CN111863151B - Polymer molecular weight distribution prediction method based on Gaussian process regression - Google Patents
Polymer molecular weight distribution prediction method based on Gaussian process regression Download PDFInfo
- Publication number
- CN111863151B CN111863151B CN202010680269.4A CN202010680269A CN111863151B CN 111863151 B CN111863151 B CN 111863151B CN 202010680269 A CN202010680269 A CN 202010680269A CN 111863151 B CN111863151 B CN 111863151B
- Authority
- CN
- China
- Prior art keywords
- molecular weight
- data
- polymer
- distribution
- process parameters
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 144
- 229920000642 polymer Polymers 0.000 title claims abstract description 108
- 238000006116 polymerization reaction Methods 0.000 claims abstract description 60
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000006243 chemical reaction Methods 0.000 claims abstract description 13
- 238000010606 normalization Methods 0.000 claims description 20
- 239000003999 initiator Substances 0.000 claims description 12
- 239000000178 monomer Substances 0.000 claims description 10
- 239000002904 solvent Substances 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 9
- 239000012295 chemical reaction liquid Substances 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 238000002474 experimental method Methods 0.000 claims description 6
- 230000035484 reaction time Effects 0.000 claims description 5
- 238000000746 purification Methods 0.000 claims description 4
- 238000012854 evaluation process Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010998 test method Methods 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims 1
- 230000002776 aggregation Effects 0.000 claims 1
- PPBRXRYQALVLMV-UHFFFAOYSA-N Styrene Chemical compound C=CC1=CC=CC=C1 PPBRXRYQALVLMV-UHFFFAOYSA-N 0.000 description 4
- 239000000047 product Substances 0.000 description 3
- OZAIFHULBGXAKX-UHFFFAOYSA-N 2-(2-cyanopropan-2-yldiazenyl)-2-methylpropanenitrile Chemical group N#CC(C)(C)N=NC(C)(C)C#N OZAIFHULBGXAKX-UHFFFAOYSA-N 0.000 description 2
- RTZKZFJDLAIYFH-UHFFFAOYSA-N Diethyl ether Chemical compound CCOCC RTZKZFJDLAIYFH-UHFFFAOYSA-N 0.000 description 2
- OZAIFHULBGXAKX-VAWYXSNFSA-N AIBN Substances N#CC(C)(C)\N=N\C(C)(C)C#N OZAIFHULBGXAKX-VAWYXSNFSA-N 0.000 description 1
- 239000002202 Polyethylene glycol Substances 0.000 description 1
- 239000004353 Polyethylene glycol 8000 Substances 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 229920001223 polyethylene glycol Polymers 0.000 description 1
- 229940068918 polyethylene glycol 400 Drugs 0.000 description 1
- 229940085678 polyethylene glycol 8000 Drugs 0.000 description 1
- 235000019446 polyethylene glycol 8000 Nutrition 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010517 secondary reaction Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
Landscapes
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Polymerisation Methods In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a polymer molecular weight distribution prediction method based on Gaussian process regression, which comprises the steps of obtaining all process parameters in a polymerization reaction process, recording historical data of corresponding polymer molecular weight concentration distribution, purifying, normalizing all the process parameters, the purified polymer molecular weight distribution and the concentration distribution, introducing the normalized polymer molecular weight distribution and the concentration distribution into a training model, training and simulating the data of the corresponding polymer molecular weight concentration distribution under all the known process parameters in the polymerization reaction process through a Gaussian process distribution model, and constructing a prediction model. The invention relates to a processing mode of polymer molecular weight distribution data and application of a Gaussian process regression model in polymerization reaction, and can analyze the existing polymer molecular weight distribution data and well predict the polymer molecular weight distribution under unknown reaction conditions through combination of the processing mode and the Gaussian process regression model.
Description
Technical Field
The invention relates to a method for predicting polymer molecular weight distribution based on Gaussian process regression.
Background
In the polymerization of polymer monomers in a solvent in the presence of an initiator, the molecular weight of the polymer of the reaction output product is not a single value, but a distribution range of polymer chain length and molecular weight, and the output is often difficult to accurately predict. While Molecular Weight Distribution (MWD) directly affects many end-use properties of the product, such as thermal properties, stress-strain properties, impact resistance, strength and hardness. For example, polyethylene glycol 400, polyethylene glycol 2000 and polyethylene glycol 8000 all have quite different uses.
