CN112044270B - Desulfurization oxidation air system control method based on data-driven multiple models - Google Patents
Desulfurization oxidation air system control method based on data-driven multiple models Download PDFInfo
- Publication number
- CN112044270B CN112044270B CN202010828597.4A CN202010828597A CN112044270B CN 112044270 B CN112044270 B CN 112044270B CN 202010828597 A CN202010828597 A CN 202010828597A CN 112044270 B CN112044270 B CN 112044270B
- Authority
- CN
- China
- Prior art keywords
- caso
- data
- absorption tower
- concentration
- oxidation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/346—Controlling the process
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/46—Removing components of defined structure
- B01D53/48—Sulfur compounds
- B01D53/50—Sulfur oxides
- B01D53/501—Sulfur oxides by treating the gases with a solution or a suspension of an alkali or earth-alkali or ammonium compound
- B01D53/502—Sulfur oxides by treating the gases with a solution or a suspension of an alkali or earth-alkali or ammonium compound characterised by a specific solution or suspension
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/74—General processes for purification of waste gases; Apparatus or devices specially adapted therefor
- B01D53/80—Semi-solid phase processes, i.e. by using slurries
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/96—Regeneration, reactivation or recycling of reactants
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D2251/00—Reactants
- B01D2251/40—Alkaline earth metal or magnesium compounds
- B01D2251/404—Alkaline earth metal or magnesium compounds of calcium
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D2258/00—Sources of waste gases
- B01D2258/02—Other waste gases
- B01D2258/0283—Flue gases
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Analytical Chemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Treating Waste Gases (AREA)
Abstract
The invention discloses a desulfurization oxidation air system control method based on a data-driven multi-model, which comprises the following steps of S1: screening modeling data by using historical operating data of unit and CaSO 3 Assay concentration data is used as modeling original data; step S2: establishing an absorption column CaSO 3 A concentration prediction model; step S3: to real-time forecast obtained absorption tower CaSO 3 Calculating the concentration by adopting a sliding average method to obtain the CaSO of the absorption tower 3 Average concentration; step S4: controlling the start and stop of the oxidation fan; the method takes the historical data and the laboratory data stored by the unit for a long time as raw materials to carry out scheme design and modeling, does not need to carry out modeling test on the system, and has no influence on the safe operation of the unit; the soft measurement of the concentration of the calcium sulfite is realized by adopting modeling modes such as a neural network and the like, new measuring equipment is not required to be added, the frequency conversion transformation of the original oxidation air system is not required, only an industrial personal computer is required to be added, and the investment cost is low.
Description
Technical Field
The invention belongs to the technical field of thermal power plant oxidation fan control, and particularly relates to a desulfurization oxidation air system control method based on a data-driven multi-model.
Background
The oxidizing air system is an important component of the wet desulphurization system, the power consumption of the system is high, but the gypsum and the slurry CaSO of the absorption tower under partial working conditions are avoided 3 The concentration exceeds the limit, a continuous operation mode is generally adopted, and the operation mode causes great energy waste, so that the on-off of the oxidizing air system according to needs has important significance for energy conservation and consumption reduction of the thermal power generating unit desulfurization system.
At present, oxidation fan operation mode mainly has 3, is respectively: 1. the continuous operation mode of the oxidation fan; 2. slurry CaSO with absorption tower 3 The concentration is measured on line in real time in an oxidation fan on-off mode according to requirements; 3. the variable frequency control mode of the oxidation fan based on the fixed excess air coefficient has certain disadvantages in the prior art,
1. the continuous operation mode of the oxidation fan is as follows: most thermal power plants avoid gypsum and absorption tower slurry CaSO under partial working conditions 3 When the concentration exceeds the limit, the continuous operation mode of the oxidation fan is adopted, and the operation mode has large operation allowance, so that great energy waste is brought to a desulfurization system;
2. slurry CaSO with absorption tower 3 The method comprises the following steps of (1) starting and stopping an oxidation fan for real-time online concentration measurement according to requirements: individual electricityA sulfite real-time online monitoring system is installed in a desulfurization system in a factory, and sulfite real-time assay concentration is used as a judgment basis for starting and stopping a fan of an oxidation air system, but the sulfite online real-time monitoring system is expensive and large in maintenance amount, has lower cost performance compared with the income brought by starting and stopping the oxidation fan, and is generally not accepted by power plant personnel;
3. the variable frequency control mode of the oxidation fan based on the fixed excess air coefficient is as follