whisker provides optimal experimental program and Accurate Prediction experimental result, greatly reduces and a large amount of people is brought due to groping property is tested Power, material resources and time and the waste of resource.
Description of the drawings
Fig. 1 is PSO Optimized BP Neural Network algorithm flow schematic diagrames.
Fig. 2 is the preparation parameter training PSO-BP nerve nets that hollow tubular basic magnesium carbonate whisker is prepared using bittern in salt lake Network schematic diagram.
Fig. 3 is the training process of PSO-BP neural networks.
Specific embodiment
The content further illustrated the present invention with reference to specific embodiment, but should not be construed as limiting the invention. Without departing from the spirit and substance of the case in the present invention, the simple modifications or substitutions made to the method for the present invention, step or condition, It all belongs to the scope of the present invention；Unless otherwise specified, technological means is well known to those skilled in the art used in embodiment Conventional means.
1 PSO-BP Neural Networks Solution examples of embodiment
Make to be described briefly PSO-BP Neural Networks Solution processes below by way of table 1.Table 1 gives 5 groups of experimental datas, Wherein the I, the II, III group is the experiment done and gives preparation parameter and experimental result, and the IVth and V group has multiple parameters (or result) is unknown (unknown-value is marked with question mark).BP neural network (training data is trained with the I, the II, III group of data Group is The more the better), the intelligence of " how to optimize and prepare whisker scheme and make the result optimal " is there has been after BP neural network is trained Energy algorithm can be instructed according to the given data inputted in the IVth group or V group of experiment, according to former generated intelligent algorithm It carries out calculating remaining unknown data.Whisker ratio is that the hollow tubular basic magnesium carbonate whisker that will be prepared is used in table 1 Flying-spot microscope is observed, and confirms the whisker being prepared for target product by XRD tests, and calculate what mass fraction obtained As a result.
For example, in the Vth group, it is known that bittern magnesium ion concentration be 3.2mol/L, volume 400.0mL, existing alkali it is dense It spends for 1.0mol/L, bittern temperature is 19 DEG C, it is assumed that prepare the whisker that ratio is 93%, may I ask other preparation parameters should It is how manyAssume again after being tested by the parameter provided, as a result ratio only has 90%, what main cause isCope with parameter What kind of adjustment madeThe intelligent algorithm that all these problems can all allow PSO-BP neural networks to be generated after training according to before is quick Obtain answer.
Wherein PSO-BP neural network algorithms flow is as shown in Figure 1, the algorithm flow of the BP neural network of PSO optimizations is： It determines network topology structure, then initial BP neural network weight threshold length, then obtains optimal power threshold value, then calculate and miss Difference, then weight threshold update, if meeting termination condition, if simulation and prediction is obtained as a result, be unsatisfactory for termination condition, Back to error is calculated, until meeting termination condition, final simulation and prediction obtains result；Best initial weights threshold value is wherein obtained to use Cluster ion algorithm obtains, and the flow of the cluster ion algorithm is ion and speed initialization, then particle fitness value calculation, Individual extreme value and group's extreme value are then looked for, then speed update and location updating, then ion fitness value calculation, then a Body extreme value and group's extreme value if meeting condition, obtain optimal power threshold value, if being unsatisfactory for condition, back to speed more New and location updating until meeting condition, obtains best initial weights threshold value.
1 PSO-BP Neural Networks Solution examples of table
Embodiment 2
1st, prepare hollow tubular basic magnesium carbonate whisker in accordance with the following methods, and record all parameters in experimentation and Process：
(1) Cha Er Han Salt Lake bittern 500mL is taken, filter and discards insoluble matter；So that magnesium ion concentration is 2.4mol/L；Match Na processed2CO3-NaHCO3The NaHCO of buffer solution and 1.500mol/L3Solution；
(2) the above-mentioned bitterns of 90mL are taken, the NaHCO of 1.500mol/L is added dropwise thereto3Solution 90mL；Na is being added dropwise2CO3- NaHCO3Buffer solution causes the pH value of reaction system to be adjusted to 9.0；
(3) above-mentioned mixed system is placed in the water-bath that water temperature is 28 DEG C and is aged 48h；Then it filters, collects whisker simultaneously It is washed with water 3 times, then is placed in baking oven (T=60 ± 3 DEG C, t=24h) drying to get hollow tubular basic magnesium carbonate whisker.It will Obtained whisker carries out XRD tests, and what is verified is hollow tubular basic magnesium carbonate whisker, and calculate its mass fraction.
Record bittern volume, magnesium ion concentration, NaHCO in above-mentioned experimentation3Concentration and volume, the pH value of system, Reaction temperature, reaction time and whisker ratio (i.e. the mass fraction of whisker).In addition to whisker ratio, and change in above-mentioned parameter One or more and repeat above-mentioned experiment 100 times, record obtains 100 groups of data.
2nd, PSO-BP neural networks training
(1) 100 groups of data obtained above are inputted into PSO-BP neural networks, training PSO-BP neural networks allow it to give birth to Into about bittern volume, magnesium ion concentration, NaHCO3Concentration and volume, the pH value of system, reaction temperature, reaction time and crystalline substance Algorithm between palpus ratio；
(2) after algorithm generation, change above-mentioned parameter and input data, PSO-BP neural networks can be according to the calculation generated Method and the parameter prediction whisker ratio of input, finally obtain and take an excellent data group, this group of data are exactly in chemical technology Optimal, the data of this optimization of direct basis produce hollow tubular basic magnesium carbonate whisker to enter in production afterwards.
The data of PSO-BP neural networks are trained in the present embodiment and the results are shown in Table 2,5 groups of representatives are given in table 2 Data, others training are not shown in the table.By whisker ratio according to being ranked sequentially from low to high, respectively with number 1,2,3, 4th, 5 represent, trained purpose be exactly make real output value infinite close to desired output, (desired output is exactly actually to test Obtained result), once achieving the goal, training terminates.As can be known from Table 2, both desired output and real output value be very It is close, error very little, it was demonstrated that accuracy of the present invention is preferable.
The prediction training of 2 BP neural network of table

