CN118430702B - Data fusion method based on object data model - Google Patents
Data fusion method based on object data model Download PDFInfo
- Publication number
- CN118430702B CN118430702B CN202410882438.0A CN202410882438A CN118430702B CN 118430702 B CN118430702 B CN 118430702B CN 202410882438 A CN202410882438 A CN 202410882438A CN 118430702 B CN118430702 B CN 118430702B
- Authority
- CN
- China
- Prior art keywords
- raw material
- data
- layer
- data set
- content
- 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
- 238000013499 data model Methods 0.000 title claims abstract description 13
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 11
- 239000000654 additive Substances 0.000 claims abstract description 92
- 230000004048 modification Effects 0.000 claims abstract description 46
- 238000012986 modification Methods 0.000 claims abstract description 46
- 239000012535 impurity Substances 0.000 claims abstract description 44
- 230000007797 corrosion Effects 0.000 claims abstract description 42
- 238000005260 corrosion Methods 0.000 claims abstract description 42
- 238000002156 mixing Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000012502 risk assessment Methods 0.000 claims abstract description 9
- 238000005984 hydrogenation reaction Methods 0.000 claims abstract description 5
- 239000002994 raw material Substances 0.000 claims description 273
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 86
- 230000000996 additive effect Effects 0.000 claims description 69
- 229910052742 iron Inorganic materials 0.000 claims description 43
- 229910052720 vanadium Inorganic materials 0.000 claims description 39
- LEONUFNNVUYDNQ-UHFFFAOYSA-N vanadium atom Chemical compound [V] LEONUFNNVUYDNQ-UHFFFAOYSA-N 0.000 claims description 39
- 238000000034 method Methods 0.000 claims description 37
- 229910001385 heavy metal Inorganic materials 0.000 claims description 32
- 230000006870 function Effects 0.000 claims description 25
- 238000010606 normalization Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000004519 manufacturing process Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 9
- 239000003795 chemical substances by application Substances 0.000 claims description 9
- 239000000463 material Substances 0.000 claims description 9
- 230000007935 neutral effect Effects 0.000 claims description 9
- 239000003607 modifier Substances 0.000 claims description 8
- 210000002569 neuron Anatomy 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 238000005530 etching Methods 0.000 claims description 7
- 230000000246 remedial effect Effects 0.000 claims description 6
- 238000009616 inductively coupled plasma Methods 0.000 claims description 5
- 238000005067 remediation Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 239000004215 Carbon black (E152) Substances 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 229930195733 hydrocarbon Natural products 0.000 claims description 3
- 150000002430 hydrocarbons Chemical class 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000008447 perception Effects 0.000 claims description 3
- 230000001902 propagating effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims 2
- 239000003208 petroleum Substances 0.000 abstract description 11
- 239000003209 petroleum derivative Substances 0.000 abstract description 7
- 238000007405 data analysis Methods 0.000 abstract description 2
- 230000001105 regulatory effect Effects 0.000 abstract 1
- 239000002585 base Substances 0.000 description 45
- 239000000243 solution Substances 0.000 description 17
- 238000013441 quality evaluation Methods 0.000 description 4
- 239000002253 acid Substances 0.000 description 3
- 150000002500 ions Chemical class 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000006056 electrooxidation reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910052751 metal Inorganic materials 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 238000010948 quality risk assessment Methods 0.000 description 2
- 101100391172 Dictyostelium discoideum forA gene Proteins 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000002378 acidificating effect Effects 0.000 description 1
- 239000003513 alkali Substances 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000003054 catalyst Substances 0.000 description 1
- 239000003518 caustics Substances 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 239000006184 cosolvent Substances 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- -1 hydrogen ions Chemical class 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000003112 inhibitor Substances 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G16C20/90—Programming languages; Computing architectures; Database systems; Data warehousing
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Probability & Statistics with Applications (AREA)
- Medical Informatics (AREA)
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
Abstract
The invention discloses a data fusion method based on an object data model, which relates to the technical field of data analysis and is used for solving the problems that the quality of petroleum is improved by carrying out modification treatment on petroleum after impurities are removed through hydrogenation in the pretreatment of petroleum processing, and impurity mixing, equipment corrosion and the like are caused by the fact that impurities are added secondarily due to improper use of additives in the modification treatment; and the risk assessment model is constructed by collecting the matched processing feedback data and the processing equipment data and integrating the processing feedback data and the processing equipment data, so that the petroleum products with higher risks are remedied and regulated, and the risks of the petroleum products are reduced.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to a data fusion method based on an object data model.
Background
Data governance techniques are a series of methods and tools for managing and controlling enterprise data assets, the primary goal of which is to ensure the availability, integrity, security, and compliance of data. Data quality risk management and control is used as one aspect of data management technology, and is safer and more reliable in petroleum processing when the data quality risk management and control is applicable to the petroleum industry.
The prior art has the following defects:
The petroleum is modified after the impurities are removed by hydrogenation during the pretreatment of petroleum processing, so that the quality of the petroleum is improved, and the problems of impurity mixing, equipment corrosion and the like are caused by secondary addition of impurities caused by improper use of the additive during the modification treatment, so that the benefit of companies is lost and the environment is improved.
The present invention proposes a solution to the above-mentioned problems.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, an embodiment of the present invention provides a data fusion method based on an object data model, so as to solve the problems set forth in the above-mentioned background art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a data fusion method based on an object data model comprises the following steps:
step S1, collecting data of a modified additive and raw material data;
s2, analyzing and classifying the modified additive data by using an AND gate logic method, classifying the raw materials according to the calculated covariance of the raw material data as a raw material classification threshold, and performing modification treatment after classifying and matching the modified additive and the raw materials;
Step S3, collecting processing feedback data and processing equipment data after modification treatment, and constructing a risk assessment model by the comprehensive processing feedback data and the processing equipment data through a multi-layer perception mechanism; analyzing the risk degree of the raw material subjected to modification treatment according to a risk assessment model;
And S4, carrying out remediation treatment on the raw material with high risk degree by using a blending method.
