CN109658987A - A kind of hydrofining catalyst carrier property prediction technique - Google Patents
A kind of hydrofining catalyst carrier property prediction technique Download PDFInfo
- Publication number
- CN109658987A CN109658987A CN201811532214.8A CN201811532214A CN109658987A CN 109658987 A CN109658987 A CN 109658987A CN 201811532214 A CN201811532214 A CN 201811532214A CN 109658987 A CN109658987 A CN 109658987A
- Authority
- CN
- China
- Prior art keywords
- catalyst carrier
- hydrofining catalyst
- data
- neural network
- model
- 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
- 239000003054 catalyst Substances 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000012549 training Methods 0.000 claims abstract description 25
- 238000003062 neural network model Methods 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 230000002068 genetic effect Effects 0.000 claims abstract description 12
- 238000004519 manufacturing process Methods 0.000 claims abstract description 9
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 claims abstract description 7
- 238000013210 evaluation model Methods 0.000 claims abstract description 4
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 239000011148 porous material Substances 0.000 claims description 32
- 229910001593 boehmite Inorganic materials 0.000 claims description 25
- FAHBNUUHRFUEAI-UHFFFAOYSA-M hydroxidooxidoaluminium Chemical compound O[Al]=O FAHBNUUHRFUEAI-UHFFFAOYSA-M 0.000 claims description 25
- 238000010521 absorption reaction Methods 0.000 claims description 18
- QAOWNCQODCNURD-UHFFFAOYSA-L Sulfate Chemical compound [O-]S([O-])(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-L 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 13
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 230000006978 adaptation Effects 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 210000004218 nerve net Anatomy 0.000 claims 1
- 238000005265 energy consumption Methods 0.000 abstract description 4
- 239000003921 oil Substances 0.000 description 5
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 4
- 239000003795 chemical substances by application Substances 0.000 description 4
- 150000002500 ions Chemical class 0.000 description 4
- 239000000446 fuel Substances 0.000 description 3
- 239000000843 powder Substances 0.000 description 3
- 229910021653 sulphate ion Inorganic materials 0.000 description 3
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 2
- GRYLNZFGIOXLOG-UHFFFAOYSA-N Nitric acid Chemical compound O[N+]([O-])=O GRYLNZFGIOXLOG-UHFFFAOYSA-N 0.000 description 2
- 241000219782 Sesbania Species 0.000 description 2
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 description 2
- 239000005864 Sulphur Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000001257 hydrogen Substances 0.000 description 2
- 229910052739 hydrogen Inorganic materials 0.000 description 2
- 238000000465 moulding Methods 0.000 description 2
- 229910017604 nitric acid Inorganic materials 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 238000007670 refining Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- WURBVZBTWMNKQT-UHFFFAOYSA-N 1-(4-chlorophenoxy)-3,3-dimethyl-1-(1,2,4-triazol-1-yl)butan-2-one Chemical compound C1=NC=NN1C(C(=O)C(C)(C)C)OC1=CC=C(Cl)C=C1 WURBVZBTWMNKQT-UHFFFAOYSA-N 0.000 description 1
- CIWBSHSKHKDKBQ-JLAZNSOCSA-N Ascorbic acid Chemical compound OC[C@H](O)[C@H]1OC(=O)C(O)=C1O CIWBSHSKHKDKBQ-JLAZNSOCSA-N 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- UCKMPCXJQFINFW-UHFFFAOYSA-N Sulphide Chemical compound [S-2] UCKMPCXJQFINFW-UHFFFAOYSA-N 0.000 description 1
- 239000004411 aluminium Substances 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000003197 catalytic effect Effects 0.000 description 1
- 238000006555 catalytic reaction Methods 0.000 description 1
- 239000010779 crude oil Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 229910001648 diaspore Inorganic materials 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 235000013312 flour Nutrition 0.000 description 1
- 125000005842 heteroatom Chemical group 0.000 description 1
- 238000005984 hydrogenation reaction Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 150000004767 nitrides Chemical class 0.000 description 1
- 125000005575 polycyclic aromatic hydrocarbon group Chemical group 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10G—CRACKING HYDROCARBON OILS; PRODUCTION OF LIQUID HYDROCARBON MIXTURES, e.g. BY DESTRUCTIVE HYDROGENATION, OLIGOMERISATION, POLYMERISATION; RECOVERY OF HYDROCARBON OILS FROM OIL-SHALE, OIL-SAND, OR GASES; REFINING MIXTURES MAINLY CONSISTING OF HYDROCARBONS; REFORMING OF NAPHTHA; MINERAL WAXES
