CN110009139A - A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling - Google Patents
A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling Download PDFInfo
- Publication number
- CN110009139A CN110009139A CN201910203634.XA CN201910203634A CN110009139A CN 110009139 A CN110009139 A CN 110009139A CN 201910203634 A CN201910203634 A CN 201910203634A CN 110009139 A CN110009139 A CN 110009139A
- Authority
- CN
- China
- Prior art keywords
- column plate
- jth
- block column
- air separation
- jth block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- 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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling, including mechanism statistics hybrid modeling module, energy consumption calculation module, energy saving intelligent optimization computing module, for carrying out operating energy saving intelligent optimization calculating to thermal coupling air separation plant, and optimal operational parameters are passed into control station and storage system by fieldbus.The present invention is in order to overcome existing thermal coupling space division model on-line operation low efficiency, the deficiency of model accuracy difference, by the mechanism model for combining high expansion, and it can use the statistical model of mass production historical data, that establishes thermal coupling air separation plant has both high-precision and efficient energy saving mixed model, the thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling is realized on this basis, it can effectively reflect the complex nonlinear physical property characteristic of thermal coupling air separation process, on-line operation is high-efficient, energy saving optimizing effect is more preferable than the optimization method based on traditional mechanism model or statistical model.
Description
Technical field
The present invention relates to the energy saving intelligent optimization field of air separation plant more particularly to a kind of thermal couplings based on hybrid modeling
Air separation plant energy conservation intelligent optimization system.
Background technique
Air separation plant is a kind of process units that industrial processes are widely used.Air separation plant separates air,
And high-purity industrial gasses such as obtain oxygen, nitrogen, argon, it is widely used in petroleum, chemical industry, the energy, electronics, metallurgy, aerospace, doctor
The various industrial circles such as health care, food and drink are treated, are had a wide range of applications in national economy and important meaning.Air separation plant
It is one of maximum equipment of industrial processes energy consumption, but is a kind of typical cryogenic rectification process, thermodynamic efficiency is simultaneously
Not high, theoretical thermodynamics efficiency is typically only 15%-20% or so, and energy cost usually accounts for the 75% of air products price,
Therefore air separation plant is also one of energy waste energy consumption equipment the most serious, is that typical energy of sacrificing exchanges setting for purity for
It is standby, there is great efficiency room for promotion.In recent years, on the one hand, due to the development of modern industry, some large scale industry projects are such as
Steel and iron industry, chemical industry, oil exploitation etc. require to provide air product by large air separation plant, and air separation plant demand is got over
Come bigger;On the other hand, since " oil crisis " twice of 1970s, energy crisis is deepened, in addition increasingly high
The ecological requirements and Air separation industry that rise benefit requirement, all to air separation process it is energy-saving propose it is more stringent
It is required that.Therefore, it improves the energy efficiency of air separation technology, reduce the energy consumption of air separation process, it is very urgent.
Currently, its model foundation is to a core of air separation plant energy saving research in the world, but space division is set at present
Standby mechanism model is sufficiently complex, so that traditional mechanism model although precision of prediction with higher, but on-line operation is imitated
Rate is too low, is unfavorable for subsequent energy saving optimizing research;On the other hand, the linear statistical mould based on data to grow up in the recent period
Type cannot effectively describe the complex nonlinear characteristic of air separation process, often have lower precision of prediction, cause based on data mould
The optimization method of type is very limited to the improvement of air separation process energy-saving effect.It turns out that establishing the high efficiency of air separation plant, height
Accuracy model is the premise for improving the production Control platform of the process, has become a crucial air separation energy saving technology.Space division
Equipment mixed model can bonding mechanism Model Extension be good, feature with high accuracy and statistical model can use a large amount of skies
It is mitogenetic to produce the high advantage of historical data, operational efficiency, it is that one kind has both high-precision and efficient novel air sub-model.In this base
On plinth, the production decision of low energy consumption can be easily obtained, significantly improves production run efficiency, reduces production energy consumption.
Summary of the invention
In order to overcome in existing air separation plant energy saving optimizing system, air separation plant modeling process cannot effectively reflect space division mistake
The low deficiency of the complex nonlinear physical property characteristic of journey, on-line operation inefficiency, accuracy, the purpose of the present invention is to provide one
Air separation plant energy conservation intelligent optimization system of the kind based on hybrid modeling, can effectively reflect the complex nonlinear physical property of air separation process
Feature, on-line operation is high-efficient, energy saving optimizing effect is more preferable.
The purpose of the present invention is achieved through the following technical solutions: a kind of air separation plant energy conservation based on hybrid modeling
Intelligent optimization system, including mechanism statistics hybrid modeling module, energy consumption calculation module, energy saving intelligent optimization computing module.
The mechanism statistics hybrid modeling module is with the temperature data of the temperature detecting element acquisition in intelligence instrument, pressure
The inspections such as the pressure data of detecting element acquisition, the data on flows of flow detecting element acquisition, the composition data of analysis element acquisition
Measured data, the data as mechanism Statistical mixture model input, and the method combined by modelling by mechanism with statistical modeling calculates
The parameters such as temperature, pressure, flow, composition distribution are believed in the air separation plant under current air separation plant service condition and operating condition out
Breath.
(1) mechanism model is realized by following formula:
Wherein, subscript i is one of components such as nitrogen, argon, oxygen, and j is the column plate serial number numbered from top to bottom, yi,jFor jth
The gas phase composition of i component, x on block column platei,jFor the liquid phase composition of i component on jth block column plate.
yi,j=ki,jxi,j (3)
Wherein, ki,jFor the vapor liquid equilibrium coefficient of i component on jth block column plate.
Lj-1xi,j-1-(Vj+Gj)yI, j-(Lj+Sj)xi,j+Vj+1yi,j+1=-Fjzi,j (4)
Wherein, Lj-1For the liquid phase flow on -1 block of column plate of jth, LjFor the liquid phase flow on jth block column plate, VjFor jth block
Gas phase flow rate on column plate, Vj+1For the gas phase flow rate on+1 block of column plate of jth, GjFlow, S are produced for the gas phase on jth block column platej
Flow, F are produced for the liquid phase on jth block column platejFor the feed rate on jth block column plate, xi,j-1For i group on -1 block of column plate of jth
The liquid phase composition divided, yi,j+1For the gas phase composition of i component on+1 block of column plate of jth, zi,jFor the charging group of i component on jth block column plate
At.
Wherein,For the liquid phase enthalpy on -1 block of column plate of jth,For the liquid phase enthalpy on jth block column plate,It is
Gas phase enthalpy on j block column plate,For the gas phase enthalpy on+1 block of column plate of jth,For the charging enthalpy on jth block column plate.
Interior coupled and heat-exchange rate U on jth block column platejValue are as follows:
Uj=αjΔTj (6)
Wherein, αjFor the thermal coupling heat exchange efficiency on jth block column plate, Δ TjIt exchanges heat for the thermal coupling on jth block column plate warm
Difference.
ki,jFrom being calculated by Peng-Robinson state equation, final calculation formula is as follows:
Wherein,For the Phase Fugacity coefficient of component i on jth block column plate,For the liquid phase of component i on jth block column plate
Fugacity coefficient.It can be calculated by following formula:
Wherein, bi,jFor the Van der waals volumes of i component on jth block column plate, ai,jFor the gravitation ginseng of i component on jth block column plate
Number, ajFor the weighted sum of all components intermolecular attraction parameter on jth block column plate, bjFor all components Van der Waals on jth block column plate
The weighted sum of volume, ZjFor the compressibility factor on jth block column plate, Aj、BjIt is fixed when to calculate jth block column plate liquid phase fugacity coefficient
The intermediate variable of justice is calculated as follows by the mixing rule of mixture:
Wherein, subscript i1And i2For two kinds of components in oxygen, nitrogen, argon,For component i on jth block column plate1Liquid phase group
At,For component i on jth block column plate2Liquid phase composition,For component i on jth block column plate1With component i2Between gravitation ginseng
Number, PjFor the pressure on jth block column plate, TjFor the temperature on jth block column plate, vjFor the molal volume on jth block column plate, R is gas
Body constant.
The Phase Fugacity coefficient of component i on jth block column plateCalculation method and liquid phase fugacity coefficientIt is identical, only need by
Liquid phase composition x in formula (9)-(13) replaces with corresponding gas phase composition y.
(2) statistical model carries out statistical modeling by the historical sample data to air separation plant, to the object in mechanism model
Property accounting equation is modified, and is realized by following formula:
yi,j=ki,jxi,j (14)
It is stated that canonical form first:
Wherein, f (X) is required regression equation form, and X is input data vector, and ω is Linear Mapping vector, subscript T
The transposition of representing matrix,It is the Nonlinear Mapping that X is projected to high-dimensional feature space, β is amount of bias.Give M thermal coupling
Close the historical sample data of air separation plant operationWherein XlFor the input vector of first of model sample, YlIt is first
The output vector of sample.Define the insensitive loss function gamma of first of sample εε(Yl,f(Xl)) are as follows:
Wherein, ε > 0 is the directly related design parameter of and function estimated accuracy, and solving purpose is construction f (X), is made and mesh
The distance between scale value is less than ε, and the parameter in above formula solves as follows:
Wherein, R () is objective function, ξl,ξl *For one group of relaxation factor of first of model sample, γ is penalty coefficient.
The energy consumption calculation module counts temperature, the pressure, flow, composition that hybrid modeling module is calculated according to mechanism
The parameter informations such as distribution, calculate the economic and technical norms such as the energy consumption of air separation process:
E=∑ Qcomp+∑Qhex (18)
E is thermal coupling air separation plant comprehensive energy consumption, and Q is the energy consumption of each equipment, the energy consumption Q including pumpcompWith change
The energy consumption Q of hot devicehexTwo classes.
The energy consumption Q of pumpcompCalculation formula are as follows:
Wherein, τ is polytropic proces coefficient, and V is flow, and R is gas constant, TinFor inlet temperature, PinFor inlet pressure,
PoutFor outlet pressure, ηcompFor the efficiency of pump.
The energy consumption Q of heat exchangerhexCalculation formula are as follows:
Qhex=V | Hin-Hout|ηhex -1 (20)
Wherein, V is flow, HinFor import enthalpy, HoutTo export enthalpy, ηhexFor efficiency of heat exchanger.
The energy conservation intelligent optimization computing module counts temperature, the pressure, stream that hybrid modeling module provides according to mechanism
The parameter informations such as amount, composition distribution, and according to economic and technical norms such as the calculated energy consumptions of energy consumption calculation module, use energy conservation
Intelligent algorithm optimizes calculating, to obtain the operating parameter optimal about energy consumption and corresponding economic indicator:
Wherein, XopFor operating condition vector, E (Xop) it is the energy consumption that energy consumption calculation module obtains, C (Xop) indicate mechanism system
The equality constraint that meter hybrid modeling module obtains, XLFor operating condition lower bound, XUFor the operating condition upper bound.The Optimized model according to
Following intelligent algorithm solves:
(1) the value X of N number of operating condition vector particle is randomly generated in D dimension problem solution spacem=(Xm1,Xm2,...,XmD)
With speed Δ Xm=(Δ Xm1,ΔXm2,...,ΔXmD), m=1,2 ..., N, wherein m indicates m-th of particle.Particle is set
Practise rate c1=c2=2, inertia weight maxima and minima w is setmax=0.8, wmin=0.2.Population quantity N=20 is set,
Maximum number of iterations iter is setmax=100.Meanwhile primary iteration number t=1 is set;
(2) speed of more new particle and position according to the following formula:
ΔXm(t+1)=w Δ Xm(t)+c1r1[pbest-Xm(t)]+c2r2[gbest-Xm(t)] (22)
Xm(t+1)=Xm(T)+ΔXm(t+1) (23)
W (t+1)=wmax-(wmax-wmin)×(t-1)/itermax (24)
Wherein, Δ XmIt (t) is speed of the particle m in the t times iteration, Δ XmIt (t) is position of the particle m in the t times iteration
It sets, Δ XmIt (t+1) is speed of the particle m in the t+1 times iteration, Δ XmIt (t+1) is position of the particle m in the t+1 times iteration
It sets, pbestIt is the locally optimal solution of particle experience;gbestIt is the globally optimal solution of all particle experiences, w is inertia weight,
c1And c2It is learning rate, r1And r2It is the random number between 0 to 1, inertia weight when w (t+1) is the t+1 times iteration,
wmaxAnd wminIt is the maximum value and minimum value of inertia weight, itermaxIt is maximum number of iterations.
(3) p is updatedbest: compare the fitness value and its individual optimal solution p of some particlebestIf fitness value is better than
pbest, then the position of using the particle current is as pbest, wherein fitness value is energy consumption E.
(4) g is updatedbest: the fitness value of more all particles and the globally optimal solution g of populationbest, select adaptive optimal control
The position of the particle of angle value is as gbest;
(5) termination condition judges: judging whether the number of iterations reaches setting value or precision whether less than 0.001, if reaching
It arrives, iteration ends, if not reaching, turns to step (2) and continue iteration.
Technical concept of the invention are as follows: by combination high expansion, high-precision mechanism model, and can use a large amount of
Production history data, efficient statistical model, it is established that air separation plant has both high-precision and efficient energy saving hybrid guided mode
Type realizes the complex nonlinear mechanism characteristic and actual motion state for accurately holding air separation plant, to overcome existing space division
The deficiency of model on-line operation low efficiency, model accuracy difference, while being saved with Calculation Method of Energy Consumption, energy saving intelligence computation method
It can optimize, the optimal operational parameters of air separation plant are calculated, finally obtain high efficiency, high-precision energy saving optimizing system.
Beneficial effects of the present invention are mainly manifested in: 1, being transported the mechanism model based on air-separating energy-saving process with based on space division
The statistical model of row real data combines, and establishes the mechanism Statistical mixture model of air separation process, can expeditiously, in high precision
The complex nonlinear mechanism characteristic and actual motion state of air separation process are taken into account in ground, have higher modeling accuracy;2, it is mixing
It carries out air separation plant energy conservation intelligent optimization on the basis of model to calculate, on-line operation is high-efficient, and energy saving optimizing effect is than traditional
Optimization method based on simple mechanism model or statistical model is more preferable.
Detailed description of the invention
Fig. 1 is the schematic diagram of the air separation plant energy conservation intelligent optimization system proposed by the invention based on hybrid modeling;
Fig. 2 is the schematic diagram of the energy saving intelligent optimization system based on hybrid modeling.
Specific embodiment
The present invention is illustrated below according to attached drawing.
Referring to Fig.1, space division production process includes air separation plant 1, intelligence instrument 2, data-interface 3, controller 4, control station
5, fieldbus 6, storage device 7 and the energy saving intelligent optimization system 8 based on hybrid modeling, wherein intelligence instrument 2, controller 4
Directly it is connected with air separation plant 1, control station 5, storage device 7 and the energy saving intelligent optimization system 8 based on hybrid modeling pass through
Fieldbus 6, the realization of data-interface 3 are connected with intelligence instrument 2, controller 4.The intelligence instrument 2 passes through temperature detection member
Part, pressure detecting element, flow detecting element measure relevant parameter, and connect with data-interface 3.It is described based on hybrid modeling
Energy saving intelligent optimization system 8 realizes the solution function of the optimal operational parameters of air separation plant 1, and optimal operational parameters are passed through now
Field bus passes to control station 5 and storage device 7.
It is analyzed according to reaction mechanism and flow process, it is contemplated that product and energy consumption are had an impact in space division production process
Each factor a tower in the Shang Ta, lower tower or argon column of air separation plant is taken in actual production process and is commonly operated
Variable and easily survey variable are as mode input variable, comprising: tower pressure interior force, feeding temperature, flow, composition, tower top tower bottom and side
The temperature of line extraction stream stock, flow, composition, operating reflux ratio boil ratio again.
Referring to Fig. 2, a kind of air separation plant energy conservation intelligent optimization system based on hybrid modeling, including mechanism statistics mixing are built
Mould module 9, energy consumption calculation module 10, energy saving intelligent optimization computing module 11.
The mechanism statistics hybrid modeling module 9 is with the temperature data of the temperature detecting element acquisition in intelligence instrument 2, pressure
The pressure data of power detecting element acquisition, the data on flows of flow detecting element acquisition, composition data of analysis element acquisition etc.
Detection data, the data as mechanism Statistical mixture model 9 input, the method combined by modelling by mechanism with statistical modeling,
Calculate the ginsengs such as temperature in the air separation plant under current air separation plant service condition and operating condition, pressure, flow, composition distribution
Number information.
(1) mechanism model is realized by following formula:
Wherein, subscript i is one of components such as nitrogen, argon, oxygen, and j is the column plate serial number numbered from top to bottom, yi,jFor jth
The gas phase composition of i component, x on block column platei,jFor the liquid phase composition of i component on jth block column plate.
yi,j=ki,jxi,j (3)
Wherein, ki,jFor the vapor liquid equilibrium coefficient of i component on jth block column plate.
Lj-1xi,j-1-(Vj+Gj)yI, j-(Lj+Sj)xi,j+Vj+1yi,j+1=-Fjzi,j (4)
Wherein, Lj-1For the liquid phase flow on -1 block of column plate of jth, LjFor the liquid phase flow on jth block column plate, VjFor jth block
Gas phase flow rate on column plate, Vj+1For the gas phase flow rate on+1 block of column plate of jth, GjFlow, S are produced for the gas phase on jth block column platej
Flow, F are produced for the liquid phase on jth block column platejFor the feed rate on jth block column plate, xi,j-1For i group on -1 block of column plate of jth
The liquid phase composition divided, yi,j+1For the gas phase composition of i component on+1 block of column plate of jth, zi,jFor the charging group of i component on jth block column plate
At.
Wherein,For the liquid phase enthalpy on -1 block of column plate of jth,For the liquid phase enthalpy on jth block column plate,It is
Gas phase enthalpy on j block column plate,For the gas phase enthalpy on+1 block of column plate of jth,For the charging enthalpy on jth block column plate.
Interior coupled and heat-exchange rate U on jth block column platejValue are as follows:
Uj=αjΔTj (6)
Wherein, αjFor the thermal coupling heat exchange efficiency on jth block column plate, Δ TjIt exchanges heat for the thermal coupling on jth block column plate warm
Difference.
ki,jFrom being calculated by Peng-Robinson state equation, final calculation formula is as follows:
Wherein,For the Phase Fugacity coefficient of component i on jth block column plate,For the liquid phase of component i on jth block column plate
Fugacity coefficient.It can be calculated by following formula:
Wherein, bi,jFor the Van der waals volumes of i component on jth block column plate, ai,jFor the gravitation ginseng of i component on jth block column plate
Number, ajFor the weighted sum of all components intermolecular attraction parameter on jth block column plate, bjFor all components Van der Waals on jth block column plate
The weighted sum of volume, ZjFor the compressibility factor on jth block column plate, Aj、BjIt is fixed when to calculate jth block column plate liquid phase fugacity coefficient
The intermediate variable of justice is calculated as follows by the mixing rule of mixture:
Wherein, subscript i1And i2For two kinds of components in oxygen, nitrogen, argon,For component i on jth block column plate1Liquid phase group
At,For component i on jth block column plate2Liquid phase composition,For component i on jth block column plate1With component i2Between gravitation ginseng
Number, PjFor the pressure on jth block column plate, TjFor the temperature on jth block column plate, vjFor the molal volume on jth block column plate, R is gas
Body constant.
The Phase Fugacity coefficient of component i on jth block column plateCalculation method and liquid phase fugacity coefficientIt is identical, only need by
Liquid phase composition x in formula (9)-(13) replaces with corresponding gas phase composition y.
(2) statistical model carries out statistical modeling by the historical sample data to air separation plant, to the object in mechanism model
Property accounting equation is modified, and is realized by following formula:
yi,j=ki,jxi,j (14)
It is stated that canonical form first:
Wherein, f (X) is required regression equation form, and X is input data vector, and ω is Linear Mapping vector, subscript T
The transposition of representing matrix,It is the Nonlinear Mapping that X is projected to high-dimensional feature space, β is amount of bias.Give M thermal coupling
Close the historical sample data of air separation plant operationWherein XlFor the input vector of first of model sample, YlIt is first
The output vector of sample.Define the insensitive loss function gamma of first of sample εε(Yl,f(Xl)) are as follows:
Wherein, ε > 0 is the directly related design parameter of and function estimated accuracy, and solving purpose is construction f (X), is made and mesh
The distance between scale value is less than ε, and the parameter in above formula solves as follows:
Wherein, R () is objective function, ξl,ξl *For one group of relaxation factor of first of model sample, γ is penalty coefficient.
The energy consumption calculation module 10 according to mechanism count hybrid modeling module 9 be calculated temperature, pressure, flow,
The parameter informations such as composition distribution, calculate the economic and technical norms such as the energy consumption of air separation process:
E=∑ Qcomp+∑Qhex (18)
E is thermal coupling air separation plant comprehensive energy consumption, and Q is the energy consumption of each equipment, the energy consumption Q including pumpcompWith change
The energy consumption Q of hot devicehexTwo classes.
The energy consumption Q of pumpcompCalculation formula are as follows:
Wherein, τ is polytropic proces coefficient, and V is flow, and R is gas constant, TinFor inlet temperature, PinFor inlet pressure,
PoutFor outlet pressure, ηcompFor the efficiency of pump.
The energy consumption Q of heat exchangerhexCalculation formula are as follows:
Qhex=V | Hin-Hout|ηhex -1 (20)
Wherein, V is flow, HinFor import enthalpy, HoutTo export enthalpy, ηhexFor efficiency of heat exchanger.
The energy conservation intelligent optimization computing module 11, according to mechanism count temperature, pressure that hybrid modeling module 9 provides,
The parameter informations such as flow, composition distribution, and according to economic and technical norms such as the calculated energy consumptions of energy consumption calculation module 10, use
Energy saving intelligent algorithm optimizes calculating, to obtain the operating parameter optimal about energy consumption and corresponding economic indicator:
Wherein, XopFor operating condition vector, E (Xop) it is the energy consumption that energy consumption calculation module 10 obtains, C (Xop) indicate mechanism
The equality constraint that statistics hybrid modeling module 9 obtains, XLFor operating condition lower bound, XUFor the operating condition upper bound.The Optimized model root
Intelligent algorithm is descended to solve accordingly:
(1) the value X of N number of operating condition vector particle is randomly generated in D dimension problem solution spacem=(Xm1,Xm2,...,XmD)
With speed Δ Xm=(Δ Xm1,ΔXm2,...,ΔXmD), m=1,2 ..., N, wherein m indicates m-th of particle.Particle is set
Practise rate c1=c2=2, inertia weight maxima and minima w is setmax=0.8, wmin=0.2.Population quantity N=20 is set,
Maximum number of iterations iter is setmax=100.Meanwhile primary iteration number t=1 is set;
(2) speed of more new particle and position according to the following formula:
ΔXm(t+1)=w Δ Xm(t)+c1r1[pbest-Xm(t)]+c2r2[gbest-Xm(t)] (22)
Xm(t+1)=Xm(T)+ΔXm(t+1) (23)
W (t+1)=wmax-(wmax-wmin)×(t-1)/itermax (24)
Wherein, Δ XmIt (t) is speed of the particle m in the t times iteration, Δ XmIt (t) is position of the particle m in the t times iteration
It sets, Δ XmIt (t+1) is speed of the particle m in the t+1 times iteration, Δ XmIt (t+1) is position of the particle m in the t+1 times iteration
It sets, pbestIt is the locally optimal solution of particle experience;gbestIt is the globally optimal solution of all particle experiences, w is inertia weight,
c1And c2It is learning rate, r1And r2It is the random number between 0 to 1, inertia weight when w (t+1) is the t+1 times iteration,
wmaxAnd wminIt is the maximum value and minimum value of inertia weight, itermaxIt is maximum number of iterations.
(3) p is updatedbest: compare the fitness value and its individual optimal solution p of some particlebestIf fitness value is better than
pbest, then the position of using the particle current is as pbest, wherein fitness value is energy consumption E.
(4) g is updatedbest: the fitness value of more all particles and the globally optimal solution g of populationbest, select adaptive optimal control
The position of the particle of angle value is as gbest;
(5) termination condition judges: judging whether the number of iterations reaches setting value or precision whether less than 0.001, if reaching
It arrives, iteration ends, if not reaching, turns to step (2) and continue iteration.
Above-described embodiment is used to illustrate the present invention, rather than limits the invention, in spirit of the invention and
In scope of protection of the claims, to any modifications and changes that the present invention makes, protection scope of the present invention is both fallen within.
Claims (3)
1. a kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling, it is characterised in that: unite including mechanism
Hybrid modeling module, energy consumption calculation module, energy saving intelligent optimization computing module are counted, the energy consumption calculation module, energy conservation intelligence are excellent
Change computing module to be connected with mechanism statistics hybrid modeling module.The mechanism statistics hybrid modeling module is in intelligence instrument
The flow that the temperature data of temperature detecting element acquisition, the pressure data of pressure detecting element acquisition, flow detecting element acquire
Detection datas, the data as mechanism Statistical mixture model such as data, the composition data of analysis element acquisition input, and pass through mechanism
The method combined with statistical modeling is modeled, the thermal coupling under current thermal coupling air separation plant service condition and operating condition is calculated
Close the parameter informations such as temperature in air separation plant, pressure, flow, composition distribution.
(2.1) mechanism model is realized by following formula:
Wherein, subscript i is one of components such as nitrogen, argon, oxygen, and j is the column plate serial number numbered from top to bottom, yi,jFor jth block tower
The gas phase composition of i component, x on platei,jFor the liquid phase composition of i component on jth block column plate.
yi,j=ki,jxi,j (3)
Wherein, ki,jFor the vapor liquid equilibrium coefficient of i component on jth block column plate.
Lj-1xi,j-1-(Vj+Gj)yI, j-(Lj+Sj)xi,j+Vj+1yi,j+1=-Fjzi,j (4)
Wherein, Lj-1For the liquid phase flow on -1 block of column plate of jth, LjFor the liquid phase flow on jth block column plate, VjFor jth block column plate
On gas phase flow rate, Vj+1For the gas phase flow rate on+1 block of column plate of jth, GjFlow, S are produced for the gas phase on jth block column platejIt is
Liquid phase on j block column plate produces flow, FjFor the feed rate on jth block column plate, xi,j-1For i component on -1 block of column plate of jth
Liquid phase composition, yi,j+1For the gas phase composition of i component on+1 block of column plate of jth, zi,jFor the feed composition of i component on jth block column plate.
Wherein,For the liquid phase enthalpy on -1 block of column plate of jth,For the liquid phase enthalpy on jth block column plate,For jth block
Gas phase enthalpy on column plate,For the gas phase enthalpy on+1 block of column plate of jth,For the charging enthalpy on jth block column plate.Jth
Interior coupled and heat-exchange rate U on block column platejValue are as follows:
Uj=αjΔTj (6)
Wherein, αjFor the thermal coupling heat exchange efficiency on jth block column plate, Δ TjFor the thermal coupling heat transfer temperature difference on jth block column plate.
ki,jFrom being calculated by Peng-Robinson state equation, final calculation formula is as follows:
Wherein,For the Phase Fugacity coefficient of component i on jth block column plate,For the liquid phase fugacity system of component i on jth block column plate
Number.It can be calculated by following formula:
Wherein, bi,jFor the Van der waals volumes of i component on jth block column plate, ai,jFor the gravitational parameter of i component on jth block column plate, aj
For the weighted sum of all components intermolecular attraction parameter on jth block column plate, bjFor all components Van der waals volumes on jth block column plate
Weighted sum, ZjFor the compressibility factor on jth block column plate, Aj、BjIt is defined when calculating jth block column plate liquid phase fugacity coefficient
Intermediate variable is calculated as follows by the mixing rule of mixture:
Wherein, subscript i1And i2For two kinds of components in oxygen, nitrogen, argon,For component i on jth block column plate1Liquid phase composition,
For component i on jth block column plate2Liquid phase composition,For component i on jth block column plate1With component i2Between gravitational parameter, PjFor
Pressure on jth block column plate, TjFor the temperature on jth block column plate, vjFor the molal volume on jth block column plate, R is gas constant.
The Phase Fugacity coefficient of component i on jth block column plateCalculation method and liquid phase fugacity coefficientIt is identical, it only need to be by formula
(9) the liquid phase composition x in-(13) replaces with corresponding gas phase composition y.
(2.2) statistical model carries out statistical modeling by the historical sample data to thermal coupling air separation plant, in mechanism model
Calculation of Physical Properties equation be modified, realized by following formula:
yi,j=ki,jxi,j (14)
It is stated that canonical form first:
Wherein, f (X) is required regression equation form, and X is input data vector, and ω is Linear Mapping vector, and subscript T is indicated
The transposition of matrix,It is the Nonlinear Mapping that X is projected to high-dimensional feature space, β is amount of bias.Given M thermal coupling is empty
The historical sample data of subset operationWherein XlFor the input vector of first of model sample, YlFor first of sample
Output vector.Define the insensitive loss function gamma of first of sample εε(Yl,f(Xl)) are as follows:
Wherein, ε > 0 is the directly related design parameter of and function estimated accuracy, and solving purpose is construction f (X), is made and target value
The distance between be less than ε, the parameter in above formula solves as follows:
Wherein, R () is objective function, ξl,ξl *For one group of relaxation factor of first of model sample, γ is penalty coefficient.
2. the thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling, feature exist according to claim 1
In: the energy consumption calculation module counts temperature, the pressure, flow, composition distribution that hybrid modeling module is calculated according to mechanism
Equal parameter informations, calculate the economic and technical norms such as the energy consumption of thermal coupling air separation process:
E=∑ Qcomp+∑Qhex (18)
E is thermal coupling air separation plant comprehensive energy consumption, and Q is the energy consumption of each equipment, the energy consumption Q including pumpcompWith heat exchanger
Energy consumption QhexTwo classes.
The energy consumption Q of pumpcompCalculation formula are as follows:
Wherein, τ is polytropic proces coefficient, and V is flow, and R is gas constant, TinFor inlet temperature, PinFor inlet pressure, PoutFor
Outlet pressure, ηcompFor the efficiency of pump.
The energy consumption Q of heat exchangerhexCalculation formula are as follows:
Qhex=V | Hin-Hout|ηhex -1 (20)
Wherein, V is flow, HinFor import enthalpy, HoutTo export enthalpy, ηhexFor efficiency of heat exchanger.
3. the thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling, feature exist according to claim 1
In: the energy conservation intelligent optimization computing module counts temperature, pressure, flow, the group of the offer of hybrid modeling module according to mechanism
At parameter informations such as distributions, and according to economic and technical norms such as the calculated energy consumptions of energy consumption calculation module, use energy saving intelligence
Optimization algorithm optimizes calculating, to obtain the operating parameter optimal about energy consumption and corresponding economic indicator:
Wherein, XopFor operating condition vector, E (Xop) it is the energy consumption that energy consumption calculation module obtains, C (Xop) indicate that mechanism statistics is mixed
Close the equality constraint that modeling module obtains, XLFor operating condition lower bound, XUFor the operating condition upper bound.The Optimized model is according to following
Intelligent algorithm solves:
(4.1) the value X of N number of operating condition vector particle is randomly generated in D dimension problem solution spacem=(Xm1,Xm2,...,XmD) and speed
Spend Δ Xm=(Δ Xm1,ΔXm2,...,ΔXmD), m=1,2 ..., N, wherein m indicates m-th of particle.Particle is set and learns speed
Rate c1=c2=2, inertia weight maxima and minima w is setmax=0.8, wmin=0.2.Population quantity N=20, setting are set
Maximum number of iterations itermax=100.Meanwhile primary iteration number t=1 is set;
(4.2) speed of more new particle and position according to the following formula:
ΔXm(t+1)=w Δ Xm(t)+c1r1[pbest-Xm(t)]+c2r2[gbest-Xm(t)] (22)
Xm(t+1)=Xm(T)+ΔXm(t+1) (23)
W (t+1)=wmax-(wmax-wmin)×(t-1)/itermax (24)
Wherein, Δ XmIt (t) is speed of the particle m in the t times iteration, Δ XmIt (t) is position of the particle m in the t times iteration,
ΔXmIt (t+1) is speed of the particle m in the t+1 times iteration, Δ XmIt (t+1) is position of the particle m in the t+1 times iteration,
pbestIt is the locally optimal solution of particle experience;gbestIt is the globally optimal solution of all particle experiences, w is inertia weight, c1With
c2It is learning rate, r1And r2It is the random number between 0 to 1, inertia weight when w (t+1) is the t+1 times iteration, wmaxWith
wminIt is the maximum value and minimum value of inertia weight, itermaxIt is maximum number of iterations.
(4.3) p is updatedbest: compare the fitness value and its individual optimal solution p of some particlebestIf fitness value is better than
pbest, then the position of using the particle current is as pbest, wherein fitness value is energy consumption E.
(4.4) g is updatedbest: the fitness value of more all particles and the globally optimal solution g of populationbest, select adaptive optimal control degree
The position of the particle of value is as gbest;
(4.5) termination condition judges: judge whether the number of iterations reach setting value or precision less than 0.001, if reaching,
Iteration ends turn to step (2) and continue iteration if not reaching.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910203634.XA CN110009139A (en) | 2019-03-18 | 2019-03-18 | A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910203634.XA CN110009139A (en) | 2019-03-18 | 2019-03-18 | A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110009139A true CN110009139A (en) | 2019-07-12 |
Family
ID=67167259
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910203634.XA Pending CN110009139A (en) | 2019-03-18 | 2019-03-18 | A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110009139A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110490386A (en) * | 2019-08-26 | 2019-11-22 | 苏州树森信息科技有限公司 | A kind of comprehensive energy dispatching method and comprehensive energy dispatch system |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101074841A (en) * | 2006-12-26 | 2007-11-21 | 浙江大学 | System and method for controlling air-separating tower dynamic matrix |
CN101760248A (en) * | 2008-12-19 | 2010-06-30 | 新奥科技发展有限公司 | Coal-based energy chemical product poly-generation system and method |
CN101884848A (en) * | 2010-06-30 | 2010-11-17 | 浙江大学 | Nonlinear observation system and method for temperature distribution in the air-separating energy-saving process |
CN203396202U (en) * | 2013-06-12 | 2014-01-15 | 开封迪尔空分实业有限公司 | Energy-saving nitrogen-cooled water pre-cooling system for air separation |
CN103558837A (en) * | 2013-10-10 | 2014-02-05 | 浙江大学 | Air separation device working condition fluctuation repairing method |
CN103793754A (en) * | 2013-12-13 | 2014-05-14 | 中冶南方工程技术有限公司 | Energy consumption prediction method of air separation system |
CN105066768A (en) * | 2015-09-09 | 2015-11-18 | 内蒙古包钢钢联股份有限公司 | Pre-cooling system energy saving method for air separation equipment |
US20180142950A1 (en) * | 2016-11-18 | 2018-05-24 | L'air Liquide, Societe Anonyme Pour L'etude Et L’Exploitation Des Procedes Georges Claude | Lng integration with cryogenic unit |
-
2019
- 2019-03-18 CN CN201910203634.XA patent/CN110009139A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101074841A (en) * | 2006-12-26 | 2007-11-21 | 浙江大学 | System and method for controlling air-separating tower dynamic matrix |
CN101760248A (en) * | 2008-12-19 | 2010-06-30 | 新奥科技发展有限公司 | Coal-based energy chemical product poly-generation system and method |
CN101884848A (en) * | 2010-06-30 | 2010-11-17 | 浙江大学 | Nonlinear observation system and method for temperature distribution in the air-separating energy-saving process |
CN203396202U (en) * | 2013-06-12 | 2014-01-15 | 开封迪尔空分实业有限公司 | Energy-saving nitrogen-cooled water pre-cooling system for air separation |
CN103558837A (en) * | 2013-10-10 | 2014-02-05 | 浙江大学 | Air separation device working condition fluctuation repairing method |
CN103793754A (en) * | 2013-12-13 | 2014-05-14 | 中冶南方工程技术有限公司 | Energy consumption prediction method of air separation system |
CN105066768A (en) * | 2015-09-09 | 2015-11-18 | 内蒙古包钢钢联股份有限公司 | Pre-cooling system energy saving method for air separation equipment |
US20180142950A1 (en) * | 2016-11-18 | 2018-05-24 | L'air Liquide, Societe Anonyme Pour L'etude Et L’Exploitation Des Procedes Georges Claude | Lng integration with cryogenic unit |
Non-Patent Citations (4)
Title |
---|
PEI-YI HAO等: ""New support vector algorithms with parametric insensitive/margin model"", 《NEURAL NETWORKS》 * |
ZHIYU WANG等: ""Characteristic Analysis and Optimal Design on Heat-Transfer Capacity for Energy Saving of Heat-Integrated Air Separation Columns"", 《INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH》 * |
邵晴: ""粒子群算法研究及其工程应用案例"", 《中国博士学位论文全文数据库 信息科技辑》 * |
闫正兵: ""内部热耦合空分塔的建模与优化研究"", 《中国博士学位论文全文数据库工程科技Ⅰ辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110490386A (en) * | 2019-08-26 | 2019-11-22 | 苏州树森信息科技有限公司 | A kind of comprehensive energy dispatching method and comprehensive energy dispatch system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN201129823Y (en) | Central air conditioner energy-saving control device based on artificial neural net technique | |
CN105886680B (en) | A kind of blast furnace ironmaking process molten iron silicon content dynamic soft measuring system and method | |
CN104651559B (en) | Blast furnace liquid iron quality online forecasting system and method based on multivariable online sequential extreme learning machine | |
CN105910169B (en) | District heating system regulating of heating net method and system based on mechanism model PREDICTIVE CONTROL | |
CN103558046B (en) | A kind of heat exchanger energy efficiency evaluation system | |
CN110906571A (en) | Solar heat pump hot water system control strategy optimization method based on machine learning | |
CN101526814B (en) | Leaching rate prediction and optimization operation method in wet metallurgical leaching process | |
CN201003930Y (en) | A cooling tower | |
Pattison et al. | Optimal design of air separation plants with variable electricity pricing | |
CN104458063B (en) | A kind of energy-saving heat quantity flow calibrating installation and method | |
CN201440088U (en) | Integrative flow rate calibration platform | |
CN107065515A (en) | Plate type heat exchanger model building method based on fuzzy-adaptation PID control | |
CN105425583A (en) | Control method of penicillin production process based on cooperative training local weighted partial least squares (LWPLS) | |
CN110009139A (en) | A kind of thermal coupling air separation plant energy conservation intelligent optimization system based on hybrid modeling | |
CN110008533A (en) | A kind of air separation plant energy-consumption monitoring system based on hybrid modeling | |
CN108446447A (en) | A kind of air cooling heat exchanger efficiency evaluation method | |
CN109934421A (en) | A kind of blast furnace molten iron silicon content prediction and compensation method towards the fluctuation working of a furnace | |
CN109948246A (en) | A kind of thermal coupling air separation plant energy saving optimizing system based on hybrid modeling | |
Bamimore et al. | Parametric effects on the performance of an industrial cooling tower | |
CN204730916U (en) | A kind of large-scale mobile steam heat and mass rate calibrating installation | |
CN107885080A (en) | A kind of internal thermally coupled air separation column control device based on concentration curve characteristic | |
CN109946993A (en) | A kind of thermal coupling air separation plant energy-consumption monitoring system based on hybrid modeling | |
CN107942660B (en) | For the internal thermally coupled air separation column control device of product design optimization of profile algorithm | |
CN109214137A (en) | A kind of prediction technique for the minimum mixing time in converter flow field | |
CN107918365B (en) | A kind of internal thermally coupled air separation column online observation device based on concentration curve characteristic |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190712 |
|
RJ01 | Rejection of invention patent application after publication |