In a preferred embodiment, in step S1, the feedstock data are feedstock acid-base difference and feedstock corrosiveness when unmodified after hydrogenation to remove impurities; the data of the modifying additive is the heavy metal content of the modifying additive.
In a preferred embodiment, in step S2, the upgrading additive is classified by detecting the heavy metal iron content and vanadium content of the upgrading additive, specifically as follows:
Collecting enough different types of modification additives, measuring the iron content and the vanadium content in the modification additives by using an inductively coupled plasma mass spectrometer, screening the modification additives with the iron content or the vanadium content of 0, marking the removed modification additives as low heavy metal content modification additives, and respectively combining the iron content data and the vanadium content data in the rest modification additives into an iron content data set and a vanadium content data set;
Respectively setting an iron content threshold value and a vanadium content threshold value according to the iron content data set and the vanadium content data set;
the modified additive is classified by an AND gate logic method, and the classification comprises the following specific parts:
determining input and output: taking the iron content and the vanadium content of the modified additive as input, and taking the additive classification as output;
setting an input rule: setting the iron content of the modifying additive to 1 as input if the iron content of the modifying additive exceeds an iron content threshold value, otherwise setting to 0 for input; if the vanadium element content of the modifying additive exceeds the vanadium content threshold, setting the vanadium content of the modifying additive to be 1 for input, otherwise setting the vanadium element content of the modifying additive to be 0 for input;
Setting an output rule: if the inputs are 1, setting the output result as 1, otherwise setting the output result as 0;
Classifying the modified additives according to the output result: classifying and marking the modified additive with the output result of 1 as a heavy metal content modifier; the upgrading additive with the output result of 0 is classified as a low heavy metal content upgrading agent and marked.
In a preferred embodiment, in step S2, the neutral difference coefficient is also calculated as the raw material acid-base difference, which is calculated by the formulaWhereinIs the coefficient of the neutral difference,Is the pH value of the raw material, 7 is the neutral pH value coefficient,
The raw material corrosion rate is calculated by calculating the raw material corrosion rate, and the raw material corrosion rate is represented by using a raw material corrosion rate numerical value, and a general corrosion rate formula can be used for calculating the raw material corrosion rate: wherein For the corrosion rate of the raw material,Is a constant, the specific value depends on the unit selected for the corrosive material,For the mass of the etching material before and after etching by the raw material,Is the surface area of the corrosion material, and time is the corrosion time;
Collecting raw material data of a plurality of production lines, and respectively combining acid-base difference normalized data and corrosion degree normalized data obtained by normalizing the collected raw material data into an acid-base difference data set and a corrosion degree data set; the normalization formula can be WhereinIs the result of the normalization process,Is the ith data in the acid-base difference data set or the corrosiveness data set,For the minimum value in the acid-base difference data set or the minimum value in the corrosiveness data set,The maximum value in the acid-base difference data set or the maximum value in the corrosiveness data set;
The acid-base difference data set and the corrosiveness data set are randomly sampled, acid-base difference normalization data in the acid-base difference data set are obtained, corrosiveness normalization data in the corrosiveness data set are obtained, covariance is calculated from the screened acid-base difference normalization data and corrosiveness normalization data to be used as a raw material classification threshold, and covariance is used as a raw material classification threshold;
the normalized data obtained by the above steps of the collected current raw material acid-base difference data and the current raw material corrosiveness data are used as the data in the calculated covariance formula AndMarking the calculation result asClassifying the current raw materials by comparing the calculation result with a raw material classification threshold;
If it is Marking the current raw material as a high-impurity raw material if the acid-base difference of the current raw material and the corrosiveness covariance of the current raw material exceed the raw material classification threshold; if it isMarking the current raw material as a low-impurity raw material if the acid-base difference of the current raw material and the corrosiveness covariance of the current raw material are lower than the raw material classification threshold;
If the current raw material is marked as a high-impurity raw material, modifying treatment is carried out by using a modifying agent with low heavy metal content; if the current feedstock is marked as a low impurity feedstock, a upgrading treatment is performed using a high heavy metal content upgrading agent.