- C10G45/00—Refining of hydrocarbon oils using hydrogen or hydrogen-generating compounds
- C10G45/02—Refining of hydrocarbon oils using hydrogen or hydrogen-generating compounds to eliminate hetero atoms without changing the skeleton of the hydrocarbon involved and without cracking into lower boiling hydrocarbons; Hydrofinishing
- C10G45/04—Refining of hydrocarbon oils using hydrogen or hydrogen-generating compounds to eliminate hetero atoms without changing the skeleton of the hydrocarbon involved and without cracking into lower boiling hydrocarbons; Hydrofinishing characterised by the catalyst used
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10G—CRACKING HYDROCARBON OILS; PRODUCTION OF LIQUID HYDROCARBON MIXTURES, e.g. BY DESTRUCTIVE HYDROGENATION, OLIGOMERISATION, POLYMERISATION; RECOVERY OF HYDROCARBON OILS FROM OIL-SHALE, OIL-SAND, OR GASES; REFINING MIXTURES MAINLY CONSISTING OF HYDROCARBONS; REFORMING OF NAPHTHA; MINERAL WAXES
- C10G2300/00—Aspects relating to hydrocarbon processing covered by groups C10G1/00 - C10G99/00
- C10G2300/70—Catalyst aspects
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Physics & Mathematics (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- General Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Organic Chemistry (AREA)
- Catalysts (AREA)
Abstract
The invention discloses a kind of hydrofining catalyst carrier property prediction technique, includes the following steps: basic data when 1) obtaining production hydrofining catalyst carrier, and data normalization is handled, form database；2) BP neural network model is established；3) the BP neural network model in step 2) is optimized using genetic algorithm, obtains the corresponding network initial weight of optimum individual in the model and threshold value；4) model training is carried out to BP neural network model using Levenberg-Marquardt algorithm, obtains training data；5) training data is inputted in established BP neural network model and obtains output valve, predicted value will obtained after numerical value renormalization, compared with desired value, by mean square error and the coefficient of determination come the superiority and inferiority of evaluation model.This method list, energy consumption is small, and precision of prediction is high, and algorithmic statement is fast, can preferably express the relationship between the key factor in hydrofining catalyst carrier production process and hydrofining catalyst carrier property.
Description
Technical field
The present invention relates to catalyst fields.More particularly, to a kind of hydrofining catalyst carrier property prediction technique.
Background technique
In recent years, with the continuous increase of oil extraction amount, world's crude oil heaviness problem is on the rise, in oil product
Sulphur, nitrogen and tenor are also continuously increased therewith.It is asked to alleviate atmosphere pollution caused by increasingly increased motor vehicle exhaust emission etc.
Stringent vehicle fuel quality standard has been formulated in topic, countries in the world in succession, to force oil plant to produce clean fuel.Hydrofinishing
Technology is the most effective means for improving oil quality and producing clean fuel, under Hydrofinishing conditions, sulfide in oil product,
Nitride is removed the hetero atoms such as sulphur, nitrogen by the way that hydrodesulfurization (HDS) and hydrodenitrogeneration (HDN) reaction occurs, while can also be sent out
The fractional saturation of raw polycyclic aromatic hydrocarbon and the hydrogenation dearomatization reaction of open loop, are improved the Cetane number of product, and hydrofinishing
The development of catalyst is the key that then this technology.
The quality of hydrofining catalyst carrier is to influence the key factor of catalyst performance.Hydrofining catalyst carrier
Mainly by screw rod banded extruder extruded moulding, carrier wet bar after molding obtained after drying, broken strip, roasting catalyst carrier at
Product generally determine catalyst carrier by indexs such as the specific surface area of analysis detection catalyst carrier, pore volume, effective hole appearances
Property.Whole process preparation process is cumbersome, and material consumption, energy consumption are larger in preparation process, and repeated labor is more.
