CN114970375A - Rectification process monitoring method based on real-time sampling data - Google Patents
Rectification process monitoring method based on real-time sampling data Download PDFInfo
- Publication number
- CN114970375A CN114970375A CN202210901926.2A CN202210901926A CN114970375A CN 114970375 A CN114970375 A CN 114970375A CN 202210901926 A CN202210901926 A CN 202210901926A CN 114970375 A CN114970375 A CN 114970375A
- Authority
- CN
- China
- Prior art keywords
- rectification process
- monitoring
- state
- data
- real
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 139
- 230000008569 process Effects 0.000 title claims abstract description 116
- 238000012544 monitoring process Methods 0.000 title claims abstract description 64
- 238000005070 sampling Methods 0.000 title claims abstract description 50
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 239000013598 vector Substances 0.000 claims description 36
- 239000011159 matrix material Substances 0.000 claims description 22
- 239000000463 material Substances 0.000 claims description 10
- 238000013075 data extraction Methods 0.000 claims description 6
- 238000004821 distillation Methods 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 2
- 239000004576 sand Substances 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 239000007788 liquid Substances 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 6
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 239000007791 liquid phase Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 239000012071 phase Substances 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- ZRTQSJFIDWNVJW-WYMLVPIESA-N Lanoconazole Chemical compound ClC1=CC=CC=C1C(CS\1)SC/1=C(\C#N)N1C=NC=C1 ZRTQSJFIDWNVJW-WYMLVPIESA-N 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000010992 reflux Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000012824 chemical production Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 239000000498 cooling water Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000012808 vapor phase Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D3/00—Distillation or related exchange processes in which liquids are contacted with gaseous media, e.g. stripping
- B01D3/14—Fractional distillation or use of a fractionation or rectification column
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2477—Temporal data queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/17—Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Analysis (AREA)
- Databases & Information Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Strategic Management (AREA)
- Probability & Statistics with Applications (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Computational Linguistics (AREA)
- Fuzzy Systems (AREA)
- Algebra (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Chemical & Material Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Development Economics (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
Abstract
The invention relates to the technical field of rectification process monitoring, and provides a rectification process monitoring method based on real-time sampling data, which comprises the following steps: sampling the rectification process in real time, and performing correlation calculation on the sampled data to obtain actual state parameters of the rectification process; constructing a state prediction network to train under each working condition state in the rectification process to obtain network output under each working condition state to form a knowledge base, comparing the output of the real-time sampling data after training with the knowledge base for calculation, and further monitoring the rectification process; and constructing a conservation relation model according to each conservation parameter in the rectification process, and monitoring the rectification process to obtain a monitoring result. The invention solves the technical problems of inaccurate monitoring and overhigh cost in the prior art, and realizes the technical effects of high-precision monitoring and low cost.
Description
Technical Field
The invention relates to the technical field of rectification process monitoring, in particular to a rectification process monitoring method based on real-time sampling data.
Background
The rectification operation is an important link in the oil refining and chemical production process. The control process of the rectifying tower directly affects the product quality, yield and energy consumption, so the monitoring of the rectifying tower is highly regarded by people.
The rectifying tower is a multi-input and multi-output object, which is composed of multi-stage tower plates, has complex internal mechanism, slow control action, complex association relation among parameters and higher control requirement, and can monitor data of the rectifying process in real time to more effectively control the rectifying process, thereby greatly improving the safety and the production efficiency of industrial production, and the invention patent application number 202111414908.3 in China provides a soft measurement method for online detection of components in a special rectifying process, which mainly comprises the following steps: and selecting soft measurement model input variables by adopting principal component analysis random forest combination variables, introducing a generalized robust loss function into a limit gradient algorithm, optimizing a loss function hyperparameter by adopting a Bayes method, training by combining the soft measurement model input variables with the limit gradient algorithm to obtain a trained ARXGboost model, and substituting real-time data into the ARXGboost model for online detection.
However, in the process of implementing the technical scheme of the invention in the embodiment of the application, the inventor of the application finds that the technology at least has the technical problems of inaccurate measurement and overhigh cost.
Disclosure of Invention
The invention provides a rectification process monitoring method based on real-time sampling data, solves the technical problems of inaccurate monitoring and high cost in the prior art, and achieves the technical effects of high-precision monitoring and low cost.
The invention provides a rectification process monitoring method based on real-time sampling data, which specifically comprises the following technical scheme:
the rectification process monitoring method based on real-time sampling data comprises the following steps:
s1, sampling in real time in a rectification process, and performing correlation calculation on sampling data to obtain actual state parameters in the rectification process;
s2, constructing a state prediction network to train under each working condition state in the rectification process to obtain network output under each working condition state to form a knowledge base, comparing the output of the real-time sampling data after training with the knowledge base for calculation, and further monitoring the rectification process;
and S3, constructing a conservation relation model according to each conservation parameter in the rectification process, and monitoring the rectification process to obtain a monitoring result.
Further, the step S1 includes:
sampling in real time under the normal operation state of the rectifying tower to obtain sampling dataX，
，NWhich represents the number of samples to be sampled,
to representtThe data vector of the rectifying tower collected at the moment,
wherein each data vector contains direct samplingThe method comprises the steps of obtaining samples, obtaining reference data, obtaining the state parameters of the rectification process by directly obtaining the parameters through a calculation formula, obtaining more accurate state parameters by calculating the concentration of each component in the rectification process through an optimization formula, further sampling a data matrix, and providing more accurate parameter basis for monitoring the rectification process.
Further, the step S2 includes:
by defining the set of conditions occurring during the rectificationConAnd concentrating the working conditions into a state parameter set under each working condition
Training the state prediction network sequentially by defining the optimized connection weight vector to obtain the final state prediction network training result
(ii) a And all training results are stored in a knowledge base and used for providing reference basis with the actual working state of the rectification process.
Further, the step S2 includes:
a set of rectification process state vector data is sampled in real time
Obtaining an output state vector through a state prediction network
(ii) a Calculating output vectors of all working condition state sets stored in a knowledge base by defining an angle approximation function Sa
Approximate angle set from actual sampled data
Further, if
In a collectionFirst, thesIf the calculation result of each element is less than 15 degrees, the sampled data is considered as the firstsAnd the data state under different working conditions realizes the monitoring of the rectification process.
Further, the step S3 includes:
by constructing a balance model, calculating each large conservation law followed in the rectification process, and ensuring the normal operation of the rectification process; the specific model is constructed as follows:
wherein,Xa set of sampled data is represented as,Frepresenting a set of sampled data extractions,Wthe material conservation relation matrix is expressed,Hotthe heat conservation relation matrix is expressed,Qthe matrix is expressed in the conservation of energy relationship,Othrepresent the other array of conservation relationships,Ysa constraint matrix is shown to define conditions,Outrepresenting the output result of the model; aggregating sampled data using a sampled data extraction setXExtracting the state parameters of rectification process related to conservation, and then making them be related to conservation relation matrixW，Hot，Q，OthCalculating and comparing, constraining under the limited condition to obtain whether each conservation relation is satisfied, and obtaining the output resultOut。
The invention has at least the following technical effects or advantages:
1. according to the invention, the state parameter of the rectification process is acquired more accurately by defining the calculation of the state parameter of the rectification process, so that a data basis is provided for accurate monitoring, the monitoring accuracy is further improved, and the monitoring cost is reduced.
2. The method obtains the output vectors of the output layers under different working conditions by carrying out network training on the state parameters of the rectification process under various working conditions through the state prediction network, basically comprises all the working conditions possibly occurring in the rectification process, has reference comprehensiveness, and more accurately monitors the rectification process.
3. According to the invention, by optimizing the connection weight vector array in the state prediction network, each working condition in the rectification process is trained more quickly and accurately, and the monitoring cost is reduced.
4. The invention calculates the approximate angles of the real-time sampling data and the elements in the state sets of various working conditions of the knowledge base by defining the angle approximate function, and compares the approximate angles to obtain the working condition states corresponding to the sampling data, thereby more accurately realizing the monitoring of the rectification process.
5. According to the invention, by constructing the conservation relation model, the rectification process is monitored according to the multi-aspect conservation law and specific limiting conditions, the accuracy of the rectification process monitoring is ensured, the waste of materials in the rectification process is avoided, and the monitoring cost is further reduced.
Drawings
FIG. 1 is a flow diagram of a rectification process according to the present invention;
FIG. 2 is a step diagram of a rectification process monitoring method based on real-time sampling data according to the present invention.
In the figure: 1-feeding preheater, 2-tower bottom product cooler, 3-condenser, 4-tower top product cooler, 5-tower plate, 6-rectifying tower and 7-reboiler.
Detailed Description
The embodiment of the application provides a rectification process monitoring method based on real-time sampling data, solves the technical problems of inaccurate monitoring and high cost in the prior art, and comprises the following specific steps:
firstly, sampling in real time in a rectification process, and carrying out correlation calculation on sampling data to obtain actual state parameters in the rectification process; constructing a state prediction network for defining an optimized connection weight vector to train under each working condition state in the rectification process to obtain network output under each working condition state to form a knowledge base, comparing the output of the real-time sampling data after training with the knowledge base for calculation, and further monitoring the rectification process; and constructing a conservation relation model according to each conservation parameter in the rectification process, and monitoring the rectification process to obtain a monitoring result. The state parameters of the rectification process are acquired more accurately by defining the calculation of the state parameters of the rectification process, so that a data basis is provided for accurate monitoring, the monitoring accuracy is further improved, and the monitoring cost is reduced; output layer output vectors under different working conditions are obtained by carrying out network training on the rectification process state parameters under various working conditions through a state prediction network, all the working conditions possibly occurring in the rectification process are basically contained, the method has reference comprehensiveness, and the rectification process is more accurately monitored; by optimizing the connection weight vector array in the state prediction network, each working condition in the rectification process is trained more quickly and accurately, and the monitoring cost is reduced; the method comprises the steps of calculating the approximate angles of real-time sampling data and elements in various working condition state sets of a knowledge base by defining an angle approximate function, comparing the approximate angles to obtain the working condition states corresponding to the sampling data, and more accurately monitoring the rectification process; by constructing a conservation relation model, the rectification process is monitored according to the multi-aspect conservation law and specific limiting conditions, the monitoring accuracy of the rectification process is ensured, the waste of materials in the rectification process is avoided, and the monitoring cost is further reduced.
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
Referring to FIG. 1, the rectification process is generally carried out in a rectification column, wherein the feed is fed from the middle of the column, and under the action of a feed preheater, the column section above the feed plate is a rectification section for increasing the concentration of volatile components in the rising gas phase plate by plate, and the column plates inside and below the feed plate are stripping sections for extracting volatile components in the falling liquid phase plate by plate; wherein, the tower plates are the places for mass transfer and heat transfer of gas-liquid two phases, and the gas-liquid two phases on each tower plate carry out bidirectional mass transfer; leading the steam led out from the tower top to a condenser, taking part of condensate as reflux liquid to return to the rectifying tower from the tower top, and taking the rest distillate as a tower top product; the liquid extracted from the tower bottom is partially gasified by a reboiler, the steam rises along the tower, and the rest liquid is used as a tower bottom product.
Referring to the attached figure 2, the rectification process monitoring method based on real-time sampling data comprises the following steps:
s1, sampling in real time in a rectification process, and performing correlation calculation on sampling data to obtain actual state parameters in the rectification process;
s11, storing and calculating the acquired data of the rectification process to obtain relevant state parameters of the rectification process;
sampling in real time under the running state of the rectifying tower to obtain sampling dataX，
，NWhich represents the number of samples to be sampled,
to representtThe data vector of the rectifying tower collected at the moment,
wherein each data vector comprises a rectification process state parameter obtained by direct sampling and obtained by utilizing a direct acquisition parameter through a calculation formula:
inlet flow rate
Temperature at inlet
Pressure at the top of the column
Outlet flow rate
Amount of reflux
Component 1 outlet flow
Component 2 outlet flow
… …, component (a)mOutlet flow rate
Outlet flow of liquid from the bottom of the column
Amount of liquid refluxed from the bottom of the column
，mTemperature of the column trays
，mPressure of the column plate
Cooling water flow of condenser
Steam flow of reboiler
Concentration of light component output at tower top and tower bottom, and heating quantity of reboiler
Cooling capacity of condenser
Concentration of light component output from each column plateCEnthalpy of feed product
Enthalpy of the overhead product
Enthalpy of bottom product
And other related calculated state parameters to obtain a set of state parameters
，
，MTo representtThe number of the state parameters of the rectification process contained in the data vector of the rectification tower collected at any moment.
In particular, the volatility of each component is obtained by looking up relevant data according to the feeding property
And other related component properties, and further calculating to obtain relative volatility and other related component characteristic parameters;
from the above, it can be known that the data matrix is sampledXCan be expressed as:
s12, further limiting the distillation process state parameters in the step S11;
calculating according to the optimized antoin equation and the gas-liquid phase equilibrium equation to obtain the concentration of the light component output:
wherein,
、
is shown astAt the sampling time ofiThe concentration of the effluent of the tray components in the liquid and vapor phases,
is shown astAt the sampling time ofiThe pressure of the tower plates of the layers,
indicates the temperature of the ith tray at the sampling time t,
，
，
，
is the antoin constant.
According to the invention, the state parameter of the rectification process is acquired more accurately by defining the calculation of the state parameter of the rectification process, so that a data basis is provided for accurate monitoring, the monitoring accuracy is further improved, and the monitoring cost is reduced.
S2, constructing a state prediction network to train under each working condition state in the rectification process to obtain network output under each working condition state to form a knowledge base, comparing the output of the real-time sampling data after training with the knowledge base for calculation, and further monitoring the rectification process;
the operating conditions include normality, feed flow fluctuation, tower bottom temperature sensor failure and other operational problems that can occur during the rectification process.
S21, collecting state parameters under various working conditions in the rectification process;
the invention defines the working condition set occurring in the rectification process as Con,
，qthe total number of the working conditions is shown,
is shown askOperating conditions in which the elements are in each caseLAnd (4) setting state parameters of the rectification process.
Defining parameters of the state prediction network:
will do soSet of State parameters for each Condition of a set of Conditions
Sequentially used as the input of the state prediction network, the number of the obtained input training vectors is L, and the number of the corresponding obtained network input nodes (the dimension of the input mode vector) is LM(ii) a Defining a network connection weight matrixW，
，RExpressing the number of neurons in an output layer;
network learning rate:
in general, are
，
，
Which represents the initial learning rate of the initial learning,nrepresents the number of learning times, Num represents the total number of learning times, and decreases as the learning time increases;
defining a neighborhood
As neurons of the output layerjIs related to the number of nodes it contains, and is followed bynThe increase of (2) is gradually reduced, and the neighborhood is determined by the determined winning neuron nodejA central, area containing several neurons, which area may be of any shape, but is generally uniformly symmetrical, a generally square or circular area,
，
the expression is to be taken to the whole,
represent
The initial value of (c).
An input layer: for input training vector
Carrying out normalization processing to obtain normalized data vector
To network connection weight matrix
To (1)jNetwork connection weight vector of individual neurons
Normalization is carried out to obtain a normalized network connection weight vector
；
Compute input layer
Connecting weight vectors to each output neuron node
European distance of
. Note the book
Is input at n times to the layeriThe output of each of the plurality of neurons,
representing input neuronsiTo output layer neuronsjThe invention defines the Euclidean distance
Comprises the following steps:
further, a minimum distance output node is calculatedjAs the best node, note
The invention adjusts and combines neurons
And the connection weight vectors of all neuron nodes in its neighborhood,
taking the next input vector as a training vector set, normalizing the next input vector, and calculating the Euclidean distance again until the training is finished
Obtaining the state prediction network training result
。
Training the elements in the formed working condition set item by item to obtain the final state prediction network training result
(ii) a And all training results are stored in a knowledge base and used for providing reference basis with the actual working state of the rectification process.
The method obtains the output vectors of the output layers under different working conditions by carrying out network training on the state parameters of the rectification process under various working conditions through the state prediction network, basically comprises all the working conditions possibly occurring in the rectification process, has reference comprehensiveness, and more accurately monitors the rectification process.
According to the invention, by optimizing the connection weight vector array in the state prediction network, each working condition in the rectification process is trained more quickly and accurately, and the monitoring cost is reduced.
S22, calculating and monitoring the real-time sampling data according to the output result of the state prediction network;
real-time sampling of a set of distillation process state vector data
Obtaining an output state vector through a state prediction network
. The invention calculates the output vector of each working condition state set stored in a knowledge base by defining an angle approximate function Sa
The set of approximate angles to the actual sampled data is:
then
Further, if
First in a setsIf the calculation result of each element is less than 15 degrees, the sampled data is considered as the firstsData state under various conditions.
The invention calculates the approximate angles of the real-time sampling data and the elements in the state sets of various working conditions of the knowledge base by defining the angle approximate function, and compares the approximate angles to obtain the working condition states corresponding to the sampling data, thereby more accurately realizing the monitoring of the rectification process.
And S3, comparing all conservation parameters in the rectification process to construct a conservation relation model, and monitoring the rectification process to obtain a monitoring result.
The invention calculates each large conservation law followed in the rectification process by constructing a conservation relation model, and ensures the normal operation of the rectification process. The specific model is constructed as follows:
wherein,Xa set of sampled data is represented as,Frepresenting a set of sampled data extractions,Wthe material conservation relation matrix is expressed,Hotthe heat conservation relation matrix is expressed,Qthe matrix is expressed in the conservation of energy relationship,Othrepresent the other array of conservation relationships,Ysa constraint matrix is shown to define conditions,Outrepresenting the output result of the model;
in particular, the qualifying conditional constraint matrixYsThe settings are made by the staff according to the specific rectification.
In particular, the conservation relation matrix comprises conservation relations existing at any stage in the rectification process, such as element inclusion, total material conservation relation, material conservation relation of light components and other related material conservation relations in the material conservation relation matrix;
aggregating sampled data using a sampled data extraction setXExtracting the state parameters of rectification process related to conservation, and then making them be related to conservation relation matrixW，Hot，Q，OthCalculating and comparing, constraining under the limited condition to obtain whether each conservation relation is satisfied, and obtaining the output resultOut. The specific process is as follows:
according to the invention, by constructing the conservation relation model, the rectification process is monitored according to the multi-aspect conservation law and specific limiting conditions, the accuracy of the rectification process monitoring is ensured, the waste of materials in the rectification process is avoided, and the monitoring cost is further reduced.
In conclusion, the rectification process monitoring method based on real-time sampling data is completed.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (5)
1. A rectification process monitoring method based on real-time sampling data is characterized by comprising the following steps:
s1, sampling in real time in a rectification process, and performing correlation calculation on sampling data to obtain actual state parameters in the rectification process;
s2, constructing a state prediction network to train under each working condition state in the rectification process to obtain network output under each working condition state to form a knowledge base, comparing the output of the real-time sampling data after training with the knowledge base for calculation, and further monitoring the rectification process;
and S3, constructing a conservation relation model according to each conservation parameter in the rectification process, and monitoring the rectification process to obtain a monitoring result.
2. The method for monitoring rectification process based on real-time sampling data as claimed in claim 1, wherein the step S1 includes:
sampling in real time under the normal operation state of the rectifying tower to obtain sampling dataX，
，NWhich represents the number of samples to be sampled,
to representtThe data vector of the rectifying tower collected at the moment,
wherein each data vector comprises direct sampling, data acquisition, distillation process state parameters obtained by using direct acquisition parameters through a calculation formula, and more accurate concentration of each component in the distillation process obtained by using an optimization formulaThe state parameter of (2).
3. The method for monitoring rectification process based on real-time sampling data as claimed in claim 1, wherein the step S2 includes:
by defining the set of conditions occurring during the rectificationConAnd concentrating the working conditions into a state parameter set under each working condition
Training the state prediction network sequentially by defining an optimized connection weight vector to obtain a final state prediction network training result
(ii) a And all training results are stored in a knowledge base and used for providing reference basis with the actual working state of the rectification process.
4. The method for monitoring rectification process based on real-time sampling data as claimed in claim 3, wherein the step S2 includes:
a set of rectification process state vector data is sampled in real time
Obtaining an output state vector through a state prediction network
(ii) a Calculating output vectors of all working condition state sets stored in a knowledge base by defining an angle approximation function Sa
Approximate angle set from actual sampled data
Further, if
First in a setsIf the calculation result of each element is less than 15 degrees, the sampled data is considered as the firstsAnd the data state under different working conditions realizes the monitoring of the rectification process.
5. The method for monitoring rectification process based on real-time sampling data as claimed in claim 1, wherein the step S3 includes:
by constructing a balance model, calculating each large conservation law followed in the rectification process, and ensuring the normal operation of the rectification process; the specific model is constructed as follows:
wherein,Xa set of sampled data is represented as,Frepresenting a set of sampled data extractions,Wthe material conservation relation matrix is represented,Hotthe heat conservation relation matrix is expressed,Qthe matrix is expressed in the conservation of energy relationship,Othrepresent the other array of conservation relationships,Ysa constraint matrix is shown to define conditions,Outrepresenting the output result of the model; aggregating sampled data using a sampled data extraction setXExtracting the state parameters of rectification process related to conservation, and then making them be related to conservation relation matrixW，Hot，Q，OthCalculating and comparing, constraining under the limited condition to obtain whether each conservation relation is satisfied, and obtaining the output resultOut。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210901926.2A CN114970375B (en) | 2022-07-29 | 2022-07-29 | Rectification process monitoring method based on real-time sampling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210901926.2A CN114970375B (en) | 2022-07-29 | 2022-07-29 | Rectification process monitoring method based on real-time sampling data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114970375A true CN114970375A (en) | 2022-08-30 |
| CN114970375B CN114970375B (en) | 2022-11-04 |
Family
ID=82968541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210901926.2A Active CN114970375B (en) | 2022-07-29 | 2022-07-29 | Rectification process monitoring method based on real-time sampling data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114970375B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117385106A (en) * | 2023-12-12 | 2024-01-12 | 山东飞扬化工有限公司 | Rapid calculation method of feeding amount, feeding method and feeding system |
Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1693297A (en) * | 2005-03-16 | 2005-11-09 | 扬子石油化工股份有限公司 | Process for advanced controlling rectifying apparatus of butadiene |
| US20070038333A1 (en) * | 2005-08-15 | 2007-02-15 | Dadebo Solomon A | Model predictive control having application to distillation |
| CN101073712A (en) * | 2006-12-26 | 2007-11-21 | 浙江大学 | Method and system for controlling high-purity rectification of rectifying tower based on generalized prediction control |
| CN101573668A (en) * | 2006-10-31 | 2009-11-04 | 搭篷技术公司 | Integrated model predictive control of distillation and dehydration sub-processes in a biofuel production process |
| US20100082139A1 (en) * | 2008-09-30 | 2010-04-01 | Rockwell Automation Technologies, Inc. | Energy optimizer for dehydrating biofuels through distillation towers and molecular sieves |
| CN102053613A (en) * | 2010-12-31 | 2011-05-11 | 北京三博中自科技有限公司 | Energy real-time monitoring system of industrial process equipment and monitoring method thereof |
| CN102183892A (en) * | 2011-05-10 | 2011-09-14 | 上海交通大学 | Load change energy consumption optimizing control method of three-column methanol distillation system |
| CN107096252A (en) * | 2017-05-04 | 2017-08-29 | 万华化学集团股份有限公司 | The method that rectifying column tower top cold is automatically controlled |
| CN108287474A (en) * | 2017-12-27 | 2018-07-17 | 上海交通大学 | Based on the probabilistic catalytic reforming reactor robust operation optimization method of raw material |
| CN109669017A (en) * | 2017-10-17 | 2019-04-23 | 中国石油化工股份有限公司 | Refinery's distillation tower top based on deep learning cuts water concentration prediction technique |
| CN110222372A (en) * | 2019-05-08 | 2019-09-10 | 中国水利水电科学研究院 | A kind of Flow of River water quality real-time predicting method and device based on data assimilation |
| CN111914889A (en) * | 2020-06-13 | 2020-11-10 | 宁波大学 | Rectifying tower abnormal state identification method based on brief kernel principal component analysis |
| CN111905396A (en) * | 2020-06-13 | 2020-11-10 | 宁波大学 | Rectification process real-time monitoring method based on online sampling data driving |
| CN112633406A (en) * | 2020-12-31 | 2021-04-09 | 天津大学 | Knowledge distillation-based few-sample target detection method |
| CN112966399A (en) * | 2021-04-15 | 2021-06-15 | 苏州大学张家港工业技术研究院 | Pulse tube refrigerator working condition prediction method and system based on machine learning |
| CN112989920A (en) * | 2020-12-28 | 2021-06-18 | 华东理工大学 | Electroencephalogram emotion classification system based on frame-level feature distillation neural network |
| CN113051828A (en) * | 2021-03-30 | 2021-06-29 | 重庆大学 | Online prediction method for natural gas water dew point driven by technological parameters |
| CN113433906A (en) * | 2021-06-24 | 2021-09-24 | 之江实验室 | Method for product prediction and distillation operation parameter optimization of distillation device |
| CN113537400A (en) * | 2021-09-14 | 2021-10-22 | 浙江捷瑞电力科技有限公司 | Branch neural network-based edge computing node allocation and exit method |
| CN113723686A (en) * | 2021-08-31 | 2021-11-30 | 江南大学 | Multitask ash box prediction method and system for energy consumption in organic silicon monomer fractionation process |
| WO2022016556A1 (en) * | 2020-07-24 | 2022-01-27 | 华为技术有限公司 | Neural network distillation method and apparatus |
| CN114239430A (en) * | 2021-12-03 | 2022-03-25 | 东南大学 | Method and system for predicting NOx at furnace outlet based on numerical simulation |
| CN114240243A (en) * | 2021-12-30 | 2022-03-25 | 无锡雪浪数制科技有限公司 | Rectifying tower product quality prediction method and device based on dynamic system identification |
| CN114460848A (en) * | 2022-02-08 | 2022-05-10 | 万华化学集团股份有限公司 | Method and device for controlling sensitive plate of rectifying tower and computer readable storage medium |
| CN114741772A (en) * | 2022-05-09 | 2022-07-12 | 北京化工大学 | Method for solving design problem of partition wall rectifying tower based on heuristic algorithm |
-
2022
- 2022-07-29 CN CN202210901926.2A patent/CN114970375B/en active Active
Patent Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1693297A (en) * | 2005-03-16 | 2005-11-09 | 扬子石油化工股份有限公司 | Process for advanced controlling rectifying apparatus of butadiene |
| US20070038333A1 (en) * | 2005-08-15 | 2007-02-15 | Dadebo Solomon A | Model predictive control having application to distillation |
| CN101573668A (en) * | 2006-10-31 | 2009-11-04 | 搭篷技术公司 | Integrated model predictive control of distillation and dehydration sub-processes in a biofuel production process |
| CN101073712A (en) * | 2006-12-26 | 2007-11-21 | 浙江大学 | Method and system for controlling high-purity rectification of rectifying tower based on generalized prediction control |
| US20100082139A1 (en) * | 2008-09-30 | 2010-04-01 | Rockwell Automation Technologies, Inc. | Energy optimizer for dehydrating biofuels through distillation towers and molecular sieves |
| CN102053613A (en) * | 2010-12-31 | 2011-05-11 | 北京三博中自科技有限公司 | Energy real-time monitoring system of industrial process equipment and monitoring method thereof |
| CN102183892A (en) * | 2011-05-10 | 2011-09-14 | 上海交通大学 | Load change energy consumption optimizing control method of three-column methanol distillation system |
| CN107096252A (en) * | 2017-05-04 | 2017-08-29 | 万华化学集团股份有限公司 | The method that rectifying column tower top cold is automatically controlled |
| CN109669017A (en) * | 2017-10-17 | 2019-04-23 | 中国石油化工股份有限公司 | Refinery's distillation tower top based on deep learning cuts water concentration prediction technique |
| CN108287474A (en) * | 2017-12-27 | 2018-07-17 | 上海交通大学 | Based on the probabilistic catalytic reforming reactor robust operation optimization method of raw material |
| CN110222372A (en) * | 2019-05-08 | 2019-09-10 | 中国水利水电科学研究院 | A kind of Flow of River water quality real-time predicting method and device based on data assimilation |
| CN111905396A (en) * | 2020-06-13 | 2020-11-10 | 宁波大学 | Rectification process real-time monitoring method based on online sampling data driving |
| CN111914889A (en) * | 2020-06-13 | 2020-11-10 | 宁波大学 | Rectifying tower abnormal state identification method based on brief kernel principal component analysis |
| WO2022016556A1 (en) * | 2020-07-24 | 2022-01-27 | 华为技术有限公司 | Neural network distillation method and apparatus |
| CN112989920A (en) * | 2020-12-28 | 2021-06-18 | 华东理工大学 | Electroencephalogram emotion classification system based on frame-level feature distillation neural network |
| CN112633406A (en) * | 2020-12-31 | 2021-04-09 | 天津大学 | Knowledge distillation-based few-sample target detection method |
| CN113051828A (en) * | 2021-03-30 | 2021-06-29 | 重庆大学 | Online prediction method for natural gas water dew point driven by technological parameters |
| CN112966399A (en) * | 2021-04-15 | 2021-06-15 | 苏州大学张家港工业技术研究院 | Pulse tube refrigerator working condition prediction method and system based on machine learning |
| CN113433906A (en) * | 2021-06-24 | 2021-09-24 | 之江实验室 | Method for product prediction and distillation operation parameter optimization of distillation device |
| CN113723686A (en) * | 2021-08-31 | 2021-11-30 | 江南大学 | Multitask ash box prediction method and system for energy consumption in organic silicon monomer fractionation process |
| CN113537400A (en) * | 2021-09-14 | 2021-10-22 | 浙江捷瑞电力科技有限公司 | Branch neural network-based edge computing node allocation and exit method |
| CN114239430A (en) * | 2021-12-03 | 2022-03-25 | 东南大学 | Method and system for predicting NOx at furnace outlet based on numerical simulation |
| CN114240243A (en) * | 2021-12-30 | 2022-03-25 | 无锡雪浪数制科技有限公司 | Rectifying tower product quality prediction method and device based on dynamic system identification |
| CN114460848A (en) * | 2022-02-08 | 2022-05-10 | 万华化学集团股份有限公司 | Method and device for controlling sensitive plate of rectifying tower and computer readable storage medium |
| CN114741772A (en) * | 2022-05-09 | 2022-07-12 | 北京化工大学 | Method for solving design problem of partition wall rectifying tower based on heuristic algorithm |
Non-Patent Citations (4)
| Title |
|---|
| 卢政印: "基于深度学习的精馏过程质量参数软测量方法研究", 《中国优秀硕士学位论文全文数据库》 * |
| 王卫霞等: "工程化精馏实训装置操作条件优化研究", 《黑龙江科技信息》 * |
| 由宏君: "精馏塔的故障诊断方案设计", 《贵州化工》 * |
| 魏奇业等: "基于神经网络的精馏塔动态模拟", 《吉林化工学院学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117385106A (en) * | 2023-12-12 | 2024-01-12 | 山东飞扬化工有限公司 | Rapid calculation method of feeding amount, feeding method and feeding system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114970375B (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104965967B (en) | A kind of yield real-time predicting method of atmospheric and vacuum distillation unit | |
| CN108803520B (en) | Dynamic process monitoring method based on variable nonlinear autocorrelation rejection | |
| CN108776831A (en) | A kind of complex industrial process Data Modeling Method based on dynamic convolutional neural networks | |
| CN114970375B (en) | Rectification process monitoring method based on real-time sampling data | |
| CN108469805B (en) | Distributed dynamic process monitoring method based on dynamic optimal selection | |
| CN110210687A (en) | A kind of Nonlinear Dynamic production process product quality prediction technique returned based on local weighted slow feature | |
| CN111142494B (en) | Intelligent control method and system for amine liquid regeneration device | |
| CN113420500B (en) | Intelligent atmospheric and vacuum system | |
| CN110362063B (en) | Fault detection method and system based on global maintenance unsupervised kernel extreme learning machine | |
| CN112597430A (en) | Method for optimizing operation parameters of complex rectifying tower | |
| Aguel et al. | Parametric study and modeling of cross-flow heat exchanger fouling in phosphoric acid concentration plant using artificial neural network | |
| CN103323484A (en) | Method for predicting crystallization state of sugarcane sugar boiling process | |
| Xie et al. | A hierarchical data reconciliation based on multiple time-delay interval estimation for industrial processes | |
| Li et al. | Knowledge-based operation optimization of a distillation unit integrating feedstock property considerations | |
| CN115014454A (en) | Soft measurement method, system, equipment and medium for main steam flow of thermal power generating unit | |
| CN104865944A (en) | Gas fractionation device control system performance evaluation method based on PCA (Principle Component Analysis)-LSSVM (Least Squares Support Vector Machine) | |
| CN114492551A (en) | Soft measurement strategy based on improved GWO-SVM | |
| CN116821695A (en) | Semi-supervised neural network soft measurement modeling method | |
| CN115631804A (en) | Method for predicting outlet concentration of sodium aluminate solution in evaporation process based on data coordination | |
| CN116662925A (en) | Industrial process soft measurement method based on weighted sparse neural network | |
| CN105259790A (en) | Time registration method for multi-parameter data during alumina production and evaporation process | |
| CN113723686B (en) | Multitask ash box prediction method and system for energy consumption in organic silicon monomer fractionation process | |
| CN111221252B (en) | Predictive controller parameter analysis setting method for industrial system with fractional hysteresis | |
| CN107694139B (en) | A kind of plate distillation column coefficient of performance on-line analysis | |
| CN116720138A (en) | Chemical tower equipment online HAZOP quantitative analysis and risk prediction method |
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 |