In a preferred embodiment, in step S3, the post-upgrading process feedback data and the processing equipment data are the post-upgrading raw material ash content and the raw material yield, respectively;
The ash content of the modified raw material is the ash content of various non-hydrocarbon impurities in the modified raw material,
The raw material yield is the ratio of the weight of the raw material after modification to the weight of the raw material before modification;
collecting the ash content and the yield of raw materials of a sufficient quantity of modified production line, respectively merging the ash content data set and the yield data set into the ash content data set and the yield data set, and constructing a multi-layer perceptron model to analyze the current risk degree of the raw materials through the ash content data set and the yield data set, wherein the method comprises the following specific steps:
The ash content data set and the raw material yield data set are used as two input features for constructing a multi-layer perceptron model, wherein the multi-layer perceptron model is provided with two neural network layers, namely an input layer, a hidden layer and an output layer; the input layer has two neurons corresponding to two input features, and the output layer has one neuron for outputting classification results;
initializing parameters: initializing the weight from an input layer to a hidden layer and the bias parameter from the hidden layer to an output layer in the multi-layer perceptron;
Forward propagation: for a given input feature, the multi-layer perceptron calculates the output from the input layer to the hidden layer through weight and bias parameters, and then transfers the output to the output layer, and converts the output to a probability value between 0 and 1 through an activation function to represent the current raw material risk level, wherein the activation function of the hidden layer can be as follows Where a is the output that the output layer transmits to the hidden layer,For the first ash content data in the ash content dataset,For the first data in the raw material yield dataset,AndThe weight from the input layer to the hidden layer is given, and c is a bias parameter; the activation function of the output layer is Sigmoid function, and the formula isWherein y is an output probability result, e is a natural base number, and x is the output transmitted from the hidden layer to the output layer;
loss calculation: according to the output probability result of the multi-layer perceptron model and a preset actual label, calculating a loss function, wherein the formula can be a mean square error formula: ; wherein the method comprises the steps of For the mean square error, n is the number of data in both data sets,For the output probability result of the ith raw material ash content data in the ash content data set and the ith raw material yield in the raw material yield data set through the multi-layer perceptron,The method comprises the steps of presetting an actual label for an ith;
back propagation: calculating the gradient of the parameters of each layer to the loss function from the output layer, forward propagating the gradient layer by layer, and calculating the weight of each layer and the partial derivative of the bias parameters to the loss function, wherein the partial derivative formula is that ,,WhereinAndRespectively obtaining the weight from the input layer to the hidden layer and the bias results of the bias parameters of the hidden layer and the output layer; updating the weight and the bias parameter by calculating the partial derivative of the loss function on the weight and the bias parameter;
Updating parameters: according to the gradient obtained by calculation, setting a learning rate L, and randomly adjusting the weight from an input layer to a hidden layer and the bias parameters of the hidden layer and an output layer; updating parameters in the multi-layer perceptron, wherein an updating rule formula is as follows: 、 And ;
Repeating the iteration: repeatedly performing forward propagation and backward propagation until the model converges, namely, the model training of the multi-layer perceptron is completed;
collecting the current raw material ash content and the current raw material yield, taking the current raw material ash content and the current raw material yield as input, calculating an output probability result through a multi-layer perceptron model, and analyzing the current raw material risk degree by comparing the output probability result with a preset probability threshold;
when the output probability result exceeds the probability threshold, marking the current raw material as a high-risk-degree raw material; otherwise, the current stock is marked as a low risk level stock.
In a preferred embodiment, in step S4, if the current stock is marked as a high risk grade stock, then remedial treatment is performed on the current stock; the method comprises a blending method, and comprises the following specific processes:
setting a blending proportion of a raw material with more impurities, namely a high-risk-degree raw material, and a raw material with lower impurities, namely a low-risk-degree raw material, mixing and blending, collecting the ash content of the current raw material, recording the ash content as HF, collecting the raw material with the lowest ash content in other production lines as a blending raw material, and marking the ash content as ; Determining and marking a reconciliation threshold asThe blending ratio is calculated by the following formula: Wherein t is the blending ratio;
and reducing the ash content of the current raw material to a blending threshold value through a blending proportion, and completing the remediation treatment of the high-risk-degree raw material.
The data fusion method based on the object data model has the technical effects and advantages that:
According to the invention, the modification additive data and the raw material data are collected, the modification additive is classified according to the modification additive data, and the modification additive is classified into a high heavy metal content modifier and a low heavy metal content modifier. And (3) carrying out quality evaluation on the raw materials by using the raw material data, and classifying the raw materials into high-impurity raw materials and low-impurity raw materials according to quality evaluation results. The classified upgrading additive and the feedstock are matched. And collecting the matched processing feedback data and processing equipment data, constructing a risk assessment model by integrating the processing feedback data and the processing equipment data, and carrying out remedial adjustment on petroleum products with higher risks, thereby reducing the risks of the petroleum products.
Drawings
FIG. 1 is a logic flow diagram of a data fusion method based on an object data model according to the present invention;
FIG. 2 is a flow chart of a data fusion method based on an object data model.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
According to the invention, the modification additive data and the raw material data are collected, the modification additive is classified according to the modification additive data, and the modification additive is classified into a high heavy metal content modifier and a low heavy metal content modifier. And (3) carrying out quality evaluation on the raw materials by using the raw material data, and classifying the raw materials into high-impurity raw materials and low-impurity raw materials according to quality evaluation results. The classified upgrading additive and the feedstock are matched. And collecting the matched processing feedback data and processing equipment data, constructing a risk assessment model by integrating the processing feedback data and the processing equipment data, and carrying out remedial adjustment on petroleum products with higher risks, thereby reducing the risks of the petroleum products.
Examples
The invention discloses a data fusion method based on an object data model, which is shown in fig. 1 and 2 and comprises the following steps:
and S1, collecting data of a modifying additive and data of a raw material.
And S2, analyzing and classifying the modified additive data by using an AND gate logic method, classifying the raw materials according to the calculated covariance of the raw material data as a raw material classification threshold, and performing modification treatment after classifying and matching the modified additive and the raw materials.
And S3, collecting the processing feedback data and the processing equipment data after modification treatment, and constructing a risk assessment model by the comprehensive processing feedback data and the processing equipment data through a multi-layer perception mechanism. And analyzing the risk degree of the raw material subjected to the modification treatment according to the risk assessment model.
And S4, carrying out remediation treatment on the raw material with high risk degree by using a blending method.
The specific implementation is as follows:
In step S1, the raw material data are the difference in acid and alkali and the corrosiveness of the raw material when the raw material is unmodified after the impurities are removed by hydrogenation. The acid-base difference of the raw materials can be detected by an infrared spectrometer, and the corrosion degree of the raw materials can be obtained by an electrochemical corrosion tester. The data of the modifying additive is the heavy metal content of the modifying additive, and can be obtained through detection by an inductively coupled plasma mass spectrometer.