In addition, during the polymerization reaction, the reaction process parameters involved are often complex and variable, and the process parameters influencing the molecular weight distribution of the final product are generally: reaction temperature, reaction time, polymer monomer concentration, initiator concentration, and the like. The time and space costs of modeling the first principle (physico-chemical formula derivation) for complex reaction process parameters are too great and imperfect. Some data-driven modeling methods do not explain them well and the accuracy of the predictions is low.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention aims to provide a method for predicting the molecular weight distribution of a polymer based on Gaussian process regression. Wherein, in the polymerization reaction of polymer monomers in solvent in the presence of initiator, the molecular weight distribution of the polymer of the output product of the reaction corresponds to the concentration distribution of the molecular weight of the polymer one by one.
The invention provides a potential model suitable for the field of polymer molecular weight distribution prediction, and provides a non-parametric Bayesian modeling method for Gaussian process regression, aiming at the defects of insufficient interpretation and poor prediction performance effect existing in the existing distributed prediction, and the model can be well interpreted by utilizing prediction uncertainty given by the model, namely providing probability prediction information, and can also be assigned with different kernel functions, such as: the linear kernel, the polynomial kernel, the Gaussian kernel, the exponential kernel and the like can obtain a predicted value with higher accuracy by adopting corresponding kernel functions for different experimental data.
The determination of network topology and generalization capability of neural networks for a given modeling task remains unresolved. Whereas gaussian process regression is greatly superior to traditional methods, both in terms of the result and interpretation of generalization errors.
The method for predicting the molecular weight distribution of the polymer based on Gaussian process regression is characterized in that the polymer monomer is subjected to polymerization reaction in a solvent in the presence of an initiator, and the method for predicting the molecular weight distribution of the polymer in a reaction liquid comprises the following steps:
1) And (3) data acquisition: acquiring all process parameters in the polymerization reaction process by adopting an orthogonal test method, grouping the process parameters according to types, recording corresponding historical data about the molecular weight concentration distribution of the polymer in the reaction liquid under all the process parameters, and establishing a historical process database;
2) And (3) data purification: purifying the historical data about the molecular weight concentration distribution of the polymer recorded in the historical process data sequence library, namely extracting the concentration data of the molecular weight of one polymer according to the difference between every two polymerization degrees, and rejecting the data of the molecular weight concentration of the other polymers to finally obtain a corresponding purified historical process data sequence library;
3) And (3) data processing: normalizing all process parameters, purified polymer molecular weight distribution and concentration distribution in a historical process data sequence database to finally obtain a corresponding historical process data sequence database after data processing;
4) Establishing a polymer molecular weight distribution prediction model: in a historical process data sequence base after data processing, each group of process parameters after normalization processing is used as input characteristics, the data of the molecular weight concentration distribution of the polymer after normalization processing is used as output characteristics, the input characteristics and the output characteristics are imported into a training model, the data of the molecular weight concentration distribution of the polymer corresponding to all known process parameters in the polymerization reaction process are trained and simulated through a Gaussian process distribution model, a prediction model is built, and finally the corresponding relation between the unknown process parameters and the molecular weight concentration distribution of the polymer is obtained.
The method for predicting the molecular weight distribution of the polymer based on Gaussian process regression is characterized in that in the step 2), the numerical value of the difference between the polymerization degrees is 30.
The method for predicting the polymer molecular weight distribution based on Gaussian process regression is characterized in that in the step 1), the grouping type of the process parameters comprises 4 groups, namely the reaction temperature, the reaction time, the concentration of polymer monomers in a solvent and the concentration of an initiator in the solvent.
The method for predicting the polymer molecular weight distribution based on Gaussian process regression is characterized in that in the step 3), all process parameters, the purified polymer molecular weight distribution and the concentration distribution are normalized, and the normalization formulas are as follows:
wherein:representing the result of normalization processing of a certain parameter;
x represents a numerical value before normalization processing of a certain parameter;
x min representing a minimum value of a certain parameter;
x max represents the maximum value of a certain parameter;
after normalization, all values of the process parameters, polymer molecular weight distribution and concentration distribution thereof are floating point numbers between [0,1 ].
The method for predicting the polymer molecular weight distribution based on Gaussian process regression is characterized in that in the step 4), a formula for establishing a prediction model based on Gaussian process regression model is as follows:
y～N(μ(X),K(X,X))
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
n represents a normal distribution (mathematical concept).
The method for predicting the polymer molecular weight distribution based on Gaussian process regression is characterized by further comprising an evaluation process of monitoring a model result: according to the prediction model established in the step 4), the accuracy of the prediction model established in the step 4) is judged by inputting unknown technological parameters to be detected and comparing the fitting condition of the prediction result of the polymer molecular weight concentration distribution based on the prediction model and the detection result of the polymer molecular weight concentration distribution based on a real polymerization reaction experiment;
the prediction formula of the polymer molecular weight concentration distribution based on the prediction model is as follows:
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
X * data representing various sets of process parameters for performing the polymerization reaction in the test data;
f * the concentration distribution value of the molecular weight of the polymer detected and output in the actual polymerization reaction experiment is shown;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
μ(X * ) A data average value of each group of process parameters for polymerization reaction in the test data is represented;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
σ n variance values of data representing the respective sets of process parameters for performing the polymerization reaction;
t represents the transpose of the matrix.
N represents a normal distribution (mathematical concept).
Compared with the prior art, the invention has the following beneficial effects:
the invention relates to a processing mode of polymer molecular weight distribution data and application of a Gaussian process regression model in polymerization reaction, and the combination of the processing mode and the Gaussian process regression model can be used for analyzing the existing polymer molecular weight distribution data and predicting the polymer molecular weight distribution under unknown reaction conditions. The invention considers that the influence of some secondary reaction conditions on the model is slight, so that the molecular weight distribution of the polymer can be predicted more accurately after screening is removed. Can provide a new reference for predicting the molecular weight distribution of the polymer in industrial engineering.
Drawings
FIG. 1 is a graph showing the results of predicting the molecular weight concentration distribution of a polymer using the prediction model established in the present application;
FIG. 2 is a graph showing the results of the molecular weight concentration distribution of the polymer obtained by the actual detection.
Detailed Description
The invention will be further illustrated with reference to specific examples, but the scope of the invention is not limited thereto.
Examples:
a method for predicting the molecular weight distribution of a polymer based on gaussian process regression during polymerization of a polymer monomer in a solvent in the presence of an initiator, comprising the steps of:
1) And (3) data acquisition: all process parameters in the polymerization reaction process are obtained by adopting an orthogonal test method, and the process parameters are grouped according to types, and if the process parameters which mainly influence the reaction result are temperature T, reaction time T, polymer monomer concentration c1 and initiator concentration c2, the process parameters are divided into a group of temperature data, a group of reaction time data, a polymer monomer concentration data and an initiator concentration data. Recording corresponding historical data about the molecular weight concentration distribution of the polymer in the reaction liquid under all process parameters, and establishing a historical process data sequence base;
2) And (3) data purification: and (3) purifying the historical data about the molecular weight concentration distribution of the polymer recorded in the historical process data sequence database, namely extracting the concentration data of the molecular weight of one polymer according to the difference of every other polymerization degree of 30, and rejecting the data of the molecular weight concentration of the other polymers. For example, the polymerization degree is 1,2,3, & gt 3000 total 3000 points, after purification treatment, the polymerization degree is 30, 60, 90, & gt 3000 total 100 points, and the data of the rest points are removed to simplify the data processing process. Finally obtaining a corresponding purified historical process data sequence base;
3) And (3) data processing: normalizing all process parameters, purified polymer molecular weight distribution and concentration distribution in a historical process data sequence database to finally obtain a corresponding historical process data sequence database after data processing;
normalizing all the technological parameters, the molecular weight distribution of the purified polymer and the concentration distribution thereof, wherein the normalization formula is as follows:
wherein:representing the result of normalization processing of a certain parameter;
x represents a numerical value before normalization processing of a certain parameter;
x min representing a minimum value of a certain parameter;
x max represents the maximum value of a certain parameter;
after normalization, all values of the process parameters, polymer molecular weight distribution and concentration distribution thereof are floating point numbers between [0,1 ].
For example, the temperature of the polymerization reaction is in the range of 100-300 ℃, then the temperature data of the polymerization reaction is normalized to be the floating point number between [0,1 ], the original temperature is 150 ℃, and the normalized floating point number is changed to be 0.25.
