CN104569035A - Method for acquiring critical property parameters of coal liquefaction oil - Google Patents
Method for acquiring critical property parameters of coal liquefaction oil Download PDFInfo
- Publication number
- CN104569035A CN104569035A CN201510058745.8A CN201510058745A CN104569035A CN 104569035 A CN104569035 A CN 104569035A CN 201510058745 A CN201510058745 A CN 201510058745A CN 104569035 A CN104569035 A CN 104569035A
- Authority
- CN
- China
- Prior art keywords
- neural network
- close
- boiling
- liquefied coal
- critical
- 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
Landscapes
- Production Of Liquid Hydrocarbon Mixture For Refining Petroleum (AREA)
Abstract
The invention discloses a method for acquiring critical property parameters of coal liquefaction oil. The method comprises the following steps: detecting narrow fractions of the coal liquefaction oil so as to acquire boiling point and group composition of the narrow fractions; and inputting the boiling point and group composition of the narrow fractions into a trained neural network model so as to acquire numerical values of multiple critical parameters of the narrow fractions, wherein the neural network model takes the group composition of the coal liquefaction oil and the boiling point of the narrow fractions of the coal liquefaction oil as the input sequence, and the multiple critical parameters of the narrow fractions are taken as the output sequence. Compared with a single group contribution method, the method disclosed by the invention has the advantages that the interaction among the various groups in organic matters is considered, and the calculation accuracy is improved.
Description
Technical field
The present invention relates to coal liquefaction field, particularly relate to a kind of acquisition methods for liquefied coal coil critical properties parameter.
Background technology
Deficient at petroleum resources, under the severe situation of liquid fuel and industrial chemicals shortage, the exploitation of coal liquefaction craft and popularization are reply oil crisis and the effective measures improving energy security guarantee.Although the exploitation of coal liquefaction craft enters industrialization phase at present substantially, in the various Basic Physical Properties Data of liquefied coal coil component and flowsheeting, the shortage of key parameter, hinders the development of coal liquefaction craft to a certain extent.
Thermodynamic parameter is not only the master data of Study on Coal liquefaction oil character, is also the foundation that liquefaction device design and amplification, coal liquefaction craft whole process are simulated.Wherein, critical properties parameter, as the important macroscopic property of liquefied coal coil, is the basis of other physical datas of association.
In coal liquefaction technology, because liquefied coal coil component is very complicated, be difficult to directly obtain its macroscopic property.So adopt is cut to multiple close-boiling cut at present more, uses for reference the ripe macroscopic property of petroleum analysis method to close-boiling cut and study, obtain the macroscopic property of liquefied coal coil.Critical parameters are generally just meaningful to single component pure compound, even it is also multicomponent mixture that liquefied coal coil cuts into close-boiling cut, its critical parameters can not be measured with the assay method of applicable single component pure compound, the critical parameters of liquefaction oil close-boiling cut be in order to Study on Coal liquefaction oil macroscopic property between relation and artificially suppose, so be called pseudo-critical parameter.But these parameters are very important to other macroscopic properties of calculating, so need to use for reference current existing research means, it is analyzed and researched.
From the current document about critical parameters research, various evaluation method can be divided into: direct measuring method, Group Contribution Method and Empirical Equation method three types.Wherein, Group Contribution Method is identical by the contribution of same group in supposition different molecular, the group organic character being regarded as constitute to the contribution of this character add and, then by methods such as Mathematical Fittings, obtain the quantitative relationship between this character and molecular radical, thus reach the function of prediction.Owing to substantially not relying on other physical property in the estimation process of the method, so more at the application in calculation of critical properties parameter.
But still there is certain defect in Group Contribution Method, such as: Group Contribution Method only considers the contribution of group in molecule, and does not consider the interaction between group, between chemical bond on the organic critical parameters of prediction; On the other hand, Group Contribution Method is relatively poor to the separating capacity of isomers, only cannot distinguish its structural difference with group at all.
The critical properties of the researchists such as the Feng Jie each close-boiling cut of coal direct liquefaction oil that adopted Group Contribution Method to calculate, but consider that the interaction of each group key in each close-boiling cut is nonlinearity, the interaction power of each group key is obtained according to mathematical regression method simple in Group Contribution Method, same group key can not be reflected well in different molecular to the difference of physico-chemical property Numerical Contributions, make precision of prediction lower, and range of application is restricted.
Therefore, need to provide applicable liquefied coal coil system and more accurately and effectively method to obtain its critical properties.