The difference in acid and base of the raw material refers to the degree of concentration of hydrogen ions in the raw material solution, and is generally expressed by the pH value, and the higher the acid value or base number in the raw material solution is, the more acidic or basic impurities contained in the raw material solution are, the worse the quality of the raw material solution is.
The corrosion degree of the raw material refers to the comprehensive index of the corrosion degree of the raw material to metal equipment and pipelines. For evaluating the influence of the raw material solution on the production apparatus, the higher the corrosiveness of the raw material solution, the more corrosive substances therein, the poorer the quality of the raw material solution.
Heavy metal content of the modifying additive, and the modifying additive added in the petroleum raw material solution is of a plurality of types, and the modifying additive can be classified according to the heavy metal content of the modifying additive so as to adapt to the requirements of the raw material solution.
The infrared spectrometer is an instrument commonly used for chemical analysis, and can obtain the vibration frequency of a substance by detecting the absorption peak of the raw material solution, and qualitatively and quantitatively analyze the substance to obtain the pH value of the raw material solution, and the pH value of the raw material solution can be used for calculating the pH value of the raw material. An electrochemical corrosion detector is an instrument for detecting and analyzing the corrosion condition of a metal material or a solution, and reflects the corrosion degree of the solution by analyzing the ac impedance response of a detection raw material solution to a metal sample. The inductively coupled plasma mass spectrometer is an analysis instrument for simultaneously measuring multiple elements with high sensitivity, can separate ions through ion mass-to-charge ratio so as to measure the signal intensity of various heavy metal ions, and automatically calculates, sorts and presents the quantitative result of heavy metals in software so as to obtain the heavy metal content of the modifying additive.
In the step S2, the heavy metal iron and vanadium content of the modification additive is high, the secondary heavy metal addition during modification is greatly influenced, and the modification additive can be classified by detecting the heavy metal iron content and vanadium content of the modification additive, and the specific steps are as follows:
Collecting enough different types of modifying additives, measuring the iron content and the vanadium content in the modifying additives by using an inductively coupled plasma mass spectrometer, screening the modifying additives with the iron content or the vanadium content of 0, marking the removed modifying additives as low heavy metal content modifying agents, and respectively merging the iron content data and the vanadium content data in the rest modifying additives into an iron content data set and a vanadium content data set. According to the iron content data set and the vanadium content data set, an iron content threshold value and a vanadium content threshold value are respectively set, and the iron content data set is used for carrying out the specific steps as follows:
calculating standard deviation of the iron content data set by using a standard deviation formula: wherein N is the number of iron content data sets, i is the number of data sequence in the iron content data set, i=1, i=2, i=3, etc.,For the ith iron content data in the iron content dataset,Is the average value of the iron content data set. Will be+As a threshold for iron content, whereinFor the threshold adjustment coefficient, adjustment may be performed according to practical situations, for example, n=1.2.
Similarly, the vanadium content threshold can be obtained, and the modified additive can be classified by using an AND gate logic method. The classification is specifically composed of the following parts:
determining input and output: the iron content and vanadium content of the modified additive are taken as inputs, and the additive classification is taken as an output.
Setting an input rule: setting the iron content of the modifying additive to 1 as input if the iron content of the modifying additive exceeds an iron content threshold value, otherwise setting to 0 for input; and similarly, if the vanadium element content of the modifying additive exceeds the vanadium content threshold, setting the vanadium content of the modifying additive to be 1 for input, otherwise setting the vanadium element content of the modifying additive to be 0 for input.
Setting an output rule: if the inputs are all 1, the output result is set to 1, otherwise, the output result is set to 0.
Classifying the modified additives according to the output result: classifying and marking the modified additive with the output result of 1 as a heavy metal content modifier; the upgrading additive with the output result of 0 is classified as a low heavy metal content upgrading agent and marked.
The pH value of the raw material can be directly measured by an infrared spectrometer, however, more impurities appear when the acidity or the alkalinity is too high, so that the neutral difference coefficient can be calculated to serve as the pH value difference of the raw material, and the calculation formula is as followsWhereinIs the coefficient of the neutral difference,The pH value of the raw material is 7, the pH value coefficient is neutral, and the neutralization coefficient is poorThe larger the feedstock, i.e., the higher the acidity or basicity of the feedstock, the more impurities in the feedstock.
The raw material corrosion rate can be calculated by calculating the raw material corrosion rate, and the raw material corrosion rate can be represented by using a numerical expression of the raw material corrosion rate, and a general corrosion rate formula can be used for calculating the raw material corrosion rate: wherein For the corrosion rate of the raw material,Is a constant, the specific value depends on the unit selected for the corrosive material,For the mass of the etching material before and after etching by the raw material,Is the surface area of the etched material and time is the etching time. The parameters in the above formulas are all conventional means and are not analyzed here.
Raw material data of a plurality of production lines are collected, and acid-base difference normalized data and corrosion degree normalized data obtained after normalization processing of the collected raw material data are respectively combined into an acid-base difference data set and a corrosion degree data set. The normalization formula can beWhereinIs the result of the normalization process,Is the ith data in the acid-base difference data set or the corrosiveness data set,For the minimum value in the acid-base difference data set or the minimum value in the corrosiveness data set,Is the maximum value in the acid-base difference data set or the maximum value in the corrosiveness data set. Such as: For the second data in the acid-base difference data set, then Is the minimum value in the acid-base difference data set,Is the maximum value in the acid-base difference data set,Is the result of normalization of the second data in the acid-base difference dataset.
