CN115169721A - Single-ton energy consumption prediction method and system in rectification process based on migration identification - Google Patents
Single-ton energy consumption prediction method and system in rectification process based on migration identification Download PDFInfo
- Publication number
- CN115169721A CN115169721A CN202210878367.8A CN202210878367A CN115169721A CN 115169721 A CN115169721 A CN 115169721A CN 202210878367 A CN202210878367 A CN 202210878367A CN 115169721 A CN115169721 A CN 115169721A
- Authority
- CN
- China
- Prior art keywords
- rectification process
- energy consumption
- ton
- migration
- matrix
- 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
Images
Classifications
-
- 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
- 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
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- 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
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- 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—Systems or methods specially adapted for 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
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Mathematical Optimization (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention discloses a rectification process single-ton energy consumption prediction method and system based on migration identification.
Description
Technical Field
The invention relates to the technical field of process industrial production and processing, in particular to a method and a system for predicting single-ton energy consumption in a rectification process based on migration identification.
Background
In the process industry, the rectification process is the separation process with the widest application, and the energy consumption of the rectification process occupies about 40 percent of the energy consumption of the whole industry. In actual production, in order to ensure the product to be qualified, conservative operation leads most of energy consumed in the rectification process to be carried away by cooling water or separation components. Therefore, the energy saving potential of the rectification process is huge. The common energy consumption prediction method lacks correlation with product quality, and accurate prediction of energy consumption required by producing single ton of qualified products in the rectification process has great guiding significance for energy conservation and consumption reduction.
The identification technique utilizes process history data to determine the structure and parameters of a single ton energy consumption prediction mathematical model. High-precision prediction models require a large amount of high-quality historical data, resulting in high identification cost. And in consideration of the similarity existing among different rectification processes, the reasonable utilization of the knowledge in the identified process model is beneficial to improving the identification precision and reducing the identification cost. The conventional identification technology without migration is limited by the assumption of independent and same distribution, and it is difficult to effectively combine the knowledge in the identification process with the process to be identified.
Disclosure of Invention
The invention aims to provide a single-ton energy consumption prediction method based on migration identification, which has high feasibility and high precision.
In order to solve the problems, the invention provides a method for predicting the energy consumption of a single ton in a rectification process based on migration identification, which comprises the following steps:
s1, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector;
s2, collecting historical data of the rectification process A, and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process A
S3, defining a single-ton energy consumption model parameter vector of the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the vector Y is output B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix;
s4, selecting a migration identification criterion function, and minimizing the migration identification criterion function to solve a migration gain matrix G;
and S5, substituting the solved migration gain matrix G into the parameter migration identification model to obtain the single-ton energy consumption predicted value of the rectification process B.
As a further improvement of the present invention, step S1 includes:
s11, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
y(k)+a 1 y(k-1)+…+a p y(k-p)=b 0 u(k)+b 1 u(k-1)+…+b q u(k-q)+v(k) (1)
wherein y (k) is an output variable at the time of k; u (k) is an input variable at the moment k; a is 1 ,a 2 ,…,a p ,b 0 ,b 1 ,…,b q Is a single ton energy consumption model parameter; p and q are the order of the single-ton energy consumption model; v (k) is measurement noise;
s12, defining output vectors, observation matrixes, parameter vectors and noise vectors as follows:
Y=[y(p+1),y(p+2)…,y(p+N)] T
θ=[a 1 ,…,a p ,b 0 ,…,b q ] T
V=[v(p+1),v(p+2)…,v(p+N)] T
wherein N is the length of the acquired data;
according to the definition, the single-ton energy consumption model (1) of the rectification process is equivalent to:
Y=Xθ+V
namely, equation (2).
As a further improvement of the present invention, step S4 includes:
s41, selecting a migration identification criterion function:
wherein, theta B The actual value of the single-ton energy consumption model parameter matrix in the rectification process B is obtained; e [ ·]Calculating the mean value; trace {. Is the trace operation of the matrix; j (-) is a criterion function for G;
and S42, substituting the formula (3) into the formula (4) to obtain:
J(G)=trace{E[(I-GX B )R(I-GX B ) T +GΣ B G T ]} (5)
wherein, the first and the second end of the pipe are connected with each other,single ton energy consumption model parameter moment for rectification process A and rectification process BThe variance of the array difference is then calculated,the difference of the single-ton energy consumption model parameter matrixes of the rectification process A and the rectification process B, wherein I is an identity matrix, sigma B Is the noise covariance matrix of rectification process B.
