CN117894386A - Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil - Google Patents
Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil Download PDFInfo
- Publication number
- CN117894386A CN117894386A CN202211225595.1A CN202211225595A CN117894386A CN 117894386 A CN117894386 A CN 117894386A CN 202211225595 A CN202211225595 A CN 202211225595A CN 117894386 A CN117894386 A CN 117894386A
- Authority
- CN
- China
- Prior art keywords
- crude oil
- near infrared
- mixed
- mixed crude
- detected
- 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
- 239000010779 crude oil Substances 0.000 title claims abstract description 235
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 72
- 238000004458 analytical method Methods 0.000 claims description 20
- 238000013145 classification model Methods 0.000 claims description 19
- 239000000203 mixture Substances 0.000 claims description 17
- 238000011176 pooling Methods 0.000 claims description 8
- 238000002835 absorbance Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 239000000126 substance Substances 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 abstract description 2
- 239000004576 sand Substances 0.000 description 20
- 239000011159 matrix material Substances 0.000 description 12
- 238000004891 communication Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 8
- 238000002156 mixing Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 239000003921 oil Substances 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 description 3
- 229910052717 sulfur Inorganic materials 0.000 description 3
- 239000011593 sulfur Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical group [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- ZAMOUSCENKQFHK-UHFFFAOYSA-N Chlorine atom Chemical compound [Cl] ZAMOUSCENKQFHK-UHFFFAOYSA-N 0.000 description 1
- 238000004566 IR spectroscopy Methods 0.000 description 1
- 238000004497 NIR spectroscopy Methods 0.000 description 1
- 241000282376 Panthera tigris Species 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000000460 chlorine Substances 0.000 description 1
- 229910052801 chlorine Inorganic materials 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The embodiment of the invention provides a method, a system and a readable medium for predicting components and physicochemical properties of mixed crude oil, which belong to the technical field of near infrared modeling data processing, wherein the method for predicting comprises the steps of obtaining near infrared spectrums of mixed crude oil to be detected; inputting the near infrared spectrum of the mixed crude oil to be detected into a pre-constructed prediction model to obtain the components and the proportions of single crude oil in the mixed crude oil to be detected; according to the components, proportions and physicochemical properties of the single-variety crude oil, the physicochemical properties of the mixed crude oil to be detected are generated by fitting, and according to the technical scheme, the components and proportions of the single-variety crude oil in the mixed crude oil to be detected are obtained based on the physicochemical properties of the single-variety crude oil and a pre-constructed prediction model, so that the physicochemical properties of the mixed crude oil to be detected are generated by fitting, and the prediction of the components and the physicochemical properties of the mixed crude oil can be realized by carrying out near infrared spectrum measurement on the mixed crude oil once, and the operation is convenient, and the prediction is quick and accurate.
Description
Technical Field
The invention relates to the technical field of near infrared modeling data processing, in particular to a method, a system and a readable medium for predicting the components and physicochemical properties of mixed crude oil.
Background
In recent years, refining enterprises in China face the problem of poor quality and heavy quality of crude oil. Crude oil blending has become an indispensable means in order to ensure the relative stability of crude oil properties. In the crude oil blending process, how to accurately and rapidly obtain the properties of the mixed crude oil is a main bottleneck for restricting the crude oil blending effect.
With the gradual maturity of near infrared spectrum technology and the progress of computer science, crude oil evaluation based on near infrared spectrum analysis technology has the characteristics of simple operation, small sample consumption, high degree of automation, short evaluation period, accurate property and the like, and can realize rapid evaluation of crude oil. The data base of the near infrared spectrum analysis is standard test analysis data of a crude oil sample, single crude oil is generally used as a standard test analysis object, however, the crude oil evaluation object based on the near infrared spectrum analysis technology is mixed crude oil after crude oil blending, the near infrared spectrum analysis object is not matched with the data base, and the problem that the available sample amount of near infrared spectrum analysis modeling is small, so that the analysis precision is affected. Therefore, research and development of a mixed crude oil analysis method based on near infrared spectrum technology realizes real-time analysis of the component content of single crude oil in mixed crude oil and the physicochemical properties of the mixed crude oil, and it is necessary to improve the analysis speed and accuracy of the mixed crude oil and the economic benefit of refineries.