The obtained acid-base difference data set and the corrosiveness data set are sampled randomly, and the acid-base difference data set is taken as an example. Randomly selecting one acid-base difference data in the acid-base difference data set, recording and returning the acid-base difference data, and repeatedly operating for m times to obtain acid-base difference normalized data in the acid-base difference data set, wherein the acid-base difference normalized data in the acid-base difference data set are respectively marked as、、And (3) performing m identical operations on the corrosiveness data set in the same way to obtain corrosiveness normalized data in the corrosiveness data set, wherein the corrosiveness normalized data are respectively marked as、、Etc. The covariance can measure the linear correlation between two random variables, and the covariance can be calculated from the screened acid-base difference normalized data and corrosion normalized data to be used as a raw material classification threshold value, which comprises the following specific steps:
the covariance calculation formula is: Wherein, the method comprises the steps of, wherein, Covariance of the screened acid-base difference normalized data and corrosion normalized data,AndThe random data in the screened acid-base difference normalized data and the corrosion normalized data are respectively selected,Is thatI.e., the average value of the normalized data of the acid-base difference selected,Is thatThe average value of the normalized data of the selected corrosiveness, m is the number of the two data selected, and the number of the two data selected is the same because the two data sets are operated m times.
Covariance (covariance)Can be represented asWhen the covariance is larger than 0, it indicates that the variation trends of the two variables are identical, and when the covariance is smaller than 0, it indicates that the variation trends of the two variables are opposite. From the following componentsAndThe nature of (2) shows that the higher the acid-base difference and corrosion degree of the raw material, the more the raw material impurities, namelyAndIs two positively correlated random variables, so。
Can be used forAs a raw material classification threshold, the normalized data obtained by processing the collected current raw material acid-base difference data and the current raw material corrosiveness data in the steps are used as the calculated covariance formulaAndMarking the calculation result asAnd comparing the calculation result with a raw material classification threshold value to classify the current raw materials.
If it isThe current raw material acid-base difference and the current raw material corrosiveness covariance exceed the raw material classification threshold, namely the current raw material has more impurities, and the current raw material is marked as a high-impurity raw material; if it isAnd the current raw material acid-base difference and the current raw material corrosiveness covariance are lower than the raw material classification threshold, namely the current raw material has few impurities, and the current raw material is marked as a low-impurity raw material.
If the current raw material is marked as a high-impurity raw material, modifying treatment is carried out by using a modifying agent with low heavy metal content; if the current feedstock is marked as a low impurity feedstock, a upgrading treatment is performed using a high heavy metal content upgrading agent. The modification treatment needs to use a plurality of additives such as a catalyst, a cosolvent, an inhibitor and the like, and the risk of modification can be prevented while the high quality improvement of petroleum products is ensured through the selective use of the additives. The specific modification treatment is a conventional means, and will not be described herein.
It should be noted that, the sample selection of the modifying additive used for setting the iron content threshold, the vanadium content threshold and the raw material classification threshold and the raw material data selection of the production line are not unique, and can be selected according to practical situations.
In step S3, the modified processing feedback data and the processing equipment data are the modified raw material ash content and the raw material yield, respectively. The ash content of the modified raw material can be detected and obtained by an automatic ash analyzer, the raw material yield can be obtained by calculating the weight before the raw material is modified and the weight after the raw material is modified, and the weight before the raw material is modified and the weight after the raw material is modified can be obtained by accessing a raw material database.
The ash content of the modified raw material is various non-hydrocarbon impurities in the modified raw material, and the higher the ash content of the raw material is, the more impurities of the raw material are.
The raw material yield is the ratio of the weight after the raw material modification to the weight before the raw material modification. The calculation formula is as follows: wherein For the product yield, as the amount of impurities in the raw material increases, the damage degree of the upgrading equipment increases, and the weight of the raw material passing through the upgrading equipment decreases, so that the product yield decreases.
Collecting the ash content and the yield of raw materials of a sufficient quantity of modified production line, respectively merging the ash content data set and the yield data set into the ash content data set and the yield data set, and constructing a multi-layer perceptron model to analyze the current risk degree of the raw materials through the ash content data set and the yield data set, wherein the method comprises the following specific steps:
The ash content data set and the raw material yield data set are used as two input features for constructing a multi-layer perceptron model, wherein the multi-layer perceptron model is provided with two neural network layers, namely an input layer, a hidden layer and an output layer. The input layer has two neurons corresponding to two input features, the hidden layer has several neurons, the specific number is set according to training rules, and the output layer has one neuron for outputting classification results.
Initializing parameters: firstly, initializing weights from an input layer to a hidden layer and bias parameters from the hidden layer to an output layer in the multi-layer perceptron. It should be explained that the weights from the input layer to the hidden layer are initialized by random assignment, if there are two neurons in the hidden layer, the weights from the input layer to the hidden layer can be initialized to [0.4, 0.6] randomly, and the bias parameters are initialized to 0.5 randomly.
Forward propagation: for a given input feature, the multi-layer perceptron calculates the output from the input layer to the hidden layer through weight and bias parameters, and then transfers the output to the output layer, and converts the output to a probability value between 0 and 1 through an activation function to represent the current raw material risk level, wherein the activation function of the hidden layer can be as followsWhere a is the output that the output layer transmits to the hidden layer,For the first ash content data in the ash content dataset,For the first data in the raw material yield dataset,AndThe weight from the input layer to the hidden layer is given, and c is a bias parameter; the activation function of the output layer can be Sigmoid function, and the formula isWhere y is the output probability result, e is the natural base, and x is the output that the hidden layer passes to the output layer.
Loss calculation: according to the output probability result of the multi-layer perceptron model and a preset actual label, calculating a loss function, wherein the formula can be a mean square error formula: . Wherein the method comprises the steps of For the mean square error, n is the number of data in both data sets,For the output probability result of the ith raw material ash content data in the ash content data set and the ith raw material yield in the raw material yield data set through the multi-layer perceptron,The i-th preset actual label.
