CN115853793A - Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters - Google Patents

Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters Download PDF

Info

Publication number
CN115853793A
CN115853793A CN202211633396.4A CN202211633396A CN115853793A CN 115853793 A CN115853793 A CN 115853793A CN 202211633396 A CN202211633396 A CN 202211633396A CN 115853793 A CN115853793 A CN 115853793A
Authority
CN
China
Prior art keywords
circulating pump
outlet
slurry
mass transfer
flue gas
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
Application number
CN202211633396.4A
Other languages
Chinese (zh)
Inventor
郑成航
高翔
赵中阳
陈竹
张悠
周灿
张涌新
李钦武
吴卫红
翁卫国
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Jiaxing Research Institute of Zhejiang University
Original Assignee
Zhejiang University ZJU
Jiaxing Research Institute of Zhejiang University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU, Jiaxing Research Institute of Zhejiang University filed Critical Zhejiang University ZJU
Priority to CN202211633396.4A priority Critical patent/CN115853793A/en
Publication of CN115853793A publication Critical patent/CN115853793A/en
Pending legal-status Critical Current

Links

Images

Abstract

The invention relates to a circulating pump intelligent regulation and control method based on accurate prediction of desulfurization process parameters, which is based on SO (SO) combined with data drive of desulfurization process mechanism 2 And the concentration prediction model is used for controlling the circulating pump to operate optimally in real time by optimizing the configuration of the circulating pump under different working conditions and adopting an intelligent circulating pump regulation and control strategy. The invention is based on SO 2 Constructing an outlet SO by using a numerical model of an absorption mechanism and a model of the number of mass transfer units 2 A concentration prediction mechanism model is adopted, an intelligent optimization algorithm is adopted, and the historical operating data is utilized to identify and correct key parameters, SO that an outlet SO with mechanism and data fused is constructed 2 The concentration prediction model has higher precision and higher interpretability on the basis of realizing second-level calculation, and meets the requirement of optimized regulation and control; the invention adjusts the optimized target value based on the actual operation requirement of the circulating pump, and realizes the real-time optimized control of the circulating pumpAnd the cost of optimizing the operation of the circulating pump is reduced.