Disclosure of Invention
It is an object of embodiments of the present invention to provide a method, system and readable medium for predicting the composition and physicochemical properties of a blended crude oil using near infrared spectroscopy of the blended crude oil, which is used to improve the above-mentioned problems in the prior art of predicting the composition and physicochemical properties of the blended crude oil.
In order to achieve the above object, the present invention provides a method for predicting a mixed crude oil component and physicochemical properties, comprising the steps of:
acquiring a near infrared spectrum of mixed crude oil to be detected;
inputting the near infrared spectrum of the mixed crude oil to be detected into a pre-constructed prediction model to obtain the components and the proportions of the single-variety crude oil in the mixed crude oil to be detected, wherein the prediction model comprises a classification model for predicting the single-variety crude oil component types in the mixed crude oil to be detected and a regression model for predicting the single-variety crude oil component proportions in the mixed crude oil to be detected;
obtaining physical and chemical properties of single crude oil in the mixed crude oil to be tested by using a standard experimental analysis method;
fitting to generate physicochemical properties of the mixed crude oil to be tested according to the components, the proportion and the physicochemical properties of the single crude oil.
Optionally, after the obtaining the near infrared spectrum of the mixed crude oil to be tested, the method further includes:
performing first derivative treatment on the near infrared spectrum of the mixed crude oil to be detected, wherein the extraction wavelength band is 4713-5345cm -1 And 5978-6611cm -1 And inputting the principal component index of the near infrared spectrum of the interval into the prediction model.
Optionally, before the near infrared spectrum of the mixed crude oil to be tested is input into a pre-constructed prediction model, the method further comprises:
respectively obtaining near infrared spectrums of the single-variety crude oils, and obtaining physicochemical properties of the single-variety crude oils by using a standard experimental analysis method;
setting types and proportions of single-variety crude oil contained in the simulated mixed crude oil, and adding absorbance vectors of near infrared spectrums of the single-variety crude oil according to the types and proportions to form a near infrared spectrum of the simulated mixed crude oil;
and extracting main component indexes in the near infrared spectrum of the simulated mixed crude oil, and establishing the prediction model.
Optionally, after the forming the near infrared spectrum simulating the mixed crude oil, the method further comprises:
performing first derivative treatment on near infrared spectrum of simulated mixed crude oil, and selecting wavelength band of 4713-5345cm -1 And 5978-6611cm -1 Near infrared spectrum of the interval.
Optionally, the classification model is a one-dimensional convolutional neural network classification model, and the classification model is established by the following operations:
leading the main component index into a one-dimensional convolution input layer, passing through two convolution layers, carrying out weight initialization and activation functions, passing through a pooling layer, and transmitting to an output layer through multiple convolution layers and pooling layers to obtain a type prediction value;
and comparing the type predicted value obtained by the output layer with a sample expected value, and if an error exists, adjusting the weight until the error between the type predicted value and the sample expected value reaches the minimum or minimum value, so as to obtain the classification model.
Optionally, the regression model is constructed according to any one of multiple linear regression, partial least squares method and artificial neural network regression modeling method, and the regression model is optimized by:
comparing the component proportion of the single-variety crude oil in the simulated mixed crude oil predicted by the regression model with the set component proportion of the single-variety crude oil to obtain a deviation result;
optimizing the regression model based on the bias results.
Optionally, the fitting to generate the physicochemical property of the mixed crude oil to be tested according to the components and the proportion of the single crude oil and the physicochemical property of the single crude oil comprises:
for physicochemical properties to be predicted that satisfy the linear condition, performing a linear fitting calculation:
a mix ＝w 1 a 1 +w 2 a 2 +…+w n a n
wherein a is mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, a n Is the physicochemical property to be predicted corresponding to the nth single variety crude oil in the mixed crude oil to be detected; and
for physicochemical properties to be predicted that satisfy the nonlinear condition, performing nonlinear fitting calculation:
wherein b mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, b n Is the physicochemical property to be predicted corresponding to the nth single crude oil in the mixed crude oil to be detected.
Accordingly, another embodiment of the present invention also provides a system for predicting physicochemical properties of a mixed crude oil component, comprising:
a memory for storing instructions executable by the processor;
and the processor is used for executing the instructions to realize the method.
In another aspect, the invention also provides a computer readable medium having stored thereon computer instructions, wherein the method described above is performed when the computer instructions are executed by a processor.
According to the technical scheme, the near infrared spectrum of the mixed crude oil to be detected is input into the pre-built prediction model, the components, the proportion and the physicochemical properties of the single-variety crude oil in the mixed crude oil to be detected are obtained, and the physicochemical properties of the mixed crude oil to be detected are generated by fitting according to the components, the proportion and the physicochemical properties of the single-variety crude oil, so that the prediction of the components and the physicochemical properties of the mixed crude oil can be realized by carrying out near infrared spectrum measurement on the mixed crude oil once, and the operation is convenient, and the prediction is quick and accurate.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain, without limitation, the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart of a method for predicting the physical and chemical properties of a blended crude oil composition in accordance with an embodiment of the present invention;
FIGS. 2-8 are near infrared spectra of a blended crude oil to be tested, a bus weight, a sand light, a Balin sand, a Kewit, a sand ultra light, a WTI according to an embodiment of the present invention;
FIG. 9 is a convolutional neural network diagram of a classification model according to an embodiment of the present invention;
FIGS. 10-12 are graphs showing the deviation degree between the predicted value and the true value of the regression model of three single crude oils in the mixed crude oil to be tested according to an embodiment of the present invention;
FIG. 13 is a flow chart showing the detailed implementation of a method for predicting the physical and chemical properties of a blended crude oil according to an embodiment of the present invention;
FIG. 14 is a schematic diagram of a system for predicting the physical and chemical properties of a blended crude oil in accordance with an embodiment of the present invention.
Detailed Description
The following describes the detailed implementation of the embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the invention, are not intended to limit the invention.
Aiming at the defects of the existing near infrared analysis technology, the invention provides a method, a system and a readable medium for predicting the components and physicochemical properties of mixed crude oil, and the method, the system and the readable medium can be used for predicting the components and the physicochemical properties of the mixed crude oil by carrying out one-time infrared spectrometry on the mixed crude oil to be detected.
FIG. 1 is a flow chart of a method for predicting the composition and physicochemical properties of a blended crude oil according to an embodiment of the present invention, specifically comprising the steps of:
step 100: and obtaining the near infrared spectrum of the mixed crude oil to be detected.
Specifically, after the near infrared spectrum of the mixed crude oil to be measured is obtained, the following steps are further executed:
performing first derivative treatment on the near infrared spectrum of the mixed crude oil to be detected, and selecting a wavelength band of 4713-5345cm -1 And 5978-6611cm -1 Near infrared spectrum of the interval.
Therefore, the near infrared spectrum of the specified band interval is favorable for extracting spectral characteristics, the prediction range is reduced, and the prediction efficiency is improved.
And extracting principal components from the near infrared spectrum to obtain principal component indexes, and inputting the principal component indexes into a prediction model.
Step 101: and inputting the near infrared spectrum of the mixed crude oil to be detected into a pre-constructed prediction model to obtain the components and proportions of the single crude oil in the mixed crude oil to be detected.
The prediction model comprises a classification model for predicting the component types of single-variety crude oil in the mixed crude oil to be detected and a regression model for predicting the component proportions of the single-variety crude oil in the mixed crude oil to be detected.
Specifically, before performing step 101, the following steps may also be performed:
s1: near infrared spectra of the single crude oils are obtained respectively, and physicochemical properties of the single crude oils are obtained by using a standard experimental analysis method.