Summary of the invention
The object of the present invention is to provide a kind of acquisition methods for liquefied coal coil critical parameters, to make up the deficiency of Group Contribution Method.
For this reason, the invention provides a kind of acquisition methods for liquefied coal coil critical parameters, comprise the following steps: the close-boiling cut of liquefied coal coil is detected, to obtain boiling point and the group composition of close-boiling cut; And the boiling point of close-boiling cut and group composition are input in housebroken neural network model, to obtain the numerical value of multiple critical parameters of close-boiling cut, wherein, neural network model using the boiling point of close-boiling cut of the group of liquefied coal coil composition and liquefied coal coil as list entries, and using multiple critical parameters of close-boiling cut as output sequence.
Further, above-mentioned multiple critical parameters are critical temperature, emergent pressure and critical volume.
Further, the training of above-mentioned neural network model comprises the following steps: liquefied coal coil is divided into multiple close-boiling cut; Detect boiling point and the group composition of each close-boiling cut in multiple close-boiling cut, and obtain the numerical value of multiple critical parameters of each close-boiling cut; Using a part of close-boiling cut in multiple close-boiling cut as training sample, neural network is trained; Deconditioning after neural network convergence, to obtain neural network model.
Further, the acquisition methods that the training of above-mentioned neural network model also comprises for liquefied coal coil critical parameters also comprises the another part in multiple close-boiling cut as test samples, the boiling point of test samples and group composition are input in the neural network model after utilizing training sample to train, measure the predicated error of neural network model according to the output of neural network model.
Further, multiple critical parameters of above-mentioned training sample comprise critical temperature, emergent pressure and critical volume, and the critical temperature of training sample, emergent pressure and critical volume are obtained by following formula respectively:
T
c＝18.293T
b 0.59525ρ
0.34742；
P
c＝0.29515×10
7T
b -2.2082ρ
2.2209；
Vc=0.82238 × 10
-4t
b 2.5111ρ
-1.6221, wherein, T
cfor critical temperature, P
cfor emergent pressure, V
cfor critical volume,
T
bfor boiling temperature, ρ is the density at 20 DEG C.
Further, above-mentioned neural network is BP neural network, comprises input layer, hidden layer and output layer.
Further, the radical amount of above-mentioned liquefied coal coil is 40, and critical parameters are 3, and the nodes of the input layer of BP neural network is 41, and the nodes of output layer is 3.
Further, above-mentioned hidden layer is one deck, and the nodes of hidden layer is 20.
Further, above-mentioned each close-boiling cut and corresponding boiling point utilize true boiling point distillation device to carry out true boiling point distillation acquisition to liquefied coal coil.
Further, the group composition of above-mentioned close-boiling cut utilizes gas chromatograph-mass spectrometer (GCMS) to record.
The method that the present invention proposes artificial neural network and Group Contribution Method to combine calculates the critical properties of liquefied coal coil, and the method has the following advantages and good effect:
1) compared with single Group Contribution Method, the method considers the reciprocation in organism between each group, improves the accuracy of calculating;
2) neural network has flexible structure, and according to the situation of the group that the sample that will predict comprises, its input layer, middle layer and output layer nodes can regulate easily; Training sample set also can carry out autotelic increase and decrease, and can increase more reliable data and constantly improve forecasting reliability by constantly screening;
3) in traditional group contribution approach, fashionable when there being new group to add, need again to return correction group contribution scale, make calculating more loaded down with trivial details.And can to increase new group in neural network be input node, expand the estimated range of network.
Except object described above, feature and advantage, other object, feature and advantage that the present invention has, will be described in further detail by reference to the accompanying drawings.
Accompanying drawing explanation
Forming the part of this instructions, showing the preferred embodiments of the present invention for understanding accompanying drawing of the present invention further, and be used for principle of the present invention is described together with instructions.In figure:
Fig. 1 is the schematic diagram according to the neural network model for liquefied coal coil critical parameters of the present invention;
Fig. 2 is the process flow diagram according to the acquisition methods for liquefied coal coil critical parameters of the present invention;
Fig. 3 is the process flow diagram of the acquisition methods according to the neural network model for liquefied coal coil critical parameters of the present invention; And
Fig. 4 is the process flow diagram of the error of calculation assay method according to the neural network model for liquefied coal coil critical parameters of the present invention.
Embodiment
Below in conjunction with accompanying drawing, embodiments of the invention are described in detail, but the multitude of different ways that the present invention can be defined by the claims and cover is implemented.