Taking the first derivative of the criterion function (5) with respect to G and making it equal to 0, one can obtain:
as a further improvement of the present invention, step S4 further includes:
s44, converting the (p) into (p) by applying a matrix inversion theorem B +q B )×(p B +q B ) The inversion of the dimensional matrix, the simplification of the available migration gain matrix, can be achieved by:
wherein, the first and the second end of the pipe are connected with each other,the transfer gain matrix G is a Fisher information matrix for single-ton energy consumption model parameter identification of the rectification process B and depends on the variance of the difference of model parameter matrixes of the rectification processes A and BAnd the quality F of the collected observation data of the rectification process B.
As a further improvement of the invention, the output variable of the single-ton energy consumption model in the rectification process is the energy consumption required by producing single-ton qualified products, and the input variable is the feeding flow, the reflux quantity and the cooling water flow required by producing single-ton qualified products.
As a further improvement of the invention, the measurement noise includes white noise that follows gaussian distribution, t distribution, poisson distribution.
As a further improvement of the invention, the historical data comprises four variables of feed flow, reflux quantity, cooling water flow and single ton energy consumption.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the program.
The invention also provides a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of any of the methods described above.
The invention also provides a system for predicting the energy consumption of single ton in the rectification process based on migration identification, which comprises the following steps:
the single-ton energy consumption model establishing module is used for dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at the sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector;
single-ton energy consumption model order and parameter vector determination of rectification process AA fixed module for collecting the historical data of the rectification process A and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process A
And the parameter migration identification model construction module of the single-ton energy consumption model in the rectification process B is used for defining the parameter vector of the single-ton energy consumption model in the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the vector Y is output B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix;
the migration gain matrix solving module is used for selecting a migration identification criterion function and minimizing the migration identification criterion function so as to solve the migration gain matrix G;
and the single-ton energy consumption prediction module of the rectification process B is used for substituting the solved migration gain matrix G into the parameter migration identification model to obtain a single-ton energy consumption prediction value of the rectification process B.
The invention has the beneficial effects that:
the parameter migration identification model provided by the invention aims at minimizing identification errors, is different from the traditional identification technology without migration by only using data knowledge of the current process to be identified, and achieves higher identification precision under the same identification cost by determining a migration gain matrix and using data of the process to be identified to adjust the model of the identified process, thereby realizing accurate prediction of single ton energy consumption in the rectification process by using a small amount of historical data.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical means of the present invention more clearly understood, the present invention may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present invention more clearly understood, the following preferred embodiments are described in detail with reference to the accompanying drawings.
Drawings
FIG. 1 is a flow chart of a method for predicting the energy consumption per ton of a rectification process based on migration identification in a preferred embodiment of the present invention;
FIG. 2 is a data plot of feed flow, reflux, and cooling water flow for distillation process A in a preferred embodiment of the present invention;
FIG. 3 is a predicted curve for the energy consumption per ton for distillation process A in a preferred embodiment of the present invention;
FIG. 4 is a data plot of feed flow, reflux, and cooling water flow for distillation process B in a preferred embodiment of the present invention;
fig. 5 is a predicted curve of the energy consumption per ton for rectification process B in a preferred embodiment of the invention.
Detailed Description
The present invention is further described below in conjunction with the drawings and the embodiments so that those skilled in the art can better understand the present invention and can carry out the present invention, but the embodiments are not to be construed as limiting the present invention.
As shown in fig. 1, a method for predicting energy consumption per ton in a rectification process based on migration identification in a preferred embodiment of the present invention includes the following steps:
s1, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector; wherein, the output variable of the single-ton energy consumption model in the rectification process is the energy consumption required by producing single-ton qualified products, and the input variable is the feeding flow, the reflux quantity and the cooling water flow required by producing single-ton qualified products.
Specifically, step S1 includes:
s11, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
y(k)+a 1 y(k-1)+…+a p y(k-p)=b 0 u(k)+b 1 u(k-1)+…+b q u(k-q)+v(k) (1)
wherein y (k) is an output variable at the moment k; u (k) is an input variable at the moment k; a is 1 ,a 2 ,…,a p ,b 0 ,b 1 ,…,b q Is a single ton energy consumption model parameter; p and q are the order of the single-ton energy consumption model; v (k) is measurement noise; the measurement noise includes white noise that follows a gaussian distribution, a t distribution, and a poisson distribution.
Step S12, defining output vectors, observation matrixes, parameter vectors and noise vectors as follows:
Y=[y(p+1),y(p+2)…,y(p+N)] T
θ=[a 1 ,…,a p ,b 0 ,…,b q ] T
V=[v(p+1),v(p+2)…,v(p+N)] T
wherein N is the length of the acquired data;
according to the definition, the single-ton energy consumption model (1) of the rectification process is equivalent to:
Y=Xθ+V
namely, equation (2).
S2, collecting historical data of the rectification process A, and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process AThe historical data comprises four variables of feeding flow, reflux quantity, cooling water flow and single-ton energy consumption.
After collecting the historical data of the rectification process A, the method further comprises the step of preprocessing the data, and specifically comprises the following steps:
step 1: rejecting non-numerical sample points in the historical data of the rectification process A, and rejecting abnormal working condition data according to the working condition records; and 2, step: removing outliers in historical data of the rectification process A, wherein the method comprises cluster analysis, a 3 sigma rule, a neighbor method, box body diagram analysis and the like; step three: filling missing values in historical data in the rectification process A, wherein the method comprises mean filling, median filling, mode filling, machine learning algorithm filling and the like; step four: the method for eliminating the dimensional difference of four variables of feed flow, reflux quantity, cooling water flow and single ton energy consumption in the rectification process A comprises the following steps: normalization, z-score normalization, centralization, hellinger transformation, parleton normalization, and the like.
Optionally, the method for determining the model order of the rectification process a includes a Hankel matrix method, an AIC criterion order method, a BIC criterion order method, an FPE criterion order method, a residual error method, an F test method, and the like.
Optionally, the method for determining the model parameters of the rectification process a includes least square estimation, maximum likelihood estimation, prediction error method, random approximation method, support vector regression, partial least square, neural network, and the like.
S3, defining a single-ton energy consumption model parameter vector of the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the vector Y is output B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix; dimension of G and data length N B It is relevant. The migration gain matrix G depends on the variance of the differences of the model parameters of the rectification processes a and B and the quality of the collected observation data of the rectification process B.
Wherein, the step of pretreatment can refer to the rectification process A.
S4, selecting a migration identification criterion function, and minimizing the migration identification criterion function to solve a migration gain matrix G; the goal of migration identification is to obtain an estimate of the migration parameters in the least variance sense.
Specifically, step S4 includes:
s41, selecting a migration identification criterion function:
wherein, theta B The actual value of the single-ton energy consumption model parameter matrix in the rectification process B is obtained; e [ ·]Calculating the mean value; trace {. Is trace operation of a matrix; j (-) is a criterion function for G;
and S42, substituting the formula (3) into the formula (4) to obtain:
J(G)=trace{E[(I-GX B )R(I-GX B ) T +GΣ B G T ]} (5)
wherein the content of the first and second substances,is the variance of the difference of the single-ton energy consumption model parameter matrixes in the rectification process A and the rectification process B,the difference of the single-ton energy consumption model parameter matrix of the rectification process A and the rectification process B, wherein I is an identity matrix, sigma B Is the noise covariance matrix of rectification process B.
Taking the first derivative of the criterion function (5) with respect to G and making it equal to 0, one can obtain:
further, step S4 further includes:
s44, converting the (p) into (p) by applying a matrix inversion theorem B +q B )×(p B +q B ) The inversion of the dimensional matrix, the simplification of the available migration gain matrix, can be achieved by:
wherein the content of the first and second substances,the Fisher information matrix for single-ton energy consumption model parameter identification of the rectification process B and the transfer gain matrix G depend on the variance of the difference of the model parameter matrixes of the rectification process A and the rectification process BAnd the quality F of the collected observation data of the rectification process B.
And S5, substituting the solved migration gain matrix G into the parameter migration identification model to obtain a single-ton energy consumption predicted value of the rectification process B.
The step S5 comprises the following steps:
substituting the calculated migration gain matrix (7) into the parameter migration identification strategy (3) to obtain:
according to the single-ton energy consumption model (2) in the rectification process, a single-ton energy consumption predicted value of the rectification process B can be obtained as follows:
optionally, the present invention further comprises the steps of:
s6, determining conditions for improving the migration identification precision:
selecting Mean Square Error (MSE) as an evaluation index of the identification precision, and defining:
wherein the content of the first and second substances,the single-ton energy consumption model parameter identification result of the rectification process B under the condition of no parameter of the migration rectification process A is obtained, MSE _ ls is MSE of the single-ton energy consumption model parameter identification result, and MSE _ tr is MSE of the parameter migration identification result.
Substituting (3) and (7) into (8), calculating to obtain:
it can be seen that whether the accuracy of the parameter migration identification is improved depends on the variance of the model parameter differences of rectification processes A and BAnd the quality F of the collected observation data of the rectification process B. When the observation data of the rectification process A and the rectification process B meet the conditions:
when the MSE _ tr is less than or equal to the MSE _ ls, the identification precision of the parameter migration is higher than that of the condition without the migration parameter. Optionally, the indicators of the identification performance analysis include mean square error, root mean square error, mean absolute percentage error, logarithm of mean square error, median absolute error, and the like.
In the process industry, the identification of existing energy consumption prediction models generally only utilizes knowledge of the current process to be identified, and is not linked to product quality. The method for predicting the energy consumption of the single ton in the rectification process based on the migration identification considers the energy consumption required by producing qualified products of the single ton, and simultaneously, the migration identification technology is adopted to improve the model precision of the process to be identified by means of the model knowledge of the identified process, reduce the identification cost, and evaluate the established model for predicting the energy consumption of the single ton.
To demonstrate the effectiveness of the present invention, in one embodiment, the historical data of rectification process a was preprocessed using the method of the present invention, and fig. 2 shows the data of feed flow, reflux amount, and cooling water flow after preprocessing. Determining the order (p) of the single-ton energy consumption prediction model by adopting a traditional identification method A ,q A ) And parametersThe predicted single ton energy consumption results for rectification process a are given in fig. 3.
Historical data of the rectification process B is further collected and preprocessed, and data of feed flow, reflux quantity and cooling water flow after preprocessing are shown in figure 4. Combining identified model parametersBy designing a migration gain matrix G, a migration identification parameter estimation is constructed
Selecting a migration identification criterion function J (G), solving a migration gain matrix G by taking the first derivative of J (G) and making the first derivative equal to 0, and further calculating a simplified and realizable migration gain matrixSubstituting it into the migration identification parameter estimationAnd obtaining a single-ton energy consumption prediction model based on migration identification in the rectification process B.
And finally, selecting Mean Square Error (MSE) of the parameter estimation value as an evaluation index of the identification precision, and discussing the relationship between the performance of the migration identification method and the performance of the traditional single-task identification method. Fig. 5 shows the prediction result of the single-ton energy consumption prediction model in the rectification process B, and it can be seen from the figure that the single-ton energy consumption model established by the migration identification method can accurately predict the single-ton energy consumption.
The preferred embodiment of the present invention also discloses an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the steps of the method in the above embodiment.
The preferred embodiment of the present invention also discloses a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method described in the above embodiments.
The preferred embodiment of the invention also discloses a system for predicting the energy consumption of single ton in the rectification process based on migration identification, which comprises the following steps:
the single-ton energy consumption model establishing module is used for dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at the sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector;
a single-ton energy consumption model order and parameter vector determination module of the rectification process A, which is used for collecting the historical data of the rectification process A and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process A
And the parameter migration identification model construction module of the single-ton energy consumption model of the rectification process B is used for defining the single-ton energy consumption model parameter vector of the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the vector Y is output B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix;
the migration gain matrix solving module is used for selecting a migration identification criterion function and minimizing the migration identification criterion function so as to solve the migration gain matrix G;
and the single-ton energy consumption prediction module of the rectification process B is used for substituting the solved migration gain matrix G into the parameter migration identification model to obtain a single-ton energy consumption prediction value of the rectification process B.
The distillation process single-ton energy consumption prediction system based on the migration identification in the embodiment of the present invention is used for implementing the aforementioned distillation process single-ton energy consumption prediction method based on the migration identification, and therefore, the specific implementation of the system can be seen in the foregoing embodiment section of the distillation process single-ton energy consumption prediction method based on the migration identification, and therefore, the specific implementation thereof can refer to the description of the corresponding respective section embodiments, and no description is provided here.
In addition, since the single-ton energy consumption prediction system based on the migration identification in this embodiment is used for implementing the single-ton energy consumption prediction method based on the migration identification in the rectification process, the function of the single-ton energy consumption prediction system based on the migration identification corresponds to the function of the method, and details are not repeated here.
The above embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (10)
1. A single-ton energy consumption prediction method in a rectification process based on migration identification is characterized by comprising the following steps:
s1, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector;
s2, collecting historical data of the rectification process A, and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process A
S3, defining a single-ton energy consumption model parameter vector of the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the fluid is deliveredOutput vector Y B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix;
s4, selecting a migration identification criterion function, and minimizing the migration identification criterion function to solve a migration gain matrix G;
and S5, substituting the solved migration gain matrix G into the parameter migration identification model to obtain the single-ton energy consumption predicted value of the rectification process B.
2. The method for predicting the energy consumption per ton in the rectification process based on the migration identification as claimed in claim 1, wherein the step S1 comprises the following steps:
s11, dividing the stable working condition of the rectification process into T sampling intervals, and establishing a single-ton energy consumption model of the rectification process at a sampling moment k according to a material and energy balance equation:
y(k)+a 1 y(k-1)+…+a p y(k-p)=b 0 u(k)+b 1 u(k-1)+…+b q u(k-q)+v(k) (1)
wherein y (k) is an output variable at the time of k; u (k) is an input variable at the moment k; a is 1 ,a 2 ,…,a p ,b 0 ,b 1 ,…,b q Is a single ton energy consumption model parameter; p and q are the order of the single-ton energy consumption model; v (k) is measurement noise;
s12, defining output vectors, observation matrixes, parameter vectors and noise vectors as follows:
Y=[y(p+1),y(p+2)…,y(p+N)] T
θ=[a 1 ,…,a p ,b 0 ,…,b q ] T
V=[v(p+1),v(p+2)…,v(p+N)] T
wherein N is the length of the acquired data;
according to the definition, the single-ton energy consumption model (1) of the rectification process is equivalent to:
Y=Xθ+V
namely, equation (2).
3. The method for predicting the energy consumption per ton in the rectification process based on the migration identification as claimed in claim 1, wherein the step S4 comprises the following steps:
s41, selecting a migration identification criterion function:
wherein, theta B The actual value of the single-ton energy consumption model parameter matrix in the rectification process B is obtained; e [ ·]Calculating the mean value; trace {. Is trace operation of a matrix; j (-) is a criterion function for G;
and S42, substituting the formula (3) into the formula (4) to obtain:
J(G)=trace{E[(I-GX B )R(I-GX B ) T +GΣ B G T ]} (5)
wherein the content of the first and second substances,is the variance of the difference of the single-ton energy consumption model parameter matrixes in the rectification process A and the rectification process B,the difference of the single-ton energy consumption model parameter matrix of the rectification process A and the rectification process B, wherein I is an identity matrix, sigma B Is the noise covariance matrix of rectification process B.
Taking the first derivative of the criterion function (5) with respect to G and making it equal to 0, one can obtain:
4. the distillation process single-ton energy consumption prediction method based on migration identification as claimed in claim 3, wherein the step S4 further comprises:
s44, converting the (p) into (p) by applying a matrix inversion theorem B +q B )×(p B +q B ) Inversion of the dimensional matrix, a simplification of the available migration gain matrix can be achieved by:
wherein, the first and the second end of the pipe are connected with each other,the Fisher information matrix for single-ton energy consumption model parameter identification of the rectification process B and the transfer gain matrix G depend on the variance of the difference of the model parameter matrixes of the rectification process A and the rectification process BAnd the quality F of the collected observation data of the rectification process B.
5. The method for predicting energy consumption of single ton in rectification process based on migration identification as claimed in claim 1, wherein the output variables of the model for energy consumption of single ton in rectification process are energy consumption required for producing single ton of qualified product, and the input variables are feed flow, reflux amount and cooling water flow required for producing single ton of qualified product.
6. The method of claim 1, wherein the measurement noise comprises white noise subject to a gaussian distribution, a t distribution, and a poisson distribution.
7. The method for predicting single-ton energy consumption of rectification process based on migration recognition according to claim 1, wherein the historical data comprises four variables of feed flow, reflux flow, cooling water flow and single-ton energy consumption.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1-7 are implemented when the program is executed by the processor.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
10. A single-ton energy consumption prediction system for a rectification process based on migration identification is characterized by comprising the following steps:
the single-ton energy consumption model building module is used for dividing the stable working condition of the rectification process into T sampling intervals, and building a single-ton energy consumption model of the rectification process at the sampling moment k according to a material and energy balance equation:
Y=Xθ+V (2)
wherein Y is an output vector, X is an observation matrix, theta is a parameter vector, and V is a noise vector;
a single-ton energy consumption model order and parameter vector determination module for the rectification process A, which is used for collecting the rectified energyHistorical data of the process A, and determining the single-ton energy consumption model order (p) of the rectification process A A ,q A ) And single ton energy consumption model parameter vector of rectification process A
And the parameter migration identification model construction module of the single-ton energy consumption model of the rectification process B is used for defining the single-ton energy consumption model parameter vector of the rectification process B:
wherein p is B =p A ,q B =q A ;
Collecting and preprocessing historical data of a rectification process B to be identified, and combining the determined single-ton energy consumption model parameters of the rectification process AConstructing a parameter migration identification model of the single-ton energy consumption model of the rectification process B:
wherein the vector Y is output B For a single ton of energy consumption of the rectification process B, the matrix X is observed B A matrix formed by historical data in the rectification process B, and G is a migration gain matrix;
the migration gain matrix solving module is used for selecting a migration identification criterion function and minimizing the migration identification criterion function so as to solve the migration gain matrix G;
and the single-ton energy consumption prediction module of the rectification process B is used for substituting the solved migration gain matrix G into the parameter migration identification model to obtain a single-ton energy consumption prediction value of the rectification process B.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210878367.8A CN115169721B (en) | 2022-07-25 | 2022-07-25 | Method and system for predicting single ton energy consumption in rectification process based on migration identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210878367.8A CN115169721B (en) | 2022-07-25 | 2022-07-25 | Method and system for predicting single ton energy consumption in rectification process based on migration identification |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115169721A true CN115169721A (en) | 2022-10-11 |
CN115169721B CN115169721B (en) | 2023-07-04 |
Family
ID=83497267
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210878367.8A Active CN115169721B (en) | 2022-07-25 | 2022-07-25 | Method and system for predicting single ton energy consumption in rectification process based on migration identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115169721B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117010576A (en) * | 2023-10-07 | 2023-11-07 | 聊城莱柯智能机器人有限公司 | Energy consumption prediction method based on elastic dynamic neural network |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110322113A (en) * | 2019-05-22 | 2019-10-11 | 中国电力科学研究院有限公司 | Based on markovian Line Loss of Distribution Network System processing method |
CN112613542A (en) * | 2020-12-14 | 2021-04-06 | 国网甘肃省电力公司营销服务中心 | Bidirectional LSTM-based enterprise decontamination equipment load identification method |
WO2021208079A1 (en) * | 2020-04-17 | 2021-10-21 | 中国科学院深圳先进技术研究院 | Method and apparatus for obtaining power battery life data, computer device, and medium |
CN113761692A (en) * | 2021-10-11 | 2021-12-07 | 中国长江三峡集团有限公司 | Migration component analysis-based multi-wind turbine generator set operation state identification method |
CN113962081A (en) * | 2021-10-20 | 2022-01-21 | 江南大学 | Rectifying tower single-ton energy consumption estimation method and system based on auxiliary measurement information |
CN114093433A (en) * | 2021-11-29 | 2022-02-25 | 江南大学 | Observer-based method and system for evaluating prediction precision of single-ton energy consumption in rectification process |
-
2022
- 2022-07-25 CN CN202210878367.8A patent/CN115169721B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110322113A (en) * | 2019-05-22 | 2019-10-11 | 中国电力科学研究院有限公司 | Based on markovian Line Loss of Distribution Network System processing method |
WO2021208079A1 (en) * | 2020-04-17 | 2021-10-21 | 中国科学院深圳先进技术研究院 | Method and apparatus for obtaining power battery life data, computer device, and medium |
CN112613542A (en) * | 2020-12-14 | 2021-04-06 | 国网甘肃省电力公司营销服务中心 | Bidirectional LSTM-based enterprise decontamination equipment load identification method |
CN113761692A (en) * | 2021-10-11 | 2021-12-07 | 中国长江三峡集团有限公司 | Migration component analysis-based multi-wind turbine generator set operation state identification method |
CN113962081A (en) * | 2021-10-20 | 2022-01-21 | 江南大学 | Rectifying tower single-ton energy consumption estimation method and system based on auxiliary measurement information |
CN114093433A (en) * | 2021-11-29 | 2022-02-25 | 江南大学 | Observer-based method and system for evaluating prediction precision of single-ton energy consumption in rectification process |
Non-Patent Citations (1)
Title |
---|
平晓静等: "基于迁移学习的有机硅单体分馏过程能耗建模" * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117010576A (en) * | 2023-10-07 | 2023-11-07 | 聊城莱柯智能机器人有限公司 | Energy consumption prediction method based on elastic dynamic neural network |
CN117010576B (en) * | 2023-10-07 | 2024-02-02 | 聊城莱柯智能机器人有限公司 | Energy consumption prediction method based on elastic dynamic neural network |
Also Published As
Publication number | Publication date |
---|---|
CN115169721B (en) | 2023-07-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109060001B (en) | Multi-working-condition process soft measurement modeling method based on feature transfer learning | |
CN109992921B (en) | On-line soft measurement method and system for thermal efficiency of boiler of coal-fired power plant | |
CN109389314B (en) | Quality soft measurement and monitoring method based on optimal neighbor component analysis | |
CN107403196B (en) | Method for predicting concentration of butane at bottom of debutanizer by instant learning modeling based on spectral clustering analysis | |
CN112947300A (en) | Virtual measuring method, system, medium and equipment for processing quality | |
CN112116198A (en) | Data-driven process industrial state perception network key node screening method | |
CN111626359A (en) | Data fusion method and device, control terminal and ship | |
CN115169721B (en) | Method and system for predicting single ton energy consumption in rectification process based on migration identification | |
CN110046377A (en) | A kind of selective ensemble instant learning soft-measuring modeling method based on isomery similarity | |
CN108984851B (en) | Weighted Gaussian model soft measurement modeling method with time delay estimation | |
CN114879628A (en) | Multi-mode industrial process fault diagnosis method based on local maximum mean difference resistance | |
CN113130014B (en) | Rare earth extraction simulation method and system based on multi-branch neural network | |
CN106599391B (en) | Association vector machine soft measurement modeling method based on dynamic weighting of triangle angle values | |
CN113203953A (en) | Lithium battery residual service life prediction method based on improved extreme learning machine | |
CN116484275B (en) | Aluminum electrolysis anode effect monitoring method based on self-adaptive space-time diagram convolutional neural network | |
CN111914471A (en) | Rectification tower fault detection method based on rapid nuclear independent component analysis | |
CN112561119A (en) | Cloud server resource performance prediction method using ARIMA-RNN combined model | |
CN114707424B (en) | Chemical process soft measurement method based on quality-related slow characteristic analysis algorithm | |
CN116470491A (en) | Photovoltaic power probability prediction method and system based on copula function | |
CN111861002A (en) | Building cold and hot load prediction method based on data-driven Gaussian learning technology | |
CN109598283B (en) | Aluminum electrolysis superheat degree identification method based on semi-supervised extreme learning machine | |
CN113151842B (en) | Method and device for determining conversion efficiency of wind-solar complementary water electrolysis hydrogen production | |
CN114861759A (en) | Distributed training method of linear dynamic system model | |
CN110288724B (en) | Batch process monitoring method based on wavelet function principal component analysis | |
CN111291020A (en) | Dynamic process soft measurement modeling method based on local weighted linear dynamic system |
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 |