Claims (5)

  1. A kind of 1. intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake, which is characterized in that including Following steps：
    S1. the multigroup experimental data for being prepared hollow tubular basic magnesium carbonate whisker using bittern in salt lake is used to that PSO to be trained to optimize BP neural network, the experimental data includes preparation parameter and experimental result, and the group number of experimental data is more than or equal to 100 groups；
    The mathematical model of the PSO is：
    ；
    Formula（1）And formula（2）In,jFor particlejDimension,iIt isiA particle；T represents current evolutionary generation；v ij (t) hidden layer TheiA node is tojWeights between a node；p ij (t) particleiPassed throughjPosition；g ij (t) populationiPassed throughj Position；x ij (t) current particlei jPosition；c1And c2All it is the restriction factor of change in displacement, is preset value；rIt is random Function；
    PSO Optimized BP Neural Networks are to make (3) formulaJValue is minimum,JFor mean square deviation index；
    ；
    Formula（3）In, N is training sample sum,It isiThe of a samplejThe target output value of a neurode isi The of a samplejThe real output value of a neurode, M are the numbers of neuron；
    S2. in step sl train after the PSO optimization BP neural network can with the physical and chemical properties change of bittern and Provide optimization preparation parameter, prognostic experiment result；Using the optimization preparation parameter, hollow tubular basic magnesium carbonate is prepared Whisker.
  2. 2. method according to claim 1, which is characterized in that the c1And c2It is 2.
  3. 3. method according to claim 1, which is characterized in that the preparation parameter include magnesium ion concentration, bittern volume, Na2CO3- NaHCO3Buffer solution volume, the volume of bicarbonate and concentration, reaction temperature, reaction time and system pH.
  4. 4. method according to claim 3, which is characterized in that the bicarbonate is sodium bicarbonate, saleratus and carbonic acid Hydrogen ammonium.
  5. 5. method according to claim 3, which is characterized in that the bicarbonate is sodium bicarbonate.
CN201711421894.1A 2017-12-25 2017-12-25 A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake Withdrawn CN108171330A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711421894.1A CN108171330A (en) 2017-12-25 2017-12-25 A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711421894.1A CN108171330A (en) 2017-12-25 2017-12-25 A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake

Publications (1)

Publication Number Publication Date
CN108171330A true CN108171330A (en) 2018-06-15

Family

ID=62520401

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711421894.1A Withdrawn CN108171330A (en) 2017-12-25 2017-12-25 A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake

Country Status (1)

Country Link
CN (1) CN108171330A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020184569A1 (en) * 2001-04-25 2002-12-05 O'neill Michael System and method for using neural nets for analyzing micro-arrays
CN104463343A (en) * 2014-10-27 2015-03-25 中国石油大学(北京) Method for predicting catalytic cracking light oil yield
CN104445304A (en) * 2014-12-09 2015-03-25 天津渤化永利化工股份有限公司 Method for preparing basic magnesium carbonate by using combinative alkali system high-salt heavy-ash mother solution

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020184569A1 (en) * 2001-04-25 2002-12-05 O'neill Michael System and method for using neural nets for analyzing micro-arrays
CN104463343A (en) * 2014-10-27 2015-03-25 中国石油大学(北京) Method for predicting catalytic cracking light oil yield
CN104445304A (en) * 2014-12-09 2015-03-25 天津渤化永利化工股份有限公司 Method for preparing basic magnesium carbonate by using combinative alkali system high-salt heavy-ash mother solution

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
刘卫林，刘丽娜: "《基于智能计算技术的水资源配置系统预测》", 1 December 2015, 中国水利水电出版社 *
吴健松等: "徐闻盐场苦卤制备Mg2(OH)2CO3•3H2O晶须", 《人工晶体学报》 *
董威: "《粗糙集理论及其数据挖掘应用》", 1 December 2009, 东北大学出版社 *

Similar Documents

Publication Publication Date Title
CN103218669B (en) A kind of live fish cultivation water quality comprehensive forecasting method of intelligence
CN103345559B (en) The dynamic evolutionary modeling method of aluminium electrolysis process art electrolysis cells energy consumption
CN105913151A (en) Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network
CN103971160B (en) particle swarm optimization method based on complex network
CN106529818B (en) Water quality assessment Forecasting Methodology based on Fuzzy Wavelet Network
CN107728477A (en) A kind of industrialized aquiculture water quality dissolved oxygen prediction control method and system
CN106802563B (en) A kind of sewage procedure optimization control method based on drosophila optimization and LSSVM
CN109934422A (en) Neural network wind speed prediction method based on time series data analysis
CN103745271A (en) Forecasting method for induction thermal deposition calcium-phosphate coating process on basis of neural network
CN111652425A (en) River water quality prediction method based on rough set and long and short term memory network
CN109308544A (en) Based on to sdpecific dispersion-shot and long term memory network cyanobacterial bloom prediction technique
CN105069220A (en) Back-propagation (BP) neural network immune genetic algorithm based microbial fermentation optimization method
CN108416460A (en) Cyanobacterial bloom prediction technique based on the random depth confidence network model of multifactor sequential-
CN107729988B (en) Blue algae bloom prediction method based on dynamic deep belief network
CN109978024B (en) Effluent BOD prediction method based on interconnected modular neural network
CN108171330A (en) A kind of intelligent method for preparing hollow tubular basic magnesium carbonate whisker using bittern in salt lake
CN109002878A (en) A kind of GA Optimized BP Neural Network
Linlin et al. Research on the back propagation neural network haze prediction model based on particle swarm optimization
CN111126827A (en) Input-output accounting model construction method based on BP artificial neural network
Deng et al. Application of neural network based on PSO algorithm in prediction model for dissolved oxygen in fishpond
CN105117616A (en) Particle swarm algorithm based optimization method for microbial fermentation
CN110276476A (en) A kind of automobile engine smoke intensity prediction technique based on PSO-RBF neural network
CN108073788A (en) A kind of method that alkali formula magnesium borate crystal whisker is prepared using bittern in salt lake
CN108118396A (en) A kind of method that alkali magnesium sulfate crystal whisker is prepared using bittern in salt lake
CN114782740A (en) Remote sensing water quality monitoring method combining genetic optimization and extreme gradient promotion

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
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20180615