Accordingly, it is desirable to provide a kind of new hydrofining catalyst carrier property prediction technique, above-mentioned to solve
Technical problem.
Summary of the invention
The purpose of the present invention is to provide a kind of hydrofining catalyst carrier property prediction techniques, and this method is simple, energy
Consume small, and precision of prediction is high, and algorithmic statement is fast, can preferably express the key in hydrofining catalyst carrier production process
Relationship between factor and hydrofining catalyst carrier property.Hydrofinishing catalysis can be fast and accurately understood according to predicted value
Agent carrier property is effectively reduced a large amount of duplication of labour and energy consumption, improves Hydrobon catalyst research and development speed.
In order to achieve the above objectives, the present invention adopts the following technical solutions:
A kind of hydrofining catalyst carrier property prediction technique, includes the following steps:
1) basic data when production hydrofining catalyst carrier is obtained, and data normalization is handled, forms data
Library；
2) BP neural network model is established, the number for influencing hydrofining catalyst carrier property in step 1) database is selected
According to the data as input data, hydrofining catalyst carrier property as output data；
3) the BP neural network model in step 2) is optimized using genetic algorithm, obtains optimum individual in the model
Corresponding network initial weight and threshold value；
4) model is carried out to the BP neural network model after optimization in step 3) using Levenberg-Marquardt algorithm
Training, obtains training data；
5) training data that step 4) obtains is inputted in established BP neural network model and obtains output valve, by numerical value
Predicted value is obtained after renormalization, is compared with desired value, by mean square error and the coefficient of determination come the superiority and inferiority of evaluation model.
In a preferred embodiment, the method for the production hydrofining catalyst carrier includes: to utilize to intend thin water
The raw materials such as aluminium mountain flour body, nitric acid, sesbania powder, pure water are uniformly mixed in varing proportions, kneaded and formed by double screw banded extruder, warp
Hydrofining catalyst carrier is made after 600 DEG C of high-temperature roastings.
In a preferred embodiment, in step 1), basic data when making hydrofining catalyst carrier includes:
BET specific surface area, the absorption total pore volume of boehmite of boehmite, sulfate ion content, pH in boehmite
Value, the mass ratio of pure water additional amount and boehmite, the BET specific surface area of hydrofining catalyst carrier, hydrofinishing are urged
The absorption total pore volume of agent carrier and the most probable pore size of hydrofining catalyst carrier.
In yet another preferred embodiment, in step 2), the data for influencing hydrofining catalyst carrier property
The including but not limited to BET specific surface area of boehmite, the absorption total pore volume of boehmite, sulfate radical in boehmite
The mass ratio of ion concentration, pH value and pure water additional amount and boehmite；The hydrofining catalyst carrier property
Data include but is not limited to the total hole of absorption of the BET specific surface area of hydrofining catalyst carrier, hydrofining catalyst carrier
The most probable pore size of appearance and hydrofining catalyst carrier.During hydrofining catalyst carrier production, to final
The factor that the performance for the hydrofining catalyst carrier being prepared has an impact may be very much, and in research of the invention
It was found that the selection to the data for influencing hydrofining catalyst carrier property, directly affects to hydrofining catalyst carrier
The precision of property prediction.
In yet another preferred embodiment, in step 2), in the BP neural network model, BP neural network is multilayer
Feedforward neural network, basic structure are divided into input layer, hidden layer and output layer.Input layer inputs above-mentioned input data, output layer
Export above-mentioned output data.
In yet another preferred embodiment, in step 3), using genetic algorithm to the BP neural network mould in step 2)
The mode that type optimizes include: using genetic algorithm in the model weight and threshold value optimize, the population of the model
In each individual include a network ownership value and threshold value, it is pre- to choose neural network by fitness function herein for individual
The sum of the Error Absolute Value between output valve and desired value after survey E is as ideal adaptation angle value F.