Description

Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters
Technical Field
The invention belongs to the technical field of slurry circulating pumps of desulfurization devices of ultralow-emission systems, and particularly relates to an intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters.
Background
The wet desulfurizing device is SO of coal-fired power plant, steel plant and the like 2 Discharge the common SO of enterprises 2 And (4) terminal treatment equipment. However, the wet desulfurization device has high energy consumption, which usually accounts for more than 40% of the energy consumption of the whole flue gas/tail gas end treatment system. Therefore, the reduction of the energy consumption of the wet desulphurization device has very important significance.
The slurry circulating pump is an important device of the wet desulphurization device, and directly influences the circulating slurry amount of the desulphurization system, thereby influencing the SO of the desulphurization device 2 And (4) removing efficiency. Meanwhile, the slurry circulating pump is also an important component of the energy consumption of the wet desulphurization device. If the number of the slurry circulating pumps is too large, the power consumption of the desulfurization system is increased; if the number of the slurry circulating pumps is insufficient, the desulfurization efficiency of the desulfurization device cannot be ensured, and the outlet SO is caused 2 The concentration exceeds the standard. Therefore, the slurry circulating pump of the desulfurization device is intelligently regulated and controlled, the potential of energy conservation and consumption reduction is exploited, and the amount of the slurry provided by the desulfurization device is adjusted in real time according to the working condition to reduce the ultralow emission systemThe total carbon emission amount and the improvement of the quality of the desulfurization byproduct gypsum are of great significance.
However, SO in the desulfurizer 2 The absorption mechanism is complicated. Since the slurry and the flue gas are usually in countercurrent contact in the absorption tower, the boundary conditions of the flue gas and the slurry are respectively positioned at the inlet and the outlet of the absorption tower, SO the traditional SO-based method is based on 2 Numerical modeling of the absorption mechanism generally requires the assumption of the outlet SO 2 The concentration of the flue gas is calculated to the flue gas inlet according to the boundary condition, and the inlet SO is compared and calculated 2 Concentration and given inlet SO 2 Concentration, and hence the assumed outlet SO, in an iterative manner 2 Correcting the flue gas concentration until calculating the inlet SO 2 Concentration and given inlet SO 2 The error between the concentrations is within an allowable range. Therefore, even if the accuracy of the numerical model is relatively high, the calculation time is too long, and the method is not suitable for real-time circulating pump optimization calculation. The traditional model based on the number of the mass transfer units is high in model calculation speed but low in accuracy due to the fact that the assumption is excessive, model parameters are difficult to obtain, and the condition of multi-layer spraying is not considered.
In recent years, with the popularization of artificial intelligence and big data technology, data-driven models are widely applied, however, a modeling method based on data only can obtain a model with high prediction accuracy under the condition of a large amount of high-quality historical operating data, but cannot obtain a satisfactory model under the condition of insufficient historical data and low data quality, and particularly can obtain a result which is contrary to a process mechanism under the condition of over-fitting or under-fitting. Meanwhile, the poor interpretability of the method also limits the application of the method in a scene with higher safety requirements.
At present, most of intelligent optimization regulation strategies for circulating pumps are mainly steady state optimization, problems such as startup and shutdown loss and the like possibly existing in actual circulating pump operation are not considered, real-time optimization of circulating pump combination is difficult to achieve under variable working conditions, theoretically optimal effects are difficult to obtain in actual application, and a series of problems such as standard exceeding, circulating pump damage, excessive energy consumption and the like are easily caused. Therefore, the intelligent optimization regulation strategy of the circulating pump has a further optimization space.
Disclosure of Invention
Aiming at the problems existing in the current intelligent regulation and control technology of the slurry circulating pump and overcoming the defects existing in the mechanism model and the data driving model prediction, the invention provides an intelligent regulation and control method of the circulating pump based on the accurate prediction of desulfurization process parameters, so as to solve the problems.
In order to achieve the purpose, the invention adopts the following technical scheme:
an intelligent control method of a circulating pump based on accurate prediction of desulfurization process parameters comprises the following steps:
the method comprises the following steps: based on SO 2 Numerical model of absorption mechanism and model based on mass transfer unit number to construct outlet SO 2 A concentration prediction mechanism model is adopted, an intelligent optimization algorithm is adopted, and the historical operation data is utilized to identify and correct key parameters, SO that an outlet SO with mechanism and data fused is constructed 2 The concentration prediction model has higher precision and higher interpretability on the basis of realizing second-level calculation, and meets the requirement of optimized regulation and control;
step two: setting related parameters for intelligent regulation and control of the circulating pump, acquiring operation parameters under the current conditions in real time, and judging when to perform the third step according to the set related parameters and the acquired real-time operation parameters;
step three: outlet SO obtained on the basis of step one 2 And the concentration prediction model calculates the optimal circulating pump configuration under the current working condition and the optimization target according to the related parameters set in the step two and the obtained real-time operation parameters.
Preferably, the outlet SO is constructed 2 The concentration prediction mechanism model comprises two sub-models, namely a liquid drop motion model and a mass transfer reaction model.
Preferably, the droplet motion model is used to calculate the surface area a of the droplet per unit volume, assuming that the droplet is a standard sphere, a is calculated using the following equation:
Figure BDA0004006307210000021
in the formula: n is l Represents the number of discharged droplets per unit time, a l Denotes the surface area of a single droplet, v l Represents the volume of the drop per unit time;
surface area a of a single droplet l Calculated from the following formula:
Figure BDA0004006307210000031
wherein: d l Represents the slurry drop diameter;
the number n of droplets ejected per unit time l Calculated from the following formula:
Figure BDA0004006307210000032
wherein: l represents the slurry flow rate;
volume v of liquid drop passing per unit time l Calculated from the following formula:
Figure BDA0004006307210000033
wherein: d represents the diameter of the absorption column, u l Represents the drop falling speed;
the simultaneous expression is as follows:
Figure BDA0004006307210000034
the falling speed of the liquid drop is obtained by a liquid drop stress model:
Figure BDA0004006307210000035
wherein: t represents the drop fall time, ρ 1 Denotes the slurry density, p g To representFlue gas density, g represents gravitational acceleration, C D Denotes the coefficient of drag, u g Indicating the flue gas velocity.
Preferably, the mass transfer reaction model describes a mass transfer process between gas and liquid and a reaction occurring inside the liquid droplet; SO of the slurry based on mass transfer theory 2 The absorption rate was calculated using the following formula:
Figure BDA0004006307210000036
wherein:
Figure BDA0004006307210000037
is SO 2 In a combination of>