The present invention adopts Group Contribution Method most widely used in the calculating of organism critical properties, utilizes neural network, trains surveying the group composition of close-boiling cut, boiling point and critical properties parameter as training sample input value neural network.When neural network is after training up, obtain neural network model, utilize this neural network model, by boiling point, the group composition of incoming inspection sample, the critical properties parameter of test samples can be calculated.
Preferably, the present invention utilizes BP neural network for the Prediction of critical property of liquefied coal coil close-boiling cut on the basis of Group Contribution Method.This BP neural network is the artificial neural network based on error back propagation (Back Propagation, BP) algorithm.
Fig. 1 is the schematic diagram according to the neural network model for liquefied coal coil critical parameters of the present invention.As shown in Figure 1, BP neural network model comprises three-layer neural network structure, i.e. input layer, hidden layer and output layer.Wherein, input layer accepts exogenous data input, and hidden layer processes input data and changes, and output layer is produce output result then.
Every one deck in neural network all comprises some neurons (also referred to as node), and the neuron number of its input layer and output layer is determined by the variable in model.
In the present invention, the variable of input layer is atmospheric boiling point Tb and the different groups composition of close-boiling cut.Group composition comprises types of radicals and this double implication of content thereof, the types of radicals of the input layer of neural network is the types of radicals that all training samples comprise, the corresponding node of each types of radicals, the N number of types of radicals comprised according to all close-boiling cuts inputs as the group of neural network, add boiling point Tb, altogether N+1 input node.
To each close-boiling cut, neural network be input as comprised group content, the group do not comprised is input as zero.
The variable of the output layer of neural network is emergent pressure Pc, critical temperature Tc, critical volume Vc totally 3 nodes of close-boiling cut.
Neuron number and the number of nodes of hidden layer are rule of thumb chosen, and the excitation function of hidden layer and output layer may be selected to be conventional linear function or nonlinear function.In addition, according to the convergence of result, excitation function can be reselected in computation process.Wherein, in conventional excitation function, S type excitation function: f (x)=1/ (1+exp (-x)) comparatively mates with the present invention.
Fig. 2 is the process flow diagram according to the acquisition methods for liquefied coal coil critical parameters of the present invention.As shown in Figure 2, this acquisition methods comprises the following steps: S12, detect the close-boiling cut of liquefied coal coil, to obtain boiling point and the group composition of close-boiling cut; S14, by the boiling point of close-boiling cut and group composition input neural network model, wherein, neural network model using the boiling point of close-boiling cut of the group of liquefied coal coil composition and liquefied coal coil as list entries, and using multiple critical parameters of close-boiling cut as output sequence; The numerical value of multiple critical parameters of S16, output close-boiling cut.
In the present invention, neural network model obtains after being through training.First select a certain amount of close-boiling cut as the training sample of neural network, preanalysis is carried out to its critical parameters, such as, to be calculated by correlation or instrumental analysis obtains, thus the list entries obtained for neural metwork training and output sequence.After determining the structure of network model, the list entries utilizing training sample to obtain and output sequence neural network training, by test after convergence, thus obtain the neural network model can predicted the critical properties of new close-boiling cut.
Fig. 3 is the process flow diagram of the acquisition methods according to the neural network model for liquefied coal coil critical parameters of the present invention.As shown in Figure 3, the acquisition methods of neural network model comprises the following steps: S22, liquefied coal coil is divided into multiple close-boiling cut; The boiling point of S24, each close-boiling cut detected in multiple close-boiling cut and group composition, and obtain the numerical value of multiple critical parameters of each close-boiling cut; S26, using a part of close-boiling cut in multiple close-boiling cut as training sample, neural network is trained; S28, neural network convergence after deconditioning, to obtain neural network model.
In step S26, be to adjust weight factor (or being called coefficient) to the training of neural network, until the output mode obtained for given input signal or output valve and desired result match.Concrete instruction is as follows: (1) initialization, namely arranges each layer weight coefficient and value at random; (2) training sample data are added to network input, calculate the output valve of each layer, output valve is obtained error signal compared with expectation value; (3) connection weight is readjusted according to error signal; (4) if be less than predictive error, then think that network is restrained and deconditioning, otherwise then return continuation training.
In addition, also select some close-boiling cuts for measuring the error of calculation of neural network model.