Wherein n is network output node number；yiFor the desired output of i-th of node of BP neural network；oiFor i-th of node
Prediction output；K is coefficient.
It chooses herein, the selection strategy of fitness ratio, i.e. roulette method, the select probability p of each individual iiAre as follows:fi=k/Fi, wherein FiFor the fitness value of individual i；K is coefficient；N is population at individual number.
Ideal adaptation angle value is calculated, genetic algorithm intersects that with mutation operation to find adaptive optimal control angle value corresponding by selection
Individual, and then obtain the corresponding network initial weight of optimum individual in the neural network and threshold value.
In yet another preferred embodiment, in step 4), carry out model training mode include: by sample data according to
The ratio of 5:1 is randomly divided into training data and test data；Training data is inputted in the BP neural network model and is instructed
Practice, using Levenberg-Marquardt algorithm, it is 1000 that frequency of training, which is arranged, training objective 0.001, and learning rate is
0.1。
Beneficial effects of the present invention are as follows:
Hydrofining catalyst carrier property prediction technique of the present invention compensates for existing to hydrofining catalyst carrier
The blank of matter prediction, and traditional neural network (such as BP algorithm) is easily trapped into local optimum, algorithmic statement is slow, and precision of prediction is low
The disadvantages of.The present invention has many advantages, such as that pace of learning is fast, avoids generating locally optimal solution.This method can be adapted for different plus hydrogen
The different carriers property of catalyst for refining.
Hydrofining catalyst carrier property prediction technique of the present invention can preferably express Hydrobon catalyst production
Relationship between key factor in process and hydrofining catalyst carrier property.Can fast and accurately it be understood according to predicted value
Hydrofining catalyst carrier property is effectively reduced a large amount of duplication of labour and energy consumption, improves hydrofinishing and urges
Agent researches and develops speed.Therefore the present invention has actual promotional value.
Detailed description of the invention
Specific embodiments of the present invention will be described in further detail with reference to the accompanying drawing.
Fig. 1 shows the flow chart of 1 prediction technique of embodiment.
Specific embodiment
In order to illustrate more clearly of the present invention, the present invention is done further below with reference to preferred embodiments and drawings
It is bright.Similar component is indicated in attached drawing with identical appended drawing reference.It will be appreciated by those skilled in the art that institute is specific below
The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.
Embodiment 1
A kind of hydrofining catalyst carrier property prediction technique, process is as shown in Figure 1, include the following steps:
1) make hydrofining catalyst carrier: using the raw materials such as boehmite powder, nitric acid, sesbania powder, pure water with
Different proportion is uniformly mixed, kneaded and formed by double screw banded extruder, and Hydrobon catalyst is made after 600 DEG C of high-temperature roastings
Carrier.
2) basic data when the above-mentioned production hydrofining catalyst carrier of acquisition: the BET specific surface area of boehmite,
The absorption total pore volume of boehmite, sulfate ion content, pH value, pure water additional amount and boehmite in boehmite
Mass ratio, the BET specific surface area of hydrofining catalyst carrier, the absorption total pore volume of hydrofining catalyst carrier and add hydrogen
The most probable pore size of refining catalytic agent carrier, and data normalization is handled, the numerical value being converted into [- 1,1] range, form number
According to library；
3) BP neural network model is established, the number for influencing hydrofining catalyst carrier property in step 2) database is selected
According to the data as input data, hydrofining catalyst carrier property as output data；
Wherein, input data includes the BET specific surface area of boehmite, the absorption total pore volume of boehmite, is intended thin
The mass ratio of sulfate ion content in diaspore, pH value and pure water additional amount and boehmite；
Output data includes the absorption of the BET specific surface area of hydrofining catalyst carrier, hydrofining catalyst carrier
The most probable pore size of total pore volume and hydrofining catalyst carrier；
Wherein, BP neural network is multilayer feedforward neural network, and basic structure is divided into input layer, hidden layer and output layer.
4) the BP neural network model in step 3) is optimized using genetic algorithm, obtains optimum individual in the model
Corresponding network initial weight and threshold value, specific method include: using genetic algorithm in the model weight and threshold value carry out
Optimize, each individual includes a network ownership value and threshold value in the population of the model, and individual passes through fitness function meter
Calculate ideal adaptation angle value, genetic algorithm by selection, intersect and mutation operation to find adaptive optimal control angle value corresponding individual, and then obtain
The corresponding network initial weight of optimum individual and threshold value into the neural network；
5) model is carried out to the BP neural network model after optimization in step 4) using Levenberg-Marquardt algorithm
Training, obtains training data, and specific method includes: that sample data is randomly divided into training data and test number according to the ratio of 5:1
According to；Training data is inputted in the BP neural network model and is trained, using Levenberg-Marquardt algorithm, if
Setting frequency of training is 1000, training objective 0.001, learning rate 0.1；
6) training data that step 5) obtains is inputted in established BP neural network model and obtains output valve, by numerical value
Predicted value is obtained after renormalization, is compared with desired value, by mean square error and the coefficient of determination come the superiority and inferiority of evaluation model.
As shown in following table 1- table 3, it can be seen that specific surface area, absorption total pore volume, the predicted value of most probable pore size and reality
Value compares.
Table 1 contains sulfate ion specific surface area predicted value and actual comparison
The predicted value of above-mentioned specific surface area and the mean square error of actual value are 0.0733, the coefficient of determination 0.8286.
Table 2 contains sulfate ion absorption total pore volume predicted value and actual comparison
The predicted value of above-mentioned absorption total pore volume and the mean square error of actual value are 0.023, the coefficient of determination 0.8539.
Table 3 contains sulfate ion most probable pore size predicted value and actual comparison
The predicted value of above-mentioned most probable pore size and the mean square error of actual value are 0.080, the coefficient of determination 0.7411.
It can be seen that the mean square error of three models is all smaller, show that model has good precision of prediction.It is from decision
Number is it can be seen that model has the good goodness of fit.
Comparative example 1
Repeat embodiment 1, difference is, by step 2) and 3) in influence factor boehmite in sulfate ion contain
Amount is deleted, and other conditions are constant, as a result as shown in following table 4- table 6, it can be seen that specific surface when without sulfate ion content
Product, absorption total pore volume, the predicted value of most probable pore size are compared with actual value.
Table 4 not sulphate-containing ion specific surface area predicted value and actual comparison
The predicted value of above-mentioned specific surface area and the mean square error of actual value are 0.1693, the coefficient of determination 0.7031.
Sulphate-containing ion does not adsorb total pore volume predicted value and actual comparison to table 5
The predicted value of above-mentioned absorption total pore volume and the mean square error of actual value are 0.027, the coefficient of determination 0.8591.
Table 3 not sulphate-containing ion most probable pore size predicted value and actual comparison
The predicted value of above-mentioned most probable pore size and the mean square error of actual value are 0.2258, the coefficient of determination 0.7291.
By upper embodiment and comparative example it is found that the model ratio of the model containing sulfate ion and not sulfate ion
Compared with the model containing sulfate ion all has better precision of prediction in specific surface area, the model prediction of most probable pore size
With the goodness of fit of model, two model performances are almost the same in absorption total pore volume.Therefore the model in no sulfate ion
Performance can decline to a great extent.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention may be used also on the basis of the above description for those of ordinary skill in the art
To make other variations or changes in different ways, all embodiments can not be exhaustive here, it is all to belong to this hair
The obvious changes or variations that bright technical solution is extended out are still in the scope of protection of the present invention.
Claims (6)
1. a kind of hydrofining catalyst carrier property prediction technique, which comprises the steps of:
1) basic data when production hydrofining catalyst carrier is obtained, and data normalization is handled, forms database；
2) BP neural network model is established, the data for influencing hydrofining catalyst carrier property in step 1) database is selected to make
It is input data, the data of hydrofining catalyst carrier property as output data；
3) the BP neural network model in step 2) is optimized using genetic algorithm, it is corresponding obtains optimum individual in the model
Network initial weight and threshold value；
4) model instruction is carried out to the BP neural network model after optimization in step 3) using Levenberg-Marquardt algorithm
Practice, obtains training data；
5) training data that step 4) obtains is inputted in established BP neural network model and obtains output valve, returned numerical value is counter
Predicted value is obtained after one change, is compared with desired value, by mean square error and the coefficient of determination come the superiority and inferiority of evaluation model.
2. prediction technique according to claim 1, which is characterized in that in step 1), make hydrofining catalyst carrier
When basic data include: the BET specific surface area of boehmite, the absorption total pore volume of boehmite, in boehmite
Sulfate ion content, pH value, the mass ratio of pure water additional amount and boehmite, hydrofining catalyst carrier BET ratio
Surface area, the absorption total pore volume of hydrofining catalyst carrier and the most probable pore size of hydrofining catalyst carrier.
3. prediction technique according to claim 2, which is characterized in that in step 2), the influence Hydrobon catalyst
The data of support include the BET specific surface area of boehmite, the absorption total pore volume of boehmite, in boehmite
The mass ratio of sulfate ion content, pH value and pure water additional amount and boehmite；The hydrofining catalyst carrier
The data of property include the BET specific surface area of hydrofining catalyst carrier, the absorption total pore volume of hydrofining catalyst carrier
And the most probable pore size of hydrofining catalyst carrier.
4. prediction technique according to claim 1, which is characterized in that in step 2), in the BP neural network model, BP
Neural network is multilayer feedforward neural network, and basic structure is divided into input layer, hidden layer and output layer.
5. prediction technique according to claim 1, which is characterized in that in step 3), using genetic algorithm in step 2)
The mode that optimizes of BP neural network model include: using genetic algorithm in the model weight and threshold value carry out it is excellent
Change, each individual includes a network ownership value and threshold value in the population of the model, and individual is calculated by fitness function
Ideal adaptation angle value, genetic algorithm by selection, intersect and mutation operation to find adaptive optimal control angle value corresponding individual, and then obtain
The corresponding network initial weight of optimum individual and threshold value in the neural network.
6. prediction technique according to claim 1, which is characterized in that in step 4), the mode for carrying out model training includes:
Sample data is randomly divided into training data and test data according to the ratio of 5:1；Training data is inputted into the BP nerve net
It is trained in network model, using Levenberg-Marquardt algorithm, it is 1000 that frequency of training, which is arranged, and training objective is
0.001, learning rate 0.1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811532214.8A CN109658987A (en) | 2018-12-14 | 2018-12-14 | A kind of hydrofining catalyst carrier property prediction technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811532214.8A CN109658987A (en) | 2018-12-14 | 2018-12-14 | A kind of hydrofining catalyst carrier property prediction technique |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109658987A true CN109658987A (en) | 2019-04-19 |
Family
ID=66113100
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811532214.8A Pending CN109658987A (en) | 2018-12-14 | 2018-12-14 | A kind of hydrofining catalyst carrier property prediction technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109658987A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807516A (en) * | 2019-10-31 | 2020-02-18 | 西安工程大学 | Junction temperature prediction method of IGBT module for driver |
| CN112133383A (en) * | 2020-08-21 | 2020-12-25 | 上海大学 | Method for predicting perovskite specific surface area based on genetic symbol regression |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090095657A1 (en) * | 2006-11-07 | 2009-04-16 | Saudi Arabian Oil Company | Automation and Control of Energy Efficient Fluid Catalytic Cracking Processes for Maximizing Value Added Products |
| CN106777922A (en) * | 2016-11-30 | 2017-05-31 | 华东理工大学 | A kind of CTA hydrofinishings production process agent model modeling method |
| CN107545122A (en) * | 2017-09-27 | 2018-01-05 | 重庆长安汽车股份有限公司 | A kind of simulation system of the vehicle gaseous effluent based on neutral net |
| CN108090658A (en) * | 2017-12-06 | 2018-05-29 | 河北工业大学 | Arc fault diagnostic method based on time domain charactreristic parameter fusion |
-
2018