Specifically, after the near infrared spectrum of each single crude oil is obtained, the near infrared spectrum of the single crude oil is subjected to first derivative treatment, and the wavelength band is selected to be 4713-5345cm -1 And 5978-6611cm -1 Near infrared spectrum of the interval.
Alternatively, single-variety crude oils include, but are not limited to, tiger, bus in, cold lake, WTI, amann, kowter, in sand light, in balin sand, sand ultralight, and the like.
In some embodiments, a single crude oil of a batch of the bargain, the sand light, the balin sand, the koraide, the sand ultralight and the WTI is obtained as a single crude oil sample, and infrared spectra of the bargain, the sand light, the balin sand, the koraide ultralight and the WTI are collected by using a brookfield Matrix-F near infrared spectrometer, and as shown in fig. 2-8, near infrared spectra of the mixed crude oil to be measured, the bargain, the sand light, the balin sand, the koraide, the sand ultralight and the WTI are respectively obtained.
Alternatively, crude oil physicochemical properties include, but are not limited to, density, kinematic viscosity, acid number, sulfur content, salt content, nitrogen content, chlorine content, water content, carbon residue, pour point, naphtha yield, diesel yield, wax oil yield, residuum yield, and the like.
In some embodiments, density, sulfur content, char, naphtha yield, diesel yield, wax oil yield, and residuum yield are obtained as crude oil physicochemical properties.
S2: setting types and proportions of single-variety crude oil contained in the simulated mixed crude oil, and adding absorbance vectors of near infrared spectrums of the single-variety crude oil according to the types and proportions to form the near infrared spectrum of the simulated mixed crude oil.
In some embodiments, for single-variety crude oils A, B and C, single-variety crude oils a-C are any three of balanose, sand light, balin sand, kowter, sand ultralight, WTI, and m: n: h (mass ratio) to obtain simulated mixed crude oil, if the near infrared spectrums of the single crude oil A, B and C are known, the absorbance of the near infrared spectrum of the mixed crude oil is the absorbance obtained by adding the absorbance of each single crude oil according to the mixing ratio according to the lambert-beer law.
S3: performing first derivative treatment on near infrared spectrum of simulated mixed crude oil, and selecting wavelength band of 4713-5345cm -1 And 5978-6611cm -1 And (3) extracting main component indexes of the near infrared spectrum in the interval near infrared spectrum, and establishing a prediction model.
Specifically, the prediction model comprises a classification model for predicting the component types of single-variety crude oil in the mixed crude oil to be detected and a regression model for predicting the component proportions of the single-variety crude oil in the mixed crude oil to be detected.
In some embodiments, the classification model is a one-dimensional convolutional neural network classification model, the neural network diagram of which is shown in fig. 9, the classification model being established by:
step 1: leading the main component index into a one-dimensional convolution input layer, passing through two convolution layers, carrying out weight initialization and activation functions, passing through a pooling layer, and transmitting to an output layer through multiple convolution layers and pooling layers to obtain a type prediction value;
in some embodiments, referring to FIG. 9, principal component index vectors are imported into a one-dimensional convolved input layer, each principal component index vector containing 80 features. Through two layers of convolution layers, the convolution kernel size is 3, the number of filters is 16, the output of the first neural network layer is a 78 x 1 matrix, the output of the second neural network layer is a 76 x 1 matrix, the output matrix is 25 x 1 through one layer of pooling layer, the convolution kernel size is 3 through one layer of convolution, the number of filters is 32, the output matrix is 23 x 1 through one layer of pooling layer, the output matrix is 7*1, the output matrix corresponding to 32 filters is flattened through the Flatten layer, the output matrix is 224 x 1 through the response layer, the output matrix is 20 x 1 through the softmax activation function, the sum of elements of the output matrix is 1, and the corresponding type with the largest value in the output matrix is taken as the final output.