Fig. 4 is the process flow diagram of the error determine method according to the neural network model for liquefied coal coil critical parameters of the present invention.As shown in Figure 4, this error determine method comprises the following steps: S32, using test samples boiling point and group composition as list entries input neural network model; The critical parameters of S34, output test samples; S36, the numerical value of multiple critical parameters exported by neural network model compare with the numerical value of multiple critical parameters of the test samples to obtain in advance, to measure the error of calculation of neural network model.
By with upper type, can obtain the error of calculation of neural network model, this error of calculation can be used for the reliability assessing this neural network model and result of calculation thereof.
Application example
Step one: Data Collection
1) adopt true boiling point distillation device to carry out close-boiling cut segmentation to coal liquefaction products, obtain 17 close-boiling cuts, wherein 13 close-boiling cuts are defined as training sample, and all the other 4 is test samples; Preferably, according to boiling point order arrangement from low to high, between adjacent two detection samples, be separated with training sample.
2) adopt gas chromatograph-mass spectrometer (GCMS) (GC-MS) to detect group composition in training sample, obtain group and the content thereof of 40 types altogether.
3) densitometer and vapour pressure permeameter is adopted to measure density and the molecular weight of 13 training samples, then by the critical parameters of the Riazi-Daubert Empirical Equation calculation training sample of improvement respectively.
The Riazi-Daubert Empirical Equation improved:
Critical temperature: T
c=18.293T
b 0.59525ρ
0.34742---------(1)
Emergent pressure: P
c=0.29515 × 10
7t
b -2.2082ρ
2.2209--------(2)
Critical volume: Vc=0.82238 × 10
-4t
b 2.5111ρ
-1.6221------------(3)
Wherein, T
bfor atmospheric boiling point, ρ is the density at 20 DEG C, g/cm3.
Step 2: neural metwork training
1) input layer: input value is the mean boiling point value of training sample and 40 group numerical value in this sample, and wherein, group A input value is the content value of A in GC-MS, if content is zero, input value is zero.
2) output layer is adopt the emergent pressure of Riazi-Daubert Empirical Equation calculate 13 the training samples improved, critical temperature, critical volume.
3) hidden layer is single layer structure, and its neuronal quantity is determined by try and error method, is finally defined as the network structure of 41 × 20 × 3.By repeatedly training, carry out deconditioning by adjustment root-mean-square error value.
Step 3: models applying
Utilize and obtain neural network the boiling point of other 4 test samples, group composition data are calculated, predict its critical properties, i.e. emergent pressure, critical temperature and critical volume.Result display is compared with the numerical value adopting the Riazi-Daubert Empirical Equation improved to obtain, and relative error is 3.5%.
The method that the present invention proposes artificial neural network and Group Contribution Method to combine calculates the critical properties of liquefied coal coil, and the method has the following advantages and good effect:
1) compared with single Group Contribution Method, the method considers the reciprocation in organism between each group, improves the accuracy of calculating.
2) artificial neural network has flexible structure, and according to the situation of the group that the sample that will predict comprises, its input layer, middle layer and output layer nodes can regulate easily; Training sample set also can carry out autotelic increase and decrease, and can increase more reliable data and constantly improve forecasting reliability by constantly screening.
3) in traditional group contribution approach, fashionable when there being new group to add, need again to return correction group contribution scale, make calculating more loaded down with trivial details.And can to increase new group at neural network algorithm be input node, expand the estimated range of network.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. for an acquisition methods for liquefied coal coil critical parameters, it is characterized in that, comprise the following steps:
The close-boiling cut of liquefied coal coil is detected, to obtain boiling point and the group composition of described close-boiling cut; And
The boiling point of described close-boiling cut and group composition are input in housebroken neural network model, to obtain the numerical value of multiple critical parameters of described close-boiling cut,
Wherein, described neural network model using the boiling point of close-boiling cut of the group of liquefied coal coil composition and described liquefied coal coil as list entries, and using multiple critical parameters of described close-boiling cut as output sequence.
2. the acquisition methods for liquefied coal coil critical parameters according to claim 1, is characterized in that, described multiple critical parameters are critical temperature, emergent pressure and critical volume.
3. the acquisition methods for liquefied coal coil critical parameters according to claim 1, is characterized in that, the training of described neural network model comprises the following steps:
Liquefied coal coil is divided into multiple close-boiling cut;
Detect boiling point and the group composition of each close-boiling cut in described multiple close-boiling cut, and obtain the numerical value of multiple critical parameters of described each close-boiling cut;
Using a part of close-boiling cut in described multiple close-boiling cut as training sample, neural network is trained;
Deconditioning after described neural network convergence, to obtain described neural network model.