Back propagation: calculating the gradient of the parameters of each layer to the loss function from the output layer, forward propagating the gradient layer by layer, and calculating the weight of each layer and the partial derivative of the bias parameters to the loss function, wherein the partial derivative formula is that,,WhereinAndThe bias results of the bias parameters of the hidden layer and the output layer are respectively the weights of the input layer to the hidden layer. The weights and bias parameters are updated by calculating the partial derivatives of the loss functions to the weights and bias parameters.
Updating parameters: and setting a proper learning rate L to randomly adjust the weight from the input layer to the hidden layer and the bias parameters of the hidden layer and the output layer according to the calculated gradient. The output result is converged, and parameters in the multi-layer perceptron are updated, wherein an updating rule formula can be as follows:、 And . For example, in an iteration, calculatedSetting the learning rate L to 0.05, and subtracting the original weight from the updated weight. Thus, by iterating constantly, the weights and bias parameters are adjusted in a direction that minimizes the loss function.
Repeating the iteration: and repeatedly performing forward propagation and backward propagation until the model converges, namely, the model training of the multi-layer perceptron is completed.
Collecting the current raw material ash content and the current raw material yield, taking the current raw material ash content and the current raw material yield as input, calculating an output probability result through a multi-layer perceptron model, and analyzing the current raw material risk degree by comparing the output probability result with a preset probability threshold.
When the output probability result exceeds the probability threshold, marking the current raw material as a high-risk-degree raw material; otherwise, the current stock is marked as a low risk level stock.
The automatic ash analyzer is an instrument for rapidly and accurately measuring the ash content in various solids or liquids, and automatically collects part of raw materials through a sample injection system for heating measurement, detects the mass of ash remained after combustion, analyzes and calculates the ash content and records data. The raw material database stores raw material information and raw material processing information of petroleum companies, and the raw material pre-upgrading weight and the raw material post-upgrading weight can be obtained by accessing the raw material database. The multi-layer perceptron is a neural network model that can be trained by inputting sufficient data to evaluate the current data. The selection of the sufficient data may be performed by listening to expert opinion in the field, and the formulas in the step are only examples, and the selection of the formulas is not unique.
In step S4, if the current stock is marked as a high risk level stock, a remedial process is performed on the current stock. The method for remedying is more, and in consideration of different economic conditions of petroleum companies, the example of remedying is carried out by using a lower-cost blending method:
The blending method is to set blending proportion of the raw materials with more impurities, namely the raw materials with high risk degree and the raw materials with lower impurities, namely the raw materials with low risk degree, for blending, so that the purposes of reducing the impurity content and improving the raw material performance can be achieved through blending. The blending proportion, that is, the proportion of the current blending amount of the raw materials divided by the blending amount of the blended raw materials, can be set according to the ash content of the raw materials, the ash content of the current raw materials is collected and recorded as HF, and the raw materials with the lowest ash content in other production lines are collected as the blended raw materials and marked as the ash content . Determining and marking a reconciliation threshold asThe impurity amount of the current raw materials after mixing and blending is controlled below a blending threshold. The blending proportion can be calculated by the following formula: wherein t is the blending ratio.
And reducing the ash content of the current raw material to a blending threshold value through the blending proportion, thereby completing the remedial treatment of the current raw material, namely the raw material with high risk degree.
The blending threshold value is set by the person skilled in the art according to the raw material requirements, and is not analyzed here.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product.
Those of ordinary skill in the art will appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and application constraints imposed on the technology. 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 application.
In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Finally: the foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (3)
1. The data fusion method based on the object data model is characterized by comprising the following steps of:
step S1, collecting data of a modified additive and raw material data;
s2, analyzing and classifying the modified additive data by using an AND gate logic method, classifying the raw materials according to the calculated covariance of the raw material data as a raw material classification threshold, and performing modification treatment after classifying and matching the modified additive and the raw materials;
Step S3, collecting processing feedback data and processing equipment data after modification treatment, and constructing a risk assessment model by the comprehensive processing feedback data and the processing equipment data through a multi-layer perception mechanism; analyzing the risk degree of the raw material subjected to modification treatment according to a risk assessment model;
s4, carrying out remediation treatment on the raw materials with high risk degree by using a blending method;
in the step S1, raw material data are raw material acid-base difference and raw material corrosiveness when unmodified after hydrogenation to remove impurities; the data of the modifying additive is the heavy metal content of the modifying additive;
In step S2, the upgrading additive is classified by detecting heavy metal iron content and vanadium content of the upgrading additive, and the specific steps are as follows:
Collecting enough different types of modification additives, measuring the iron content and the vanadium content in the modification additives by using an inductively coupled plasma mass spectrometer, screening the modification additives with the iron content or the vanadium content of 0, marking the removed modification additives as low heavy metal content modification additives, and respectively combining the iron content data and the vanadium content data in the rest modification additives into an iron content data set and a vanadium content data set;
Respectively setting an iron content threshold value and a vanadium content threshold value according to the iron content data set and the vanadium content data set;
the modified additive is classified by an AND gate logic method, and the classification comprises the following specific parts:
determining input and output: taking the iron content and the vanadium content of the modified additive as input, and taking the additive classification as output;
setting an input rule: setting the iron content of the modifying additive to 1 as input if the iron content of the modifying additive exceeds an iron content threshold value, otherwise setting to 0 for input; if the vanadium element content of the modifying additive exceeds the vanadium content threshold, setting the vanadium content of the modifying additive to be 1 for input, otherwise setting the vanadium element content of the modifying additive to be 0 for input;
Setting an output rule: if the inputs are 1, setting the output result as 1, otherwise setting the output result as 0;