Thus, the characteristic information of single crude oil in the mixed crude oil simulation spectrum can be obtained by utilizing the strong characteristic extraction capability of the convolutional neural network.
Step 2: and comparing the type predicted value obtained by the output layer with a sample expected value, and if an error exists, adjusting the weight until the error between the type predicted value and the sample expected value reaches the minimum or minimum value, so as to obtain the classification model.
In some embodiments, the performance of the classification model is judged through the confusion matrix, a better classification model is selected, the confusion matrix is shown in table 1, the accuracy rate can reach 73%, the recall rate can reach 79%, and the model classification effect is better.
TABLE 1
In some embodiments, the regression model is constructed according to any one of multiple linear regression, partial least squares, and artificial neural network regression modeling methods, the regression model being optimized by:
step 1: comparing the component proportion of the single-variety crude oil in the simulated mixed crude oil predicted by the regression model with the set component proportion of the single-variety crude oil to obtain a deviation result;
step 2: optimizing the regression model based on the bias results.
In some embodiments, the obtained data set is divided into three initial training sets according to the types of single crude oil contained in the simulated mixed crude oil, the mixed crude oil contains single crude oil A, B and C, the single crude oil A-C is any three of Barbary weight, sand light, balin sand, kowitt, sand ultra light and WTI, a one-dimensional convolutional neural network is adopted to respectively establish regression models of the contents of the three single crude oils in the mixed crude oil to be tested, and the root mean square error RMSEP and the decision coefficient R are predicted 2 And judging the performance of the regression model.
RMSEP can be expressed by the following formula:
where RMSEP is the predicted root mean square error, a smaller value indicates a better fit, assuming y i Is true, f (x i ) For the predicted value, m is the number of samples.
Determining the coefficient R 2 Has a value of between 0 and 1, R 2 The closer to 1, the better the effect of the description model, the closer to 0, the worse the effect of the description model, of course R 2 Negative values are also present, at this time the model is very poorly effective, determining the coefficient R 2 The expression can be used as follows:
let y be i To be a true value of the value,is the average of the true values, m is the number of samples, f (x i ) Is a predicted value.
The deviation degree of the single-variety crude oil proportion predicted value in the simulated mixed crude oil and the single-variety crude oil component proportion in the set simulated mixed crude oil obtained by the crude oil regression model is shown in figures 10-12, the predicted root mean square error RMSEP of the three regression models is less than 0.03, and the coefficient R is determined 2 All the regression models are larger than 0.98, so that the three regression models can be seen to have small prediction errors and good fitting effect.
And (3) performing first derivative treatment on the near infrared spectrum of the mixed crude oil to be detected, extracting main component indexes of the near infrared spectrum in a specified band interval, inputting the main component indexes into a classification model and a regression model, and predicting the components of the obtained mixed crude oil to be detected as shown in a table 2.
TABLE 2
Step 102: and obtaining the physicochemical properties of single crude oil in the mixed crude oil to be tested by using a standard experimental analysis method.
Step 103: fitting according to the components, proportions and physicochemical properties of the single crude oil to generate the physicochemical properties of the mixed crude oil to be tested.
Specifically, for physicochemical properties to be predicted that satisfy the linear condition, a linear fitting calculation is performed:
a mix ＝w 1 a 1 +w 2 a 2 +…+w n a n
wherein a is mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, a n Is the physicochemical property to be predicted corresponding to the nth single variety crude oil in the mixed crude oil to be detected; and
for physicochemical properties to be predicted that satisfy the nonlinear condition, performing nonlinear fitting calculation:
wherein b mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, b n Is the physicochemical property to be predicted corresponding to the nth single crude oil in the mixed crude oil to be detected.
In this example, according to the predicted result of the components of the mixed crude oil and the physicochemical properties of the single crude oil, a fitting value of the physicochemical properties of the mixed crude oil is generated according to a fitting calculation method, and the comparison result of the fitting value and the reference value of the physicochemical properties of the mixed crude oil obtained by a standard test analysis method is shown in table 3.
TABLE 3 Table 3
The average relative error absolute values of the density, sulfur content, carbon residue, naphtha yield, diesel yield, wax oil yield and residual oil yield of the predicted mixed crude oil are respectively 0.54%,7.05%,6.84%,4.36%,4.88%,2.15% and 5.60%, so that the accuracy requirements of crude oil analysis of a refinery are met.
In some embodiments, referring to fig. 13, a flowchart of a method for predicting a mixed crude oil component and physicochemical properties according to an embodiment of the present invention is shown, and specifically includes the following steps:
step 130: near infrared spectra of the single crude oils are obtained respectively, and physicochemical properties of the single crude oils are obtained by using a standard experimental analysis method.
Step 131: setting types and proportions of single-variety crude oil contained in the simulated mixed crude oil, and adding absorbance vectors of near infrared spectrums of the single-variety crude oil according to the types and proportions to form the near infrared spectrum of the simulated mixed crude oil.
Step 132: performing first derivative treatment on near infrared spectrum of simulated mixed crude oil, and selecting wavelength band of 4713-5345cm -1 And 5978-6611cm -1 Extracting main component indexes in the near infrared spectrum, and establishing a prediction model.
Step 133: and obtaining the near infrared spectrum of the mixed crude oil to be detected.
Step 134: performing first derivative treatment on near infrared spectrum of mixed crude oil to be detected, and selecting a wavelength band of 4713-5345cm -1 And 5978-6611cm -1 The near infrared spectrum of the interval is used as the near infrared spectrum input into the prediction model.
Step 135: inputting the near infrared spectrum of the mixed crude oil to be detected into a pre-constructed prediction model, extracting the same main component from the input near infrared spectrum in the model, obtaining main component indexes, and predicting to obtain the components and proportions of single crude oil in the mixed crude oil to be detected.
Step 136: fitting according to the components, proportions and physicochemical properties of the single crude oil to generate the physicochemical properties of the mixed crude oil to be tested.
Correspondingly, the embodiment of the invention also provides a prediction system for mixing the components and physicochemical properties of crude oil, which comprises the following components:
a memory for storing instructions executable by the processor;
and the processor is used for executing the instructions to realize the method.
In some embodiments, a block diagram of a predictive system for blending crude oil components and physicochemical properties is shown in FIG. 14, including an internal communication bus 141, a processor (processor) 142, a Read Only Memory (ROM) 143, a Random Access Memory (RAM) 144, a communication port 145, and a hard disk 147. The internal communication bus 141 may enable data communication between predicted system components that mix crude components and physicochemical properties. Processor 142 may make decisions and issue prompts. In some embodiments, processor 142 may be comprised of one or more processors.
The communication port 145 may enable data transmission and communication between the prediction system of the blended crude oil components and physicochemical properties and external input/output devices. In some embodiments, the prediction system of the blended crude oil components and physicochemical properties may send and receive information and data from the network through the communication port 145. In some embodiments, the prediction system of the blended crude oil components and physicochemical properties may be in data transmission and communication in a wired form with external input/output devices via input/output 146.
The prediction system for the composition and physicochemical properties of the blended crude oil may also include various forms of program storage units as well as data storage units, such as a hard disk 147, read Only Memory (ROM) 143 and Random Access Memory (RAM) 144, capable of storing various data files for computer processing and/or communication, and possible program instructions for execution by processor 142. The processor 142 executes these instructions to implement the main part of the method. The results processed by the processor 142 are transmitted to an external output device through the communication port 145 and displayed on the user interface of the output device.
For example, the implementation process files of the above-described method for predicting the physicochemical properties and components of the mixed crude oil may be computer programs, stored in the hard disk 147, and recorded into the processor 142 for execution to implement the method of the present application.
In another aspect, the invention also provides a computer readable medium having stored thereon computer instructions, wherein the method described above is performed when the computer instructions are executed by a processor.
When the process file for the prediction method of the physical and chemical properties of the mixed crude oil is a computer program, the process file may be stored in a computer readable storage medium as an article of manufacture. For example, computer-readable storage media may include, but are not limited to, magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact Disk (CD), digital Versatile Disk (DVD)), cards, and flash memory devices (e.g., electrically erasable programmable read-only memory (EPROM), cards, sticks, key drives). Moreover, various storage media described herein can represent one or more devices and/or other machine-readable media for storing information. The term "machine-readable medium" can include, without being limited to, wireless channels and various other media (and/or storage media) capable of storing, containing, and/or carrying code and/or instructions and/or data.
The terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly indicates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Those of skill in the art would understand that information, signals, and data may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The various illustrative logical modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a web site, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk (disk) and disc (disk) as used herein include Compact Disc (CD), laser disc, optical disc, digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks (disk) usually reproduce data magnetically, while discs (disk) reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (9)
1. A method for predicting a blend of crude oil components and physicochemical properties, the method comprising:
acquiring a near infrared spectrum of mixed crude oil to be detected;
inputting the near infrared spectrum of the mixed crude oil to be detected into a pre-constructed prediction model to obtain the components and the proportions of the single-variety crude oil in the mixed crude oil to be detected, wherein the prediction model comprises a classification model for predicting the single-variety crude oil component types in the mixed crude oil to be detected and a regression model for predicting the single-variety crude oil component proportions in the mixed crude oil to be detected;
obtaining physical and chemical properties of single crude oil in the mixed crude oil to be tested by using a standard experimental analysis method;
fitting to generate physicochemical properties of the mixed crude oil to be tested according to the components, the proportion and the physicochemical properties of the single crude oil.
2. The method according to claim 1, further comprising, after the obtaining the near infrared spectrum of the mixed crude oil to be measured:
performing first derivative treatment on the near infrared spectrum of the mixed crude oil to be detected, wherein the extraction wavelength band is 4713-5345cm -1 And 5978-6611cm -1 And inputting the principal component index of the near infrared spectrum of the interval into the prediction model.
3. The prediction method according to claim 1, further comprising, before said inputting the near infrared spectrum of the blended crude oil to be measured into a pre-constructed prediction model:
respectively obtaining near infrared spectrums of the single-variety crude oils, and obtaining physicochemical properties of the single-variety crude oils by using a standard experimental analysis method;
setting types and proportions of single-variety crude oil contained in the simulated mixed crude oil, and adding absorbance vectors of near infrared spectrums of the single-variety crude oil according to the types and proportions to form a near infrared spectrum of the simulated mixed crude oil;
and extracting main component indexes in the near infrared spectrum of the simulated mixed crude oil, and establishing the prediction model.
4. The method of predicting as set forth in claim 3, further comprising, after said forming the near infrared spectrum simulating the blended crude oil:
performing first derivative treatment on the near infrared spectrum of the simulated mixed crude oil, and selecting a wavelength band of 4713-5345cm -1 And 5978-6611cm -1 Near infrared spectrum of the interval.
5. A prediction method according to claim 3, wherein the classification model is a one-dimensional convolutional neural network classification model, which is built by:
leading the main component index into a one-dimensional convolution input layer, passing through two convolution layers, carrying out weight initialization and activation functions, passing through a pooling layer, and transmitting to an output layer through multiple convolution layers and pooling layers to obtain a type prediction value;
and comparing the type predicted value obtained by the output layer with a sample expected value, and if an error exists, adjusting the weight until the error between the type predicted value and the sample expected value reaches the minimum or minimum value, so as to obtain the classification model.
6. A prediction method according to claim 3, wherein the regression model is constructed according to any one of a multiple linear regression, a partial least squares method and an artificial neural network regression modeling method, and is optimized by:
comparing the component proportion of the single-variety crude oil in the simulated mixed crude oil predicted by the regression model with the set component proportion of the single-variety crude oil to obtain a deviation result;
optimizing the regression model based on the bias results.