4. the acquisition methods for liquefied coal coil critical parameters according to claim 3, it is characterized in that, the training of described neural network model also comprises the another part in described multiple close-boiling cut as test samples, the boiling point of described test samples and group composition are input in the described neural network model after utilizing described training sample to train, measure the predicated error of described neural network model according to the output of described neural network model.
5. the acquisition methods for liquefied coal coil critical parameters according to claim 3, it is characterized in that, multiple critical parameters of described training sample comprise critical temperature, emergent pressure and critical volume, and the described critical temperature of described training sample, emergent pressure and critical volume are obtained by following formula respectively:
T
c＝18.293T
b 0.59525ρ
0.34742；
P
c＝0.29515×10
7T
b -2.2082ρ
2.2209；
Vc=0.82238 × 10
-4t
b 2.5111ρ
-1.6221, wherein, T
cfor critical temperature, P
cfor emergent pressure, V
cfor critical volume,
T
bfor boiling temperature, ρ is the density at 20 DEG C.
6. the acquisition methods for liquefied coal coil critical parameters according to claim 3, is characterized in that, described neural network is BP neural network, comprises input layer, hidden layer and output layer.
7. the acquisition methods for liquefied coal coil critical parameters according to claim 6, it is characterized in that, the radical amount of described liquefied coal coil is 40, and described critical parameters are 3, the nodes of the input layer of described BP neural network is 41, and the nodes of described output layer is 3.
8. the acquisition methods for liquefied coal coil critical parameters according to claim 6, is characterized in that, described hidden layer is one deck, and the nodes of described hidden layer is 20.
9. the acquisition methods for liquefied coal coil critical parameters according to any one of claim 1 to 3, is characterized in that, described each close-boiling cut and corresponding boiling point utilize true boiling point distillation device to carry out true boiling point distillation acquisition to liquefied coal coil.
10. the acquisition methods for liquefied coal coil critical parameters according to any one of claim 1 to 3, is characterized in that, the group composition of described close-boiling cut utilizes gas chromatograph-mass spectrometer (GCMS) to record.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510058745.8A CN104569035A (en) | 2015-02-04 | 2015-02-04 | Method for acquiring critical property parameters of coal liquefaction oil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510058745.8A CN104569035A (en) | 2015-02-04 | 2015-02-04 | Method for acquiring critical property parameters of coal liquefaction oil |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104569035A true CN104569035A (en) | 2015-04-29 |
Family
ID=53085598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510058745.8A Pending CN104569035A (en) | 2015-02-04 | 2015-02-04 | Method for acquiring critical property parameters of coal liquefaction oil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104569035A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407635A (en) * | 2015-07-31 | 2017-02-15 | 中国石油化工股份有限公司 | Method for calculating boiling point of hydrocarbon compound |
| CN111223519A (en) * | 2018-11-23 | 2020-06-02 | 中国石油化工股份有限公司 | System for determining basic physical property parameters of organic matters and application thereof |
| CN111899814A (en) * | 2020-06-12 | 2020-11-06 | 中国石油天然气股份有限公司 | Method, equipment and storage medium for calculating physical properties of single molecule and mixture |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11265197A (en) * | 1997-12-13 | 1999-09-28 | Hyundai Electronics Ind Co Ltd | Voice recognizing method utilizing variable input neural network |
| CN103559542A (en) * | 2013-10-31 | 2014-02-05 | 华北水利水电大学 | Extension neural network pattern recognition method based on priori knowledge |
| CN104008751A (en) * | 2014-06-18 | 2014-08-27 | 周婷婷 | Speaker recognition method based on BP neural network |
-
2015
- 2015-02-04 CN CN201510058745.8A patent/CN104569035A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11265197A (en) * | 1997-12-13 | 1999-09-28 | Hyundai Electronics Ind Co Ltd | Voice recognizing method utilizing variable input neural network |
| CN103559542A (en) * | 2013-10-31 | 2014-02-05 | 华北水利水电大学 | Extension neural network pattern recognition method based on priori knowledge |
| CN104008751A (en) * | 2014-06-18 | 2014-08-27 | 周婷婷 | Speaker recognition method based on BP neural network |
Non-Patent Citations (6)
| Title |
|---|
| 周山花等: "人工神经网络在石油分析中的应用研究（I）——BP神经网络预测石油馏分临界性质", 《石油化工高等学校学报》 * |
| 姜慧婧: "塔里木馏分油物性的预测研究", 《中国石油大学硕士学位论文》 * |
| 徐振刚等: "《煤化工技术理论与实践》", 30 November 2009 * |
| 旷戈等: "人工神经网络基团贡献法估算纯有机物的临界参数", 《计算机与应用化学》 * |
| 朱肖曼: "煤液化油基本性质的研究", 《煤炭科学研究总院硕士硕士学位论文》 * |
| 李立权: "《加氢裂化装置工艺计算与技术分析》", 30 September 2009 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407635A (en) * | 2015-07-31 | 2017-02-15 | 中国石油化工股份有限公司 | Method for calculating boiling point of hydrocarbon compound |
| CN106407635B (en) * | 2015-07-31 | 2019-03-08 | 中国石油化工股份有限公司 | A method of calculating hydrocarbon compound boiling point |
| CN111223519A (en) * | 2018-11-23 | 2020-06-02 | 中国石油化工股份有限公司 | System for determining basic physical property parameters of organic matters and application thereof |
| CN111223519B (en) * | 2018-11-23 | 2023-06-13 | 中国石油化工股份有限公司 | System for determining basic physical property parameters of organic matters and application thereof |
| CN111899814A (en) * | 2020-06-12 | 2020-11-06 | 中国石油天然气股份有限公司 | Method, equipment and storage medium for calculating physical properties of single molecule and mixture |
| CN111899814B (en) * | 2020-06-12 | 2024-05-28 | 中国石油天然气股份有限公司 | Single molecule and mixture physical property calculation method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105205224B (en) | Time difference Gaussian process based on fuzzy curve analysis returns soft-measuring modeling method | |
| Han et al. | A hybrid PSO-SVM-based model for determination of oil recovery factor in the low-permeability reservoir | |
| Pang et al. | Evaluation method and application of the relative contribution of marine hydrocarbon source rocks in the Tarim Basin: a case study from the Tazhong area | |
| CN101266299B (en) | Method for forecasting oil gas utilizing earthquake data object constructional features | |
| CN109060702B (en) | Infrared spectrum nonlinear quantitative analysis method | |
| Goel et al. | Prediction of API values of crude oils by use of saturates/aromatics/resins/asphaltenes analysis: computational-intelligence-based models | |
| CN104569035A (en) | Method for acquiring critical property parameters of coal liquefaction oil | |
| CN104376434A (en) | Comprehensive evaluation method for multipoint processing quality of cigarettes | |
| CN103760307B (en) | Method for evaluating type of kerogen | |
| Ping et al. | Thermodynamic modeling of petroleum inclusions: the prediction of the saturation pressure of crude oils | |
| Cai et al. | On-line multi-gas component measurement in the mud logging process based on Raman spectroscopy combined with a CNN-LSTM-AM hybrid model | |
| Liu et al. | TOC determination of Zhangjiatan shale of Yanchang formation, Ordos Basin, China, using support vector regression and well logs | |
| Samodurov | Complimentarity of different approaches for assessing vertical turbulent exchange intensity in natural stratified basins | |
| CN103984856A (en) | System and method for predicting longitudinal crack width of wall body based on neural network | |
| CN109655394B (en) | Nuclear magnetic resonance T2 spectrum permeability calculation method under constraint of throat structure parameters | |
| CN116856905A (en) | Crude oil density real-time quantitative prediction method based on multicomponent gas detection system | |
| CN102539370B (en) | Filtering method for fourier transform infrared spectrum online analysis of multi-component gas | |
| CN118095607A (en) | Method and device for determining distribution range of exploration target | |
| Tan et al. | Research on measurement model of water content of oil well based on data fusion | |
| da Silva et al. | Evaluating the Splintex model for estimating the soil water retention curve for a wide range of soils | |
| Duan et al. | Modeling the characterization of the plus fractions by using continuous distribution function | |
| Niu et al. | Developing novel ensemble models for predicting soil hydraulic properties in China’s arid region | |
| CN106970039A (en) | A kind of pipeline blending quantitative detecting method based on terahertz time-domain spectroscopy | |
| Zhang et al. | Calibration method of multi-parameter compensation for optical dissolved oxygen sensor in seawater based on machine learning algorithm | |
| CN116067905A (en) | Method and apparatus for detecting oil blending |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150429 |
|
| RJ01 | Rejection of invention patent application after publication |