For example, the molecular weight of the polymer ranges from 100 to 20000, then after normalization, the data of the molecular weight of the polymer is normalized to be a floating point number between [0,1 ], the original molecular weight is 1000, and after normalization, the data is converted to be a floating point number of 0.0452.
4) Establishing a polymer molecular weight distribution prediction model: in a historical process data sequence base after data processing, each group of process parameters after normalization processing is used as input characteristics, the data of the molecular weight concentration distribution of the polymer after normalization processing is used as output characteristics, the input characteristics and the output characteristics are imported into a training model, the data of the molecular weight concentration distribution of the polymer corresponding to all known process parameters in the polymerization reaction process are trained and simulated through a Gaussian process distribution model, a prediction model is built, and finally the corresponding relation between the unknown process parameters and the molecular weight concentration distribution of the polymer is obtained.
The formula for establishing the prediction model based on the Gaussian process regression model is as follows:
y～N(μ(X),K(X,X))
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
n represents a normal distribution (mathematical concept);
5) Monitoring the evaluation process of the model result, and according to the prediction model established in the step 4), comparing the fitting condition of the prediction result of the polymer molecular weight concentration distribution based on the prediction model and the detection result of the polymer molecular weight concentration distribution based on the real polymerization reaction experiment by inputting the unknown process parameters to be detected so as to judge the accuracy of the prediction model established in the step 4);
the prediction formula of the polymer molecular weight concentration distribution based on the prediction model is as follows:
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
X * data representing various sets of process parameters for performing the polymerization reaction in the test data;
f * concentration distribution values representing the molecular weight of the polymer subjected to polymerization reaction detection output in the test data;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
μ(X * ) A data average value of each group of process parameters for polymerization reaction in the test data is represented;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
σ n variance values of data representing the respective sets of process parameters for performing the polymerization reaction;
n represents a normal distribution (mathematical concept);
t represents the transpose of the matrix.
Example 1:
styrene is used as a reaction raw material, petroleum ether is used as a solvent, an initiator is azodiisobutyronitrile AIBN, for example, the initial styrene concentration in a reaction liquid is 9.16mol/L at the experimental condition temperature of 20 ℃, the initial initiator concentration is 5.98mol/L, and after continuous reaction for 4 hours, the molecular weight concentration distribution of a polymer in the reaction liquid is tested.
Using the above formulaf * I.e. the result of predicting the molecular weight concentration distribution of the obtained polymer, as shown in fig. 1.
The results of the molecular weight concentration distribution of the polymer obtained by directly detecting the reaction liquid can be very directly and truly detected, as shown in fig. 2.
Comparing fig. 1 and fig. 2, it can be seen that the fitting degree of the prediction result of the polymer molecular weight concentration distribution based on the prediction model and the detection result of the polymer molecular weight concentration distribution based on the real polymerization reaction experiment is better, the feasibility of the gaussian regression model on the prediction of the polymerization process molecular weight distribution is verified, and a new reference idea can be provided for the prediction of the industrial engineering polymer molecular weight distribution.
What has been described in this specification is merely an enumeration of possible forms of implementation for the inventive concept and may not be considered limiting of the scope of the present invention to the specific forms set forth in the examples.
Claims (6)
1. The polymer molecular weight distribution prediction method based on Gaussian process regression is characterized in that a polymer monomer is subjected to polymerization reaction in a solvent in the presence of an initiator, and the polymer molecular weight distribution prediction method in a reaction liquid in the polymerization reaction process comprises the following steps:
1) And (3) data acquisition: acquiring all process parameters in the polymerization reaction process by adopting an orthogonal test method, grouping the process parameters according to types, recording corresponding historical data about the molecular weight concentration distribution of the polymer in the reaction liquid under all the process parameters, and establishing a historical process database;
2) And (3) data purification: purifying the historical data about the molecular weight concentration distribution of the polymer recorded in the historical process data sequence library, namely extracting the concentration data of the molecular weight of one polymer according to the difference between every two polymerization degrees, and rejecting the data of the molecular weight concentration of the other polymers to finally obtain a corresponding purified historical process data sequence library;
3) And (3) data processing: normalizing all process parameters, purified polymer molecular weight distribution and concentration distribution in a historical process data sequence database to finally obtain a corresponding historical process data sequence database after data processing;
4) Establishing a polymer molecular weight distribution prediction model: in a historical process data sequence base after data processing, each group of process parameters after normalization processing is used as input characteristics, the data of the molecular weight concentration distribution of the polymer after normalization processing is used as output characteristics, the input characteristics and the output characteristics are imported into a training model, the data of the molecular weight concentration distribution of the polymer corresponding to all known process parameters in the polymerization reaction process are trained and simulated through a Gaussian process distribution model, a prediction model is built, and finally the corresponding relation between the unknown process parameters and the molecular weight concentration distribution of the polymer is obtained.
2. The method for predicting molecular weight distribution of polymer based on gaussian process regression according to claim 1, wherein in step 2), the value of the difference between polymerization degrees is 30.
3. The method for predicting molecular weight distribution of polymer based on gaussian process regression according to claim 1, wherein in step 1), the grouping type of the process parameters includes 4 groups of reaction temperature, reaction time, concentration of polymer monomer in solvent and concentration of initiator in solvent, respectively.
4. The method for predicting molecular weight distribution of polymer based on gaussian process regression according to claim 1, wherein in step 3), all process parameters, purified molecular weight distribution of polymer and concentration distribution thereof are normalized, and the formula of normalization is:
wherein:representing the result of normalization processing of a certain parameter;
x represents a numerical value before normalization processing of a certain parameter;
x min representing a minimum value of a certain parameter;
x max represents the maximum value of a certain parameter;
after normalization, all values of the process parameters, polymer molecular weight distribution and concentration distribution thereof are floating point numbers between [0,1 ].
5. The method for predicting molecular weight distribution of polymer based on gaussian process regression according to claim 1, wherein in step 4), the formula for establishing the prediction model based on gaussian process regression model is:
y～N(μ(X),K(X,X))
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
n represents a normal distribution.
6. The method for predicting molecular weight distribution of a polymer based on gaussian process regression according to claim 1, further comprising the step of monitoring the evaluation process of the model results: according to the prediction model established in the step 4), the accuracy of the prediction model established in the step 4) is judged by inputting unknown technological parameters to be detected and comparing the fitting condition of the prediction result of the polymer molecular weight concentration distribution based on the prediction model and the detection result of the polymer molecular weight concentration distribution based on a real polymerization reaction experiment;
the prediction formula of the polymer molecular weight concentration distribution based on the prediction model is as follows:
wherein y represents a concentration distribution value of the molecular weight of the polymer outputted in the training data;
x represents the data of each group of process parameters for polymerization reaction in the training data;
X * indicating that aggregation was performed in the test dataCombining the data of each set of process parameters of the reaction;
f * the concentration distribution value of the molecular weight of the polymer detected and output in the actual polymerization reaction experiment is shown;
μ (X) represents the data mean of the sets of process parameters for which the polymerization reaction is performed in the training data;
μ(X * ) A data average value of each group of process parameters for polymerization reaction in the test data is represented;
k represents covariance matrix of data of each set of process parameters for performing polymerization reaction;
σ n variance values of data representing the respective sets of process parameters for performing the polymerization reaction;
t represents the transpose of the matrix;
n represents a normal distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010680269.4A CN111863151B (en) | 2020-07-15 | 2020-07-15 | Polymer molecular weight distribution prediction method based on Gaussian process regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010680269.4A CN111863151B (en) | 2020-07-15 | 2020-07-15 | Polymer molecular weight distribution prediction method based on Gaussian process regression |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111863151A CN111863151A (en) | 2020-10-30 |
| CN111863151B true CN111863151B (en) | 2024-01-30 |
Family
ID=72983727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010680269.4A Active CN111863151B (en) | 2020-07-15 | 2020-07-15 | Polymer molecular weight distribution prediction method based on Gaussian process regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111863151B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112687355A (en) * | 2020-12-04 | 2021-04-20 | 复旦大学 | Machine learning-assisted polymer synthesis inverse analysis method |
| CN116092585B (en) * | 2023-01-30 | 2024-04-19 | 上海睿璟生物科技有限公司 | Multiple PCR amplification optimization method, system, equipment and medium based on machine learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106971240A (en) * | 2017-03-16 | 2017-07-21 | 河海大学 | The short-term load forecasting method that a kind of variables choice is returned with Gaussian process |
| CN107451101A (en) * | 2017-07-21 | 2017-12-08 | 江南大学 | It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method |
| CN108369659A (en) * | 2015-09-30 | 2018-08-03 | 扎斯特有限公司 | The system and method for entity with destination properties for identification |
| CN108491970A (en) * | 2018-03-19 | 2018-09-04 | 东北大学 | A kind of Predict Model of Air Pollutant Density based on RBF neural |
| CN109492265A (en) * | 2018-10-18 | 2019-03-19 | 南京林业大学 | The kinematic nonlinearity PLS soft-measuring modeling method returned based on Gaussian process |
| CN110032069A (en) * | 2019-04-02 | 2019-07-19 | 东华大学 | A kind of polyester fiber spinning process segmentation parameter configuration method based on error compensation |
-
2020
- 2020-07-15 CN CN202010680269.4A patent/CN111863151B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108369659A (en) * | 2015-09-30 | 2018-08-03 | 扎斯特有限公司 | The system and method for entity with destination properties for identification |
| CN106971240A (en) * | 2017-03-16 | 2017-07-21 | 河海大学 | The short-term load forecasting method that a kind of variables choice is returned with Gaussian process |
| CN107451101A (en) * | 2017-07-21 | 2017-12-08 | 江南大学 | It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method |
| CN108491970A (en) * | 2018-03-19 | 2018-09-04 | 东北大学 | A kind of Predict Model of Air Pollutant Density based on RBF neural |
| CN109492265A (en) * | 2018-10-18 | 2019-03-19 | 南京林业大学 | The kinematic nonlinearity PLS soft-measuring modeling method returned based on Gaussian process |
| CN110032069A (en) * | 2019-04-02 | 2019-07-19 | 东华大学 | A kind of polyester fiber spinning process segmentation parameter configuration method based on error compensation |
Non-Patent Citations (1)
| Title |
|---|
| 基于表界面吸附作用的聚合物基QCM传感器研究;王炉煜;《中国博士学位论文全文数据库 工程科技Ⅰ辑》(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111863151A (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111863151B (en) | Polymer molecular weight distribution prediction method based on Gaussian process regression | |
| CN108875772B (en) | Fault classification model and method based on stacked sparse Gaussian Bernoulli limited Boltzmann machine and reinforcement learning | |
| Guh | Robustness of the neural network based control chart pattern recognition system to non‐normality | |
| CN110569566B (en) | Method for predicting mechanical property of plate strip | |
| CN108829878B (en) | Method and device for detecting abnormal points of industrial experimental data | |
| Yeganeh et al. | Monitoring linear profiles using Artificial Neural Networks with run rules | |
| CN105678409A (en) | Adaptive and distribution-free time series abnormal point detection method | |
| CN111639304B (en) | CSTR fault positioning method based on Xgboost regression model | |
| CN104915679A (en) | Large-scale high-dimensional data classification method based on random forest weighted distance | |
| CN117273551B (en) | Intelligent production management method, system and storage medium for modified asphalt | |
| Yu et al. | A framework to identify and respond to weak signals of disastrous process incidents based on FRAM and machine learning techniques | |
| CN111651932A (en) | Online dynamic security assessment method for power system based on integrated classification model | |
| CN110808947B (en) | Automatic vulnerability quantitative evaluation method and system | |
| CN110057588B (en) | Bearing early fault detection and diagnosis method and system based on fusion of singular value and graph theory characteristics | |
| CN108920428B (en) | Fuzzy distance discrimination method based on joint fuzzy expansion principle | |
| CN108053093A (en) | A kind of k- neighbour's method for diagnosing faults based on the conversion of average influence Value Data | |
| CN109767430B (en) | Quality detection method and quality detection system for valuable bills | |
| KR101748013B1 (en) | Method for evaluating the an identity of a copolymer and system using the same | |
| CN111797574A (en) | Integrated Gaussian process regression model method for polymer molecular weight distribution | |
| Manavazhahan | A study of activation functions for neural networks | |
| CN114118680A (en) | Network security situation assessment method and system | |
| CN114660931A (en) | Method and system for diagnosing and identifying petrochemical process fault | |
| Olsen et al. | Predicting dactyloscopic examiner fingerprint image quality assessments | |
| CN115580445B (en) | Unknown attack intrusion detection method, unknown attack intrusion detection device and computer readable storage medium | |
| Teodorescu et al. | Analysis of the predictability of time series obtained from genomic sequences by using several predictors |
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 |