7. The method according to claim 1, wherein the fitting to generate the physicochemical properties of the blended crude oil to be tested based on the components and proportions of the single crude oil and the physicochemical properties of the single crude oil comprises:
for physicochemical properties to be predicted that satisfy the linear condition, performing a linear fitting calculation:
a mix ＝w 1 a 1 +w 2 a 2 +…+w n a n
wherein a is mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, a n Is the physicochemical property to be predicted corresponding to the nth single variety crude oil in the mixed crude oil to be detected; and
for physicochemical properties to be predicted that satisfy the nonlinear condition, performing nonlinear fitting calculation:
wherein b mix Is the physicochemical property to be predicted of the mixed crude oil to be measured, w n Is the mass fraction of the nth single crude oil in the mixed crude oil to be detected, b n Is the physicochemical property to be predicted corresponding to the nth single crude oil in the mixed crude oil to be detected.
8. A system for predicting the physical and chemical properties of a blended crude oil composition, comprising:
a memory for storing instructions executable by the processor;
a processor for executing the instructions to implement the method of any one of claims 1-7.
9. A computer readable medium having stored thereon computer instructions, wherein the computer instructions, when executed by a processor, perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211225595.1A CN117894386A (en) | 2022-10-09 | 2022-10-09 | Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211225595.1A CN117894386A (en) | 2022-10-09 | 2022-10-09 | Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117894386A true CN117894386A (en) | 2024-04-16 |
Family
ID=90651301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211225595.1A Pending CN117894386A (en) | 2022-10-09 | 2022-10-09 | Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117894386A (en) |
-
2022
- 2022-10-09 CN CN202211225595.1A patent/CN117894386A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yun et al. | A hybrid variable selection strategy based on continuous shrinkage of variable space in multivariate calibration | |
| CN101988895B (en) | Method for predicting single-type crude oil content in mixed crude oil by near infrared spectrum | |
| CA2740845C (en) | Estimating detailed compositional information from limited analytical data | |
| CN103534578B (en) | Determine spectral device and the method for composition present in sample | |
| US8269961B2 (en) | System and method for determining the asphaltene content of crude oil | |
| CN110687072B (en) | Calibration set and verification set selection and modeling method based on spectral similarity | |
| CN104990894B (en) | A kind of gasoline property detection method based on weighting absorbance and similar sample | |
| CN104089911A (en) | Spectral model transmission method based on unary linear regression | |
| CN107976419B (en) | Method for predicting properties of oil product by near infrared spectrum | |
| WO2017007845A1 (en) | Method for correlating physical and chemical measurement data sets to predict physical and chemical properties | |
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN114676792A (en) | Near infrared spectrum quantitative analysis dimensionality reduction method and system based on stochastic projection algorithm | |
| CN117894386A (en) | Method, system and readable medium for predicting components and physicochemical properties of mixed crude oil | |
| CN113533133A (en) | Asphalt dynamic viscoelastic characteristic prediction method based on machine learning | |
| CN109145887B (en) | Threshold analysis method based on spectral latent variable confusion discrimination | |
| CN114783539A (en) | Traditional Chinese medicine component analysis method and system based on spectral clustering | |
| CN115326749A (en) | Method and apparatus for measuring contaminants | |
| CN112749466A (en) | Intelligent manufacturing measurement control integrated model and establishing method and application thereof | |
| WO2024011687A1 (en) | Method and apparatus for establishing oil product physical property fast evaluation model | |
| CN116413417A (en) | Oil product data analysis method, system and equipment based on spectrum data | |
| CN117116375A (en) | Method and system for predicting component parts of mixed sample | |
| CN116246728A (en) | Crude oil blending method, system and storage medium | |
| Rinnan et al. | Significance of the structure of data in partial least squares regression predictions involving both natural and human experimental design | |
| CN118016199A (en) | Column chromatography prediction method based on machine learning | |
| CN110907570B (en) | Organic matter maturity evaluation method and terminal equipment |
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 |