Figure BDA0004006307210000038
Is SO 2 Based on the total mass transfer coefficient of the air film, and>
Figure BDA0004006307210000039
is SO in flue gas 2 Is based on the mole fraction of->
Figure BDA00040063072100000310
SO being in equilibrium with the slurry phase 2 The mole fraction of (c);
Figure BDA00040063072100000311
derived from henry's law:
Figure BDA00040063072100000312
wherein:
Figure BDA00040063072100000313
is SO 2 Based on the Henry coefficient of (A), based on the number of cells in the cell>
Figure BDA00040063072100000314
Is SO in slurry 2 A mole fraction;
total gas film mass transfer coefficient
Figure BDA00040063072100000315
Obtained by the following formula:
Figure BDA0004006307210000041
wherein:
Figure BDA0004006307210000042
is SO 2 The mass transfer coefficient of the air film is greater or less>
Figure BDA0004006307210000043
Is SO 2 The mass transfer coefficient of the liquid film, E is a reaction enhancement coefficient;
the reaction enhancement factor E is a parameter related to the pH of the slurry and is described by the following fit:
E＝βpH α
wherein alpha and beta are constants, and the pH is the pH value of the slurry;
mass transfer coefficient of air film
Figure BDA0004006307210000044
The solution is solved simultaneously by:
Figure BDA0004006307210000045
Figure BDA0004006307210000046
Figure BDA0004006307210000047
Figure BDA0004006307210000048
wherein:
Figure BDA0004006307210000049
is SO 2 P is the total pressure in the column, T is the temperature in the column, sh is the shewood number, sc is the Schmidt number, M is the gas phase diffusion coefficient of air Is the molar mass of air, is present in the blood>
Figure BDA00040063072100000410
Is SO 2 Molar mass of (b), V air Is the molar volume of air>
Figure BDA00040063072100000411
Is SO 2 The molar volume of (c);
Figure BDA00040063072100000412
the liquid film mass transfer coefficient is calculated by the following formula:
Figure BDA00040063072100000413
Figure BDA00040063072100000414
/>
Figure BDA00040063072100000415
wherein:
Figure BDA00040063072100000416
is SO 2 Diffusion coefficient in the serum, f is the drop oscillation frequency,. Sup.,. Sup.>
Figure BDA00040063072100000417
Is the gas molecular volume, μ l Is the dynamic viscosity of water;
based on conservation of material, fromSO removed from flue gas 2 In an amount equal to the SO absorbed by the slurry 2 Expressed by the following formula:
Figure BDA0004006307210000051
wherein: n is a radical of hydrogen A Is SO 2 Mass transfer flux of gas, S is the cross-sectional area of the desulfurizing tower, G is the flow rate of flue gas m 3 /s，p A Is SO 2 The partial pressure of the gas;
ignore
Figure BDA0004006307210000052
Obtaining by integration:
Figure BDA0004006307210000053
wherein: y is in Indicating the flue gas inlet SO 2 Mole fraction, y out Is an outlet SO 2 Mole fraction, z out Is the height of the flue gas outlet, z in The height of a flue gas inlet is defined, and NTU is the number of mass transfer units;
when the absorption tower is provided with a plurality of spraying layers, SO between two spraying layers or between the lowest spraying layer and the flue gas inlet 2 The absorption capacity is described by the following formula:
y n ＝y n-1 exp(-NTU n )
wherein: y is n For flue gases SO at different locations 2 The mole fraction is the flue gas SO at the inlet of the absorption tower when n =0 2 Mole fraction when n>At 0, y n-1 For SO at different spray level positions 2 Mole fraction, NTU n The number of mass transfer units of the slurry between the nth spraying layer and the (n-1) th layer or between the lowest spraying layer and the flue gas inlet is the number;
NTU n the first part is NTU of slurry sprayed from the nth spraying layer; the second part is NTU of the slurry sprayed from the upper spraying layer; NTU 'of slurry discharged from No. n shower layer' n ：
Figure BDA0004006307210000054
Wherein:
Figure BDA0004006307210000055
SO of slurry sprayed from the n-th spray layer 2 Total gas film mass transfer coefficient, a n Is the surface area of the slurry sprayed from the n-th spray layer in a unit volume, z n Is the height of the flue gas outlet, z n-1 Is the flue gas inlet height;
assuming that the NTU of the slurry sprayed from the upper spray level is proportional to the NTU n-1 Coefficient of proportionality of k n Then NTU n Represented by the formula:
NTU n ＝NTU′ n +k n NTU n-1
calculating downwards layer by layer from the upper layer by the formula, and calculating all the mass transfer units, and further calculating the SO of the flue gas outlet layer by layer 2 The mole fraction is shown as the following formula:
Figure BDA0004006307210000061
wherein: y is outlet Is a flue gas outlet SO 2 Mole fraction, y 0 Is an inlet SO 2 The mole fraction, m is the total number of spraying layers, and K is a correction factor.
Preferably, the SO established above 2 In the concentration prediction model, historical operating data is adopted to identify key parameters, and the identification process is described by the following optimization problems:
Figure BDA0004006307210000062
wherein:
Figure BDA0004006307210000063
is the ith actual outlet flue gasSO 2 Mole fraction, y outlet，i Calculating the outlet flue gas SO for the ith 2 Mole fraction, x is the total number of identification data samples, one.
The relation between the parameter to be identified and the model is nonlinear, a heuristic intelligent algorithm is adopted for optimizing, after the parameter range is determined, the optimization is further refined in a mode including grid search or gradient reduction, so that the optimal identification parameter is obtained.
Preferably, the related parameters intelligently controlled by the circulation pump include: setting a shortest optimization interval t interual,min H, the time interval between two adjustments is longer than a certain time, which can be generally set to 1-6 hours; setting optimized patience interval delta p ，mg/m 3 When discharging SO 2 When the concentration value is within the upper limit and the lower limit of the tolerance interval, the circulating pump cannot be adjusted; setting optimization objectives
Figure BDA0004006307210000064
mg/m 3 Target SO as optimization system 2 Outlet concentration.
The circulating pump is used as large-scale equipment, and the service life of the circulating pump can be greatly shortened by frequent switching of the circulating pump, so that the safety production of a desulfurization system is damaged. The shortest optimization interval is set during actual control, so that the time interval between two adjustments is longer than a certain time length. Setting an optimized tolerance interval as the outlet SO 2 When the concentration value is within the upper limit and the lower limit of the tolerance interval, the configuration of the circulating pump cannot be adjusted according to the optimized value. The aim of this is to reduce the number of times the circulation pump is adjusted as much as possible, and to further reduce the energy consumption of the circulation pump. Moreover, the outlet SO is adopted in the actual use process 2 The measured value of the concentration is used as a basis for judging whether the desulfurization device operates in a tolerance interval, the robustness and the stability of the system can be further improved, and the phenomenon of exceeding standards caused by model mismatch and the like is prevented.
In the actual operation process, due to the limitation of emission standards, the number of exceeding standards needs to be reduced as much as possible. To preventStopping at the outlet SO within the shortest optimum interval 2 The concentration exceeds the standard, and the optimization target is slightly lower than the outlet SO 2 Concentration emission limit.
Preferably, the real-time acquisition of the operating parameters under the current conditions includes, but is not limited to: inlet SO 2 Concentration, outlet SO 2 Concentration, circulation pump opening state, slurry pH value and flue gas flow.
Preferably, in step three, assuming that there are m spraying layers, the optimization problem of the operation cost of the desulfurization system is represented by the following formula (an equation for solving the lowest cost of the optimal operation of the circulating pump):
Figure BDA0004006307210000071
Figure BDA0004006307210000072
in the formula: s n The running state of the circulating pump for supplying slurry to the nth spraying layer is expressed as s n When the value is not less than 0, the circulation pump is in a closed state n When the power consumption is not less than 1, the circulation pump is in an on state, cn is the power of the circulation pump for supplying slurry to the nth spraying layer, and the power consumption includes the running power consumption, the equipment depreciation cost, the unit, y outlet For calculated outlet flue gas SO 2 Mole fraction, y outlet，target Is a set outlet SO 2 Optimization goal of mole fraction.
Preferably, in the optimization process, because the total combination number of the circulating pumps is limited, in order to ensure the calculation precision, the optimal circulating pump configuration under the current working condition is obtained by exhaustive solution in a parallel calculation mode.
Compared with the prior art, the invention has the beneficial effects that:
1. outlet SO for establishing mechanism and data driven fusion 2 The concentration prediction model has higher precision on the basis of higher calculation speed, and meets the requirement of optimized regulation and control;
2. an intelligent circulating pump regulation and control method (strategy) considering the actual running requirement of the circulating pump is constructed; based on the actual operation requirement of the circulating pump, the optimization target value is adjusted, the shortest optimization interval and the optimized tolerance interval are added, the real-time optimization control of the circulating pump is realized, and the cost of the optimization operation of the circulating pump is reduced.
Drawings
FIG. 1 is a flow chart of an intelligent control method for a circulation pump based on accurate prediction of desulfurization process parameters in an embodiment of the present invention;
FIG. 2 is a schematic diagram of the optimal regulation of the circulation pump in the embodiment of the invention;
FIG. 3 is a flow chart of an optimal regulation strategy of a circulation pump according to an embodiment of the present invention;
FIG. 4 is a combination diagram of a circulating pump configuration of a desulfurization unit in an embodiment of the present invention;
FIG. 5 is a comparison graph of the combined energy consumption of the circulating pump configuration of the desulfurization unit in the example of the present invention;
FIG. 6 shows the actual SO of the desulfurizer in the embodiment of the present invention 2 Concentration value and SO 2 A comparison graph of the predicted values of the prediction model;
FIG. 7 shows the configuration of the front and rear circulation pumps and the SO outlet are optimized under different working conditions in the embodiment of the present invention 2 And (4) concentration.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
Referring to fig. 1, the method for intelligently regulating and controlling the circulating pump based on the accurate prediction of desulfurization process parameters is based on the SO based on the desulfurization process mechanism and combined with data drive 2 The concentration prediction model adopts an intelligent circulating pump regulation and control strategy to control the circulating pump to operate optimally in real time by optimizing the configuration of the circulating pump under different working conditions, and specifically comprises the following steps:
step one (construction of export SO) 2 Concentration prediction model): combined with conventional SO-based 2 Numerical model of absorption mechanism and model based on mass transfer unit number to construct outlet SO 2 A concentration prediction mechanism model is adopted, an intelligent optimization algorithm is adopted, and the historical operating data is utilized to identify and correct key parameters, SO that an outlet SO with mechanism and data fused is constructed 2 The concentration prediction model has higher precision and higher interpretability on the basis of realizing second-level calculation, and meets the requirement of optimized regulation and control;
step two: setting related parameters (setting control system parameters) for intelligent control of the circulating pump, including setting shortest optimization interval, optimizing tolerance interval, optimizing target and the like, and acquiring operation parameters (acquiring real-time parameters) under the current conditions in real time, including but not limited to inlet SO 2 Concentration, outlet SO 2 Judging whether to perform the third step (judging whether to perform optimization) according to the set relevant parameters and the obtained real-time running parameters, such as concentration, the opening state of a circulating pump, the pH value of the slurry, the flow rate of flue gas and the like;
step three (calculating the optimal circulating pump combination according to the set parameters and the real-time parameters): outlet SO based on mechanism and data fusion obtained in step one 2 And the concentration prediction model calculates the optimal circulating pump configuration under the current working condition and the optimization target according to the related parameters (such as the set optimization target number) set in the step two and the obtained real-time operation parameters.
Constructed Outlet SO 2 The concentration prediction mechanism model comprises two part submodels, namely a liquid drop motion model and a mass transfer reaction model.
The droplet motion model is mainly used for calculating the surface area a of a droplet in a unit volume, and assuming that the droplet is in a standard spherical shape, the calculation can be performed in the following way:
Figure BDA0004006307210000091
in the formula: n is l Represents the number of ejected droplets per unit time, a l Denotes the surface area of a single droplet, m 2 ，v l Denotes the volume of the passage of a droplet per unit time, m 3 /s；
Surface area a of a single droplet l Can be calculated from:
Figure BDA0004006307210000092
wherein: a is l Denotes the surface area of a single droplet, m 2 ，d l Represents the diameter of the slurry drop, m;
number of ejected droplets n per unit time l Can be calculated from:
Figure BDA0004006307210000093
wherein: n is l The number of ejected droplets per unit time is expressed, L represents the flow rate of the slurry, m 3 /s；
Volume v of liquid drop passing per unit time l Can be calculated from:
Figure BDA0004006307210000094
wherein: v. of l Denotes the volume of the passage of a droplet per unit time, m 3 S, D absorption column diameter, m, u l Represents the drop falling speed, m/s;
the formula above can be combined to obtain:
Figure BDA0004006307210000095
the falling speed of the liquid drop can be obtained by a liquid drop stress model:
Figure BDA0004006307210000096
wherein: t represents the drop fall time, s, ρ 1 Denotes the density of the slurry in kg/m 3 ，ρ g Denotes the smoke density kg/m 3 And g represents the acceleration of gravity of 9.8m/s 2 ，C D Denotes the coefficient of drag, u g Representing the flue gas velocity, m/s; the falling speed and the relative flue gas speed of the liquid drops can be obtained according to the formula, so that the surface area a of the liquid drops in unit volume can be obtained.
The mass transfer reaction model describes the mass transfer process between gas and liquid and the reaction inside the liquid drop; SO of slurry based on mass transfer theory 2 The absorption rate can be calculated using the following formula:
Figure BDA0004006307210000101
wherein:
Figure BDA0004006307210000102
is SO 2 Absorption rate of (1), mol/(m) 2 ·s)，/>
Figure BDA0004006307210000103
Is SO 2 Total gas film mass transfer coefficient, mol/(m) 2 ·s)，/>
Figure BDA0004006307210000104
Is SO in flue gas 2 Is based on the mole fraction of->
Figure BDA0004006307210000105
SO being in equilibrium with the slurry phase 2 Can be obtained from henry's law:
Figure BDA0004006307210000106
wherein:
Figure BDA0004006307210000107
is SO 2 Has a Henry coefficient of->
Figure BDA0004006307210000108
Is SO in slurry 2 A mole fraction;
total gas film mass transfer coefficient
Figure BDA0004006307210000109
Can be obtained by the following formula:
Figure BDA00040063072100001010
/>
wherein:
Figure BDA00040063072100001011
is SO 2 Mass transfer coefficient of gas film, mol/(m) 2 ·s)，/>
Figure BDA00040063072100001012
Is SO 2 Mass transfer coefficient of liquid film, mol/(m) 2 S), E is the reaction enhancement factor;
the reaction enhancement factor E is a parameter related to the pH of the slurry and is described by the following fit:
E＝βpH α
wherein alpha and beta are constants, and the pH is the pH value of the slurry;
mass transfer coefficient of air film
Figure BDA00040063072100001013
The solution can be solved simultaneously by:
Figure BDA00040063072100001014
Figure BDA00040063072100001015
Figure BDA00040063072100001016
Figure BDA0004006307210000111
wherein:
Figure BDA0004006307210000112
is SO 2 Gas phase diffusion coefficient of (c), m 2 P is the total pressure in the column, pa, T is the temperature in the column, K, sh is the shewood number, sc is the Schmidt number, M is the internal pressure in the column air Is the molar mass of air, g/mol->
Figure BDA0004006307210000113
Is SO 2 Molar mass of (2), g/mol, V air Is the molar volume of air, cm 3 /mol，/>
Figure BDA0004006307210000114
Is SO 2 Molar volume of (c), cm 3 /mol；
Figure BDA0004006307210000115
The liquid film mass transfer coefficient can be calculated by:
Figure BDA0004006307210000116
Figure BDA0004006307210000117
Figure BDA0004006307210000118
wherein:
Figure BDA0004006307210000119
is SO 2 In slurryDiffusion coefficient m 2 /s，/>
Figure BDA00040063072100001110
Is the volume of gas molecule cm 3 /mol，μ l Is the dynamic viscosity of water, MPa.s, f is the oscillation frequency of the liquid drop, hz;
SO removed from flue gas based on material conservation 2 In an amount equal to the SO absorbed by the slurry 2 This can be represented by the following formula:
Figure BDA00040063072100001111
wherein: n is a radical of A Is SO 2 Gas mass transfer flux mol/(s.m) 2 ) And a is the surface area m of the liquid drops in the unit volume of the desulfurizing tower 2 /m 3 S is the cross-sectional area of the desulfurizing tower, m 2 G is the flue gas flow m 3 /s，p A Is SO 2 Partial pressure of gas, pa;
according to neglect
Figure BDA00040063072100001112
Integration can give: />
Figure BDA00040063072100001113
Wherein: y is in Indicating the flue gas inlet SO 2 Mole fraction, y out Is an outlet SO 2 Mole fraction, z out Is the height of the flue gas outlet, m, z in Is the height of a flue gas inlet, m and G are the flow rates of the flue gas, mol/m 3 (ii) a NTU is the number of mass transfer units and can describe SO of slurry 2 The absorption capacity.
When the absorption tower is provided with a plurality of spraying layers, SO between two spraying layers or between the lowest spraying layer and the flue gas inlet 2 The absorption capacity can also be described by the above formula:
y n ＝y n-1 exp(-NTU n )
wherein: y is n For flue gases SO of different locations 2 The mole fraction is the flue gas SO at the inlet of the absorption tower when n =0 2 Mole fraction of when n>At 0, y n-1 For SO at different spray level positions 2 Mole fraction, NTU n Is between the nth spraying layer and the (n-1) th layer>1 hour) or the number of mass transfer units of the slurry between the lowest spraying layer and the flue gas inlet (n =1 hour);
NTU n typically consisting of two parts, the first part being the NTU of the slurry sprayed from the nth spray level. The second part is the NTU of the slurry sprayed from the upper spray level. NTU 'of slurry discharged from spray layer n' n ：
Figure BDA0004006307210000121
Wherein:
Figure BDA0004006307210000122
SO of slurry sprayed from the n-th spray layer 2 Total gas film mass transfer coefficient, mol/(m) 2 ·s)，a n Is the surface area of the slurry sprayed from the n-th spray layer in a unit volume, m 2 /m 3 ，z n Is the height of the flue gas outlet, m, z n-1 Is the flue gas inlet height, m;
assuming that the NTU of the slurry sprayed from the upper spray level is proportional to the NTU n-1 The proportionality coefficient is k n Then NTU n Can be represented by the following formula:
NTU n ＝NTU′ n +k n NTU n-1
the above formula can be used for calculating downwards layer by layer from the upper layer, namely all the mass transfer units can be calculated, and further the SO of the flue gas outlet can be calculated layer by layer 2 The mole fraction is shown as the following formula:
Figure BDA0004006307210000123
wherein: y is outlet Is a flue gas outlet SO 2 Mole fraction, y 0 Is an inlet SO 2 Mole fraction, m is the total number of sprayed layers, layers (n refers to the number of sprayed layers, the first one such, the second one, m refers to the total number of sprayed layers, for a total of m layers);
in order to further improve the accuracy of the model, a correction factor K is added in the model to modify errors caused by uneven mixing of gas and liquid, liquid drop collision and the like in the desulfurization tower:
Figure BDA0004006307210000131
preferably, the SO established above 2 Part of unknown parameters in the concentration prediction model still need to be identified by adopting historical operating data, and the Root Mean Square Error (RMSE) is used as a loss function to better describe SO of an actual desulfurization device 2 To improve the prediction accuracy, the identification process can be described by the following optimization problem:
Figure BDA0004006307210000132
wherein:
Figure BDA0004006307210000133
is the ith actual outlet flue gas SO 2 Mole fraction, y outlet，i Calculating the outlet flue gas SO for the ith 2 Mole fraction, x is the total number of identification data samples, one.
The relationship between the above-mentioned parameter to be identified and the model is non-linear. A heuristic intelligent algorithm such as a group intelligent algorithm can be adopted for optimizing, after the rough range of the parameters is determined, the optimization is further refined by adopting a grid search or gradient descending mode and the like, so as to obtain the optimal identification parameters.
Referring to fig. 2, the related parameters (parameters to be set by a policy) for intelligently regulating the circulation pump include: setting a shortest optimization interval t interval,min H, so that it is adjusted twiceThe time interval is longer than a certain time length, which can be generally set to 1-6 hours (specifically, corresponding adjustment can be performed according to actual conditions); setting optimized patience interval delta p ，mg/m 3 When discharging SO 2 When the concentration value is within the upper limit and the lower limit of the tolerance interval, the circulating pump cannot be adjusted; setting optimization objectives
Figure BDA0004006307210000134
mg/m 3 Target SO as optimization system 2 Outlet concentration.
The circulating pump is used as large-scale equipment, and the service life of the circulating pump can be greatly shortened by frequent switching of the circulating pump, so that the safety production of a desulfurization system is damaged. The shortest optimization interval is set during actual control, so that the time interval between two adjustments is longer than a certain time length. Setting an optimized tolerance interval as the outlet SO 2 When the concentration value is within the upper limit and the lower limit of the tolerance interval, the configuration of the circulating pump cannot be adjusted according to the optimized value. The aim of this is to reduce the number of times the circulation pump is adjusted as much as possible, and to further reduce the energy consumption of the circulation pump. Moreover, the outlet SO is adopted in the actual use process 2 The measured value of the concentration is used as a basis for judging whether the desulfurization device operates in a tolerance interval, the robustness and the stability of the system can be further improved, and the phenomenon of exceeding standards caused by model mismatch and the like is prevented.
In the actual operation process, due to the limitation of emission standards, the number of exceeding standards needs to be reduced as much as possible. To prevent SO from being discharged within the shortest optimum interval 2 The concentration exceeds the standard, and the optimization target is slightly lower than that of the outlet SO 2 Concentration emission limit.
Referring to FIG. 3, operating parameters under current conditions, including but not limited to outlet SO, are obtained in real time 2 Concentration of
Figure BDA0004006307210000144
The pH value of the slurry, the flue gas flow G and the like, and judging the outlet SO 2 Whether the concentration is not in the range of tolerance to heart and the interval t from the last optimization interval Whether greater than shortestOptimizing interval t interval,min When the two conditions are simultaneously yes, the optimal circulating pump configuration combination calculated in the step three is started, and the two conditions for judging to start to execute the step three can be expressed by the following method:
Figure BDA0004006307210000141
t interval >t interval,min
outlet SO based on mechanism and data fusion obtained in step one 2 And (3) according to the optimization target number set in the second step and the obtained real-time parameters, assuming that m spraying layers exist, the optimization problem of the operation cost of the desulfurization system is represented by the following formula (solving an equation of the lowest cost of the optimal operation of the circulating pump):
Figure BDA0004006307210000142
Figure BDA0004006307210000143
in the formula: s n The running state of the circulating pump for supplying slurry to the nth spraying layer is expressed as s n When the value is not less than 0, the circulation pump is in a closed state n When =1, it means that the circulation pump is in the on state, c n Power of circulating pump for supplying slurry to spray layer n, including power consumption, depreciation cost, unit, y outlet For calculated outlet flue gas SO 2 Mole fraction, y outlet，target Is a set outlet SO 2 Optimization goal of mole fraction.
In the optimization process, because the total combination number of the circulating pumps is limited, in order to ensure the calculation precision, the parallel calculation mode can be adopted for exhaustive solution, and the optimal circulating pump configuration under the current working condition is obtained.
Example 2
In order to verify the effectiveness of the method, engineering practice is carried out on a limestone-gypsum wet desulphurization device of an ultra-low emission system matched with a certain 1000MW coal-fired power generating set, and specific application results and analysis are as follows:
as shown in fig. 4, the desulfurization apparatus has 1, 2, 3, 4, and 5 circulation pumps, each corresponding to a spray level. Considering all the on-off states of the circulating pumps, there are 32 circulating pump combination configurations, and the energy consumption costs of the different configurations are shown in fig. 5.
The outlet SO provided by the invention 2 The comparison result between the predicted value and the actual value of the concentration prediction model is shown in fig. 6, and the curves of the predicted value and the actual value are relatively close. Especially, when the actual value is in the rising or falling trend, the predicted value can be accurately matched with the change of the actual value, and the change shows that the outlet SO is 2 The concentration prediction model has high prediction accuracy and can better reflect the outlet SO 2 And (4) concentration change rule. The RMSE for this model was 2.20mg/m 3 。
To identify the obtained outlet SO 2 Constructing an intelligent control system of the circulating pump on the basis of the concentration prediction model, and when the shortest optimization interval is set to be 1h, respectively setting the upper limit and the lower limit of the patience interval to be 35mg/m 3 And 10mg/m 3 The optimization target is set to 15mg/m 3 In the process, the configuration of the front and rear circulating pumps and the SO outlet are optimized 2 The concentration results are shown in FIG. 7. It can be seen that when the optimized regulation system parameters are configured as above, the outlet SO of the desulfurization device 2 The concentration does not exceed the standard within 168h, and the energy consumption of the circulating pump is greatly reduced compared with that before optimization.
The invention combines the traditional SO-based 2 Numerical model of absorption mechanism and model based on mass transfer unit number to construct outlet SO 2 A concentration prediction mechanism model is adopted, an intelligent optimization algorithm is adopted, and the historical operating data is utilized to identify and correct key parameters, SO that an outlet SO with mechanism and data fused is constructed 2 A concentration prediction model; on the basis of realizing second-level (even millisecond-level) calculation, the method has higher precision and higher interpretability, and meets the requirement of optimized regulation. And an intelligent circulating pump regulation and control strategy considering the actual running requirement of the circulating pump is constructed. Based on actual operation of circulating pumpAnd the optimization target value is adjusted, the shortest optimization interval and the optimized tolerance interval are added, the real-time optimization control of the circulating pump is realized, and the optimization operation cost of the circulating pump is reduced.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the spirit of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. An intelligent control method of a circulating pump based on accurate prediction of desulfurization process parameters is characterized by comprising the following steps:
the method comprises the following steps: based on SO 2 Numerical model of absorption mechanism and model based on mass transfer unit number to construct outlet SO 2 A concentration prediction mechanism model is adopted, an intelligent optimization algorithm is adopted, and the historical operating data is utilized to identify and correct key parameters, SO that an outlet SO with mechanism and data fused is constructed 2 A concentration prediction model;
step two: setting related parameters for intelligent regulation and control of the circulating pump, acquiring operation parameters under the current conditions in real time, and judging when to perform the third step according to the set related parameters and the acquired real-time operation parameters;
step three: outlet SO based on mechanism and data fusion obtained in step one 2 And the concentration prediction model is used for calculating the optimal circulating pump configuration under the current working condition and the optimization target according to the related parameters set in the step two and the obtained real-time operation parameters.
2. The intelligent circulating pump regulating and controlling method based on accurate desulfurization process parameter prediction as claimed in claim 1, wherein: constructed Outlet SO 2 The concentration prediction mechanism model comprises two sub-models, namely droplet transportationKinetic models and mass transfer reaction models.
3. The intelligent control method for the circulating pump based on the accurate prediction of the desulfurization process parameters, according to claim 2, is characterized in that: the drop motion model is used to calculate the surface area a of the drop per unit volume, assuming the drop is a standard sphere, a is calculated using the following equation:
Figure FDA0004006307200000011
in the formula: n is l Represents the number of discharged droplets per unit time, a l Denotes the surface area of a single droplet, v l Represents the volume of the drop per unit time;
surface area a of a single droplet l Calculated from the following formula:
Figure FDA0004006307200000012
wherein: d l Represents the slurry drop diameter;
number of ejected droplets n per unit time l Calculated from the following formula:
Figure FDA0004006307200000013
wherein: l represents the slurry flow rate;
volume v of liquid drop passing per unit time l Calculated from the following formula:
Figure FDA0004006307200000014
wherein: d represents the diameter of the absorption column, u l Represents the drop falling speed;
the simultaneous expression is as follows:
Figure FDA0004006307200000021
the falling speed of the liquid drop is obtained by a liquid drop stress model:
Figure FDA0004006307200000022
wherein: t represents the drop fall time, ρ l Denotes the slurry density, p g Denotes the smoke density, g denotes the acceleration of gravity, C D Denotes the coefficient of drag, u g Indicating the flue gas velocity.
4. The intelligent circulating pump regulating and controlling method based on accurate prediction of desulfurization process parameters as claimed in claim 3, wherein: the mass transfer reaction model describes the mass transfer process between gas and liquid and the reaction inside the liquid drop; SO of slurry based on mass transfer theory 2 The absorption rate was calculated using the following formula:
Figure FDA0004006307200000023
wherein:
Figure FDA0004006307200000024
is SO 2 Is taken up and taken up>
Figure FDA0004006307200000025
Is SO 2 The total mass transfer coefficient of the air film, and>
Figure FDA0004006307200000026
is SO in flue gas 2 Is based on the mole fraction of->
Figure FDA0004006307200000027
SO being in equilibrium with the slurry phase 2 Is prepared from (A) and (B)A mole fraction;
Figure FDA0004006307200000028
derived from henry's law:
Figure FDA0004006307200000029
wherein:
Figure FDA00040063072000000210
is SO 2 Has a Henry coefficient of->
Figure FDA00040063072000000211
Is SO in slurry 2 A mole fraction;
total gas film mass transfer coefficient
Figure FDA00040063072000000212
Obtained by the following formula:
Figure FDA00040063072000000213
wherein:
Figure FDA00040063072000000214
is SO 2 The mass transfer coefficient of the air film, based on the mass transfer coefficient of the blood vessel>
Figure FDA00040063072000000215
Is SO 2 The mass transfer coefficient of the liquid film, E is a reaction enhancement coefficient;
the reaction enhancement factor E is a parameter related to the pH of the slurry and is described by the following fit:
E＝βpH α
wherein alpha and beta are constants, and the pH is the pH value of the slurry;
mass transfer coefficient of air film
Figure FDA0004006307200000031
The solution is solved simultaneously by:
Figure FDA0004006307200000032
Figure FDA0004006307200000033
Figure FDA0004006307200000034
Figure FDA0004006307200000035
wherein:
Figure FDA0004006307200000036
is SO 2 P is the total pressure in the column, T is the temperature in the column, sh is the shewood number, sc is the Schmidt number, M is the gas phase diffusion coefficient of air Is the molar mass of air, is present in the blood>
Figure FDA0004006307200000037
Is SO 2 Molar mass of (A), V air Is the molar volume of air>
Figure FDA0004006307200000038
Is SO 2 The molar volume of (a);
Figure FDA0004006307200000039
the liquid film mass transfer coefficient is calculated by the following formula:
Figure FDA00040063072000000310
Figure FDA00040063072000000311
/>
Figure FDA00040063072000000312
wherein:
Figure FDA00040063072000000313
is SO 2 Diffusion coefficient in the serum, f is the drop oscillation frequency,. Sup.,. Sup.>
Figure FDA00040063072000000314
Is the gas molecular volume, μ l Is the dynamic viscosity of water;
SO removed from flue gas based on material conservation 2 In an amount equal to the SO absorbed by the slurry 2 Expressed by the following formula:
Figure FDA00040063072000000315
wherein: n is a radical of hydrogen A Is SO 2 Mass transfer flux of gas, S is cross-sectional area of desulfurizing tower, G is flow rate of flue gas 3 /s，p A Is SO 2 Partial pressure of gas;
ignore
Figure FDA0004006307200000041
Integrating to obtain:
Figure FDA0004006307200000042
wherein: y is in To representFlue gas inlet SO 2 Mole fraction, y out Is an outlet SO 2 Mole fraction, z out Is the height of the flue gas outlet, z in The height of a flue gas inlet is defined, and NTU is the number of mass transfer units;
when the absorption tower is provided with a plurality of spraying layers, SO between two spraying layers or between the lowest spraying layer and the flue gas inlet 2 The absorption capacity is described by the following formula:
y n ＝y n-1 exp(-NTU n )
wherein: y is n For flue gases SO of different locations 2 The mole fraction is the flue gas SO at the inlet of the absorption tower when n =0 2 Mole fraction of when n>At 0, y n-1 For SO at different spray level positions 2 Mole fraction, NTU n The number of mass transfer units of the slurry between the nth spraying layer and the (n-1) th layer or between the lowest spraying layer and the flue gas inlet is the number;
NTU n the first part is NTU of slurry sprayed from the nth spraying layer; the second part is NTU of the slurry sprayed from the upper spraying layer; NTU 'of slurry discharged from n-th shower layer' n ：
Figure FDA0004006307200000043
Wherein:
Figure FDA0004006307200000044
SO of slurry sprayed from the n-th spray layer 2 Total gas film mass transfer coefficient, a n Is the surface area of the slurry sprayed from the n-th spray layer in a unit volume, z n Is the height of the flue gas outlet, z n-1 Is the flue gas inlet height;
assuming that the NTU of the slurry sprayed from the upper spray level is proportional to the NTU n-1 Coefficient of proportionality of k n Then NTU n Represented by the formula:
NTU n ＝NTU′ n +k n NTU n-1
from the upper layer to the layer by layer according to the formulaCalculating to obtain the total mass transfer unit number and the layer-by-layer calculated SO 2 The mole fraction is shown as the following formula:
Figure FDA0004006307200000051
wherein: y is outlet Is a flue gas outlet SO 2 Mole fraction, y 0 Is an inlet SO 2 The mole fraction, m is the total number of spraying layers, and K is a correction factor.
5. The intelligent circulating pump regulating and controlling method based on accurate prediction of desulfurization process parameters, according to claim 4, is characterized in that: SO thus established 2 In the concentration prediction model, historical operating data is adopted to identify key parameters, and the identification process is described by the following optimization problems:
Figure FDA0004006307200000052
wherein:
Figure FDA0004006307200000053
is the ith actual outlet flue gas SO 2 Mole fraction, y outlet,i Calculating the outlet flue gas SO for the ith 2 Mole fraction, x is the total number of identification data samples.
6. The intelligent circulating pump regulating and controlling method based on accurate prediction of desulfurization process parameters as recited in claim 5, wherein: the relation between the parameters to be identified and the model is nonlinear, a heuristic intelligent algorithm is adopted for optimizing, after the parameter range is determined, the optimization is further refined in a mode including grid search or gradient descent, and the optimal identification parameters are obtained.
7. Desulfurization-based according to claim 1The intelligent control method for the circulating pump with the accurately predicted process parameters is characterized by comprising the following steps of: the related parameters of the intelligent control of the circulating pump comprise: setting the shortest optimization interval to ensure that the time interval between two times of adjustment is longer than a certain duration; setting an optimized tolerance interval when SO is discharged 2 When the concentration value is within the upper limit and the lower limit of the tolerance interval, the circulating pump cannot be adjusted; setting an optimization objective as an objective SO for an optimization system 2 Outlet concentration.
8. The intelligent circulating pump regulating and controlling method based on accurate desulfurization process parameter prediction as claimed in claim 1, wherein: obtaining the operating parameters in real time under the current conditions includes, but is not limited to: inlet SO 2 Concentration, outlet SO 2 Concentration, circulation pump open state, slurry pH value, and flue gas flow.
9. The intelligent circulating pump regulating and controlling method based on accurate desulfurization process parameter prediction as claimed in claim 1, wherein: in step three, assuming that there are m spray levels, the optimization problem of the operating cost of the desulfurization system is represented by the following formula:
Figure FDA0004006307200000054
Figure FDA0004006307200000061
in the formula: s n The running state of the circulating pump for supplying slurry to the nth spraying layer is expressed as s n When the value is not less than 0, the circulation pump is in a closed state n When =1, it means that the circulation pump is in the on state, c n The power of the circulating pump for supplying slurry to the nth spraying layer comprises the power consumption of operation, the depreciation cost of equipment and y outlet For calculated outlet flue gas SO 2 Mole fraction, y outlet,target Is a set outlet SO 2 Optimization goal of mole fraction.
10. The intelligent circulating pump regulating and controlling method based on accurate desulfurization process parameter prediction of claim 9, wherein: in the optimization process, in order to ensure the calculation precision, the parallel calculation mode is adopted for exhaustive solution, and the optimal circulating pump configuration under the current working condition is obtained.
CN202211633396.4A 2022-12-19 2022-12-19 Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters Pending CN115853793A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211633396.4A CN115853793A (en) 2022-12-19 2022-12-19 Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211633396.4A CN115853793A (en) 2022-12-19 2022-12-19 Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters

Publications (1)

Publication Number Publication Date
CN115853793A true CN115853793A (en) 2023-03-28

Family

ID=85674151

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211633396.4A Pending CN115853793A (en) 2022-12-19 2022-12-19 Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters

Country Status (1)

Country Link
CN (1) CN115853793A (en)

Similar Documents

Publication Publication Date Title
CN111850591B (en) Combined diaphragm regulating valve device, alkaline water electrolysis hydrogen production system and control method
CN103771553B (en) The adaptive fuzzy control method of wet flue gas desulfurization waste water evaporation process
CN108710972B (en) Reservoir flood control scheduling method based on POA algorithm
CN107096371B (en) Wide-load-adjustable limestone slurry supply system and adjusting method
CN109126408A (en) A kind of wet desulphurization device and intelligent control method
CN115853793A (en) Intelligent circulating pump regulation and control method based on accurate prediction of desulfurization process parameters
CN113593653B (en) Wet desulfurization device overall economic benefit optimal control method
CN113198294B (en) Seawater desulfurization control method and equipment
CN115440310A (en) Desulfurization tower slurry circulating pump combination optimization method and system based on prediction technology
JP2529244B2 (en) Absorption liquid circulation controller for wet flue gas desulfurization equipment
CN114606523B (en) Variable-power hydrogen production system and automatic control system and method thereof
CN113655824B (en) Wet desulphurization spraying amount and spraying slurry pH value coupling control optimization method
CN110937616B (en) Intelligent setting system and method for alumina evaporation process
CN104944382A (en) Closed multistage absorption system for preparing sulfurous acid from sulfur dioxide
CN115651998B (en) Steel slag flue gas treatment system, system control method, electronic equipment and storage medium
CN111471822A (en) Control method of converter vaporization cooling system
CN115253652B (en) Desulfurization single-tower double-circulation automatic slurry supply control method and device
CN208356448U (en) A kind of tail gas absorber
CN204702508U (en) The closed circuit multistage absorption system of a kind of sulfurous gas sulfurous acid
CN113521998B (en) Sensitive and economical wet desulfurization outlet flue gas SO x Concentration control method
CN211925937U (en) Blowdown system of laboratory constant temperature and humidity air conditioner humidifier
CN213725766U (en) Slurry concentration control system of wet desulphurization absorption tower of thermal power generating unit
JPH02180616A (en) Control of waste gas desulfurization apparatus
CN214261370U (en) Atomizing system suitable for semidry flue gas desulfurization
CN211462651U (en) Spray scattering tower

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