- 2018-12-14 CN CN201811532214.8A patent/CN109658987A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090095657A1 (en) * | 2006-11-07 | 2009-04-16 | Saudi Arabian Oil Company | Automation and Control of Energy Efficient Fluid Catalytic Cracking Processes for Maximizing Value Added Products |
| CN106777922A (en) * | 2016-11-30 | 2017-05-31 | 华东理工大学 | A kind of CTA hydrofinishings production process agent model modeling method |
| CN107545122A (en) * | 2017-09-27 | 2018-01-05 | 重庆长安汽车股份有限公司 | A kind of simulation system of the vehicle gaseous effluent based on neutral net |
| CN108090658A (en) * | 2017-12-06 | 2018-05-29 | 河北工业大学 | Arc fault diagnostic method based on time domain charactreristic parameter fusion |
Non-Patent Citations (1)
| Title |
|---|
| 余夕志: "《汽油和柴油馏分加氢脱硫催化剂及反应动力学研究》", 《中国优秀博硕士学位论文全文数据库（博士） 工程科技Ⅰ辑（月刊）》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807516A (en) * | 2019-10-31 | 2020-02-18 | 西安工程大学 | Junction temperature prediction method of IGBT module for driver |
| CN112133383A (en) * | 2020-08-21 | 2020-12-25 | 上海大学 | Method for predicting perovskite specific surface area based on genetic symbol regression |
| CN112133383B (en) * | 2020-08-21 | 2023-06-13 | 上海大学 | Method for predicting perovskite specific surface area based on genetic symbolic regression |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dong et al. | Drivers of the decoupling indicator between the economic growth and energy-related CO2 in China: A revisit from the perspectives of decomposition and spatiotemporal heterogeneity | |
| CN108228716B (en) | SMOTE _ Bagging integrated sewage treatment fault diagnosis method based on weighted extreme learning machine | |
| CN109658988A (en) | A kind of Hydrobon catalyst performance prediction method | |
| CN109658987A (en) | A kind of hydrofining catalyst carrier property prediction technique | |
| CN109934422B (en) | Neural network wind speed prediction method based on time series data analysis | |
| CN1979496A (en) | Copper-alloy pipe-material casting-milling technology parameter designing and optimizing method | |
| CN108509601B (en) | Food flavor assessment method based on big data analysis | |
| CN107273633B (en) | Varying delay method of estimation and flow time lag method of estimation is hydrocracked between multiple operation | |
| CN104951847A (en) | Rainfall forecast method based on kernel principal component analysis and gene expression programming | |
| CN105740960B (en) | A kind of optimization method of industry hydrocracking reaction condition | |
| CN109831790A (en) | Cooperation caching method based on brainstorming optimization algorithm in mist wireless access network | |
| CN108985323A (en) | A kind of short term prediction method of photovoltaic power | |
| Jin et al. | Fractal topography and complexity assembly in multifractals | |
| CN105404142A (en) | Aluminum electrolytic multi-parameter control method based on BP neural network and MBFO algorithm | |
| CN103164631A (en) | Intelligent coordinate expression gene analyzer | |
| CN106326677B (en) | A kind of flexible measurement method of PTA device acetic acid consumption | |
| Tasneem et al. | Organic catalysts for hydrogen production from noodle wastewater: Machine learning and deep learning-based analysis | |
| CN114202060A (en) | Method for predicting methylene blue adsorption performance of biomass activated carbon based on deep neural network | |
| CN108064779B (en) | The breeding method of the Penaeus monodon of resistance to less salt new product germline | |
| CN106874644A (en) | The real-time predicting method and its system of hydrogenolysis degree in a kind of hydrofinishing | |
| CN109002878A (en) | A kind of GA Optimized BP Neural Network | |
| CN112365071A (en) | Real-time optimization method of S Zorb device based on improved differential evolution algorithm | |
| CN103145516B (en) | Method for selective hydrogenation of unsaturated compound | |
| CN110276476A (en) | A kind of automobile engine smoke intensity prediction technique based on PSO-RBF neural network | |
| CN102629255A (en) | Regional population structure analysis method based on clustering analysis of mitochondrial deoxyribonucleic acid (mtDNA) genetic information |
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: 20190419 |
|
| RJ01 | Rejection of invention patent application after publication |