Classifying the modified additives according to the output result: classifying and marking the modified additive with the output result of 1 as a heavy metal content modifier; classifying and marking the modified additive with the output result of 0 as a low heavy metal content modifier;
In step S3, the modified processing feedback data and the processing equipment data are respectively the ash content and the raw material yield of the modified raw material;
The ash content of the modified raw material is the ash content of various non-hydrocarbon impurities in the modified raw material,
The raw material yield is the ratio of the weight of the raw material after modification to the weight of the raw material before modification;
collecting the ash content and the yield of raw materials of a sufficient quantity of modified production line, respectively merging the ash content data set and the yield data set into the ash content data set and the yield data set, and constructing a multi-layer perceptron model to analyze the current risk degree of the raw materials through the ash content data set and the yield data set, wherein the method comprises the following specific steps:
The ash content data set and the raw material yield data set are used as two input features for constructing a multi-layer perceptron model, wherein the multi-layer perceptron model is provided with two neural network layers, namely an input layer, a hidden layer and an output layer; the input layer has two neurons corresponding to two input features, and the output layer has one neuron for outputting classification results;
initializing parameters: initializing the weight from an input layer to a hidden layer and the bias parameter from the hidden layer to an output layer in the multi-layer perceptron;
Forward propagation: for a given input feature, the multi-layer perceptron calculates the output from the input layer to the hidden layer through weight and bias parameters, and then transfers the output to the output layer, and converts the output to a probability value between 0 and 1 through an activation function to represent the current raw material risk degree, wherein the activation function of the hidden layer is that Where a is the output that the output layer transmits to the hidden layer,For the first ash content data in the ash content dataset,For the first data in the raw material yield dataset,AndThe weight from the input layer to the hidden layer is given, and c is a bias parameter; the activation function of the output layer is Sigmoid function, and the formula isWherein y is an output probability result, e is a natural base number, and x is the output transmitted from the hidden layer to the output layer;
loss calculation: calculating a loss function according to the output probability result of the multi-layer perceptron model and a preset actual label, wherein the formula is a mean square error formula: ; wherein the method comprises the steps of For the mean square error, n is the number of data in both data sets,For the output probability result of the ith raw material ash content data in the ash content data set and the ith raw material yield in the raw material yield data set through the multi-layer perceptron,The method comprises the steps of presetting an actual label for an ith;
back propagation: calculating the gradient of the parameters of each layer to the loss function from the output layer, forward propagating the gradient layer by layer, and calculating the weight of each layer and the partial derivative of the bias parameters to the loss function, wherein the partial derivative formula is that ,,WhereinAndRespectively obtaining the weight from the input layer to the hidden layer and the bias results of the bias parameters of the hidden layer and the output layer; updating the weight and the bias parameter by calculating the partial derivative of the loss function on the weight and the bias parameter;
Updating parameters: according to the gradient obtained by calculation, setting a learning rate L, and randomly adjusting the weight from an input layer to a hidden layer and the bias parameters of the hidden layer and an output layer; updating parameters in the multi-layer perceptron, wherein an updating rule formula is as follows: 、 And ;
Repeating the iteration: repeatedly performing forward propagation and backward propagation until the model converges, namely, the model training of the multi-layer perceptron is completed;
collecting the current raw material ash content and the current raw material yield, taking the current raw material ash content and the current raw material yield as input, calculating an output probability result through a multi-layer perceptron model, and analyzing the current raw material risk degree by comparing the output probability result with a preset probability threshold;
when the output probability result exceeds the probability threshold, marking the current raw material as a high-risk-degree raw material; otherwise, the current stock is marked as a low risk level stock.
2. The method for data fusion based on the object data model according to claim 1, wherein:
in step S2, the neutral difference coefficient is calculated as the raw material acid-base difference, and the calculation formula is WhereinIs the coefficient of the neutral difference,Is the pH value of the raw material, 7 is the neutral pH value coefficient,
The raw material corrosion rate is calculated by calculating the raw material corrosion rate, and the raw material corrosion rate is represented by using a raw material corrosion rate numerical value, and a general corrosion rate formula can be used for calculating the raw material corrosion rate: wherein For the corrosion rate of the raw material,Is a constant, the specific value depends on the unit selected for the corrosive material,For the mass of the etching material before and after etching by the raw material,Is the surface area of the corrosion material, and time is the corrosion time;
Collecting raw material data of a plurality of production lines, and respectively combining acid-base difference normalized data and corrosion degree normalized data obtained by normalizing the collected raw material data into an acid-base difference data set and a corrosion degree data set; the normalization formula is as follows WhereinIs the result of the normalization process,Is the ith data in the acid-base difference data set or the corrosiveness data set,For the minimum value in the acid-base difference data set or the minimum value in the corrosiveness data set,The maximum value in the acid-base difference data set or the maximum value in the corrosiveness data set;
The acid-base difference data set and the corrosiveness data set are randomly sampled, acid-base difference normalization data in the acid-base difference data set are obtained, corrosiveness normalization data in the corrosiveness data set are obtained, covariance is calculated from the screened acid-base difference normalization data and corrosiveness normalization data to be used as a raw material classification threshold, and covariance is used as a raw material classification threshold;
the normalized data obtained by the above steps of the collected current raw material acid-base difference data and the current raw material corrosiveness data are used as the data in the calculated covariance formula AndMarking the calculation result asClassifying the current raw materials by comparing the calculation result with a raw material classification threshold;
If it is Marking the current raw material as a high-impurity raw material if the acid-base difference of the current raw material and the corrosiveness covariance of the current raw material exceed the raw material classification threshold; if it isMarking the current raw material as a low-impurity raw material if the acid-base difference of the current raw material and the corrosiveness covariance of the current raw material are lower than the raw material classification threshold;
If the current raw material is marked as a high-impurity raw material, modifying treatment is carried out by using a modifying agent with low heavy metal content; if the current feedstock is marked as a low impurity feedstock, a upgrading treatment is performed using a high heavy metal content upgrading agent.
3. The method for data fusion based on the object data model according to claim 1, wherein:
in step S4, if the current raw material is marked as a high risk level raw material, performing a remedial process on the current raw material; the method comprises a blending method, and comprises the following specific processes:
setting a blending proportion of a raw material with more impurities, namely a high-risk-degree raw material, and a raw material with lower impurities, namely a low-risk-degree raw material, mixing and blending, collecting the ash content of the current raw material, recording the ash content as HF, collecting the raw material with the lowest ash content in other production lines as a blending raw material, and marking the ash content as ; Determining and marking a reconciliation threshold asThe blending ratio is calculated by the following formula: Wherein t is the blending ratio;
and reducing the ash content of the current raw material to a blending threshold value through a blending proportion, and completing the remediation treatment of the high-risk-degree raw material.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410882438.0A CN118430702B (en) | 2024-07-03 | 2024-07-03 | Data fusion method based on object data model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410882438.0A CN118430702B (en) | 2024-07-03 | 2024-07-03 | Data fusion method based on object data model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN118430702A CN118430702A (en) | 2024-08-02 |
CN118430702B true CN118430702B (en) | 2024-09-20 |
Family
ID=92310775
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410882438.0A Active CN118430702B (en) | 2024-07-03 | 2024-07-03 | Data fusion method based on object data model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN118430702B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN86104452A (en) * | 1985-06-27 | 1987-04-29 | 泰克萨科开发公司 | Method for partical oxidation |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5405513A (en) * | 1994-07-08 | 1995-04-11 | Saudi Arabian Oil Company | Method and apparatus for an electrochemical test cell |
US6168709B1 (en) * | 1998-08-20 | 2001-01-02 | Roger G. Etter | Production and use of a premium fuel grade petroleum coke |
US7818156B2 (en) * | 2007-04-18 | 2010-10-19 | General Electric Company | Corrosion assessment method and system |
CN102768702B (en) * | 2012-07-02 | 2014-07-02 | 清华大学 | Oil refining production process schedule optimization modeling method on basis of integrated control optimization |
CN107578100B (en) * | 2017-08-31 | 2019-08-16 | 江苏康缘药业股份有限公司 | A kind of the result feedback method and device of Kernel-based methods parameter |
EP3795661B1 (en) * | 2019-09-17 | 2022-03-09 | Infineum International Limited | Asphaltene solvation and dispersion process |
US20240036025A1 (en) * | 2022-07-27 | 2024-02-01 | Phillips 66 Company | Quantitating corrosive naphthenic acids in crude oil |
CN116385237A (en) * | 2023-02-09 | 2023-07-04 | 新疆维吾尔自治区环境保护科学研究院 | Dangerous characteristic evaluation method and system for petroleum pollutants |
CN117127952A (en) * | 2023-09-15 | 2023-11-28 | 北京石大油源科技开发有限公司 | Underground in-situ modification system and method for hot steam-assisted multicomponent miscible fluid |
-
2024
- 2024-07-03 CN CN202410882438.0A patent/CN118430702B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN86104452A (en) * | 1985-06-27 | 1987-04-29 | 泰克萨科开发公司 | Method for partical oxidation |
Also Published As
Publication number | Publication date |
---|---|
CN118430702A (en) | 2024-08-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117195135B (en) | Water pollution anomaly traceability detection method and system | |
White et al. | Methodological tools | |
Garces et al. | Outliers detection in environmental monitoring databases | |
CN111338972A (en) | Machine learning-based software defect and complexity incidence relation analysis method | |
CN110619691B (en) | Prediction method and device for slab surface cracks | |
CN117517596B (en) | Method and system for monitoring combustible and toxic harmful gases in real time based on Internet of things | |
CN117538492B (en) | On-line detection method and system for pollutants in building space | |
Erginel | Fuzzy rule-based and control charts | |
CN112904810A (en) | Process industry nonlinear process monitoring method based on effective feature selection | |
CN119397163B (en) | Fixed pollution source data collection calibration method, system, medium and program product | |
CN111178672A (en) | Intelligent inspection method based on balance | |
CN118586765A (en) | A quality management risk control method and system based on big data | |
CN118430702B (en) | Data fusion method based on object data model | |
CN111898673A (en) | A Dissolved Oxygen Content Prediction Method Based on EMD and LSTM | |
CN118863552B (en) | A risk early warning method and system based on food business risk information database | |
CN118521324B (en) | True and false testing method for mixed-lap of multi-cigarette samples | |
CN118428608B (en) | Essence production quality full-flow tracing method and system based on data analysis | |
CN112949680A (en) | Pollution source identification method based on corresponding analysis and multiple linear regression | |
CN118333235A (en) | Behavior fraud risk prediction method and device and electronic equipment | |
CN117009767A (en) | Soil benchmark formulation and risk assessment method based on bioavailability | |
Lee et al. | Advanced yield enhancement: computer-based spatial pattern analysis. Part 1 | |
CN116380816A (en) | Method, program and storage medium for judging faults of coal quality online detection system | |
Cundeva-Blajer | Data Science for Testing, Inspection and Certification | |
Wahyono et al. | Application of AI to filter anomalous data from sensors in an online water quality monitoring system | |
CN117689114B (en) | A pollution monitoring system for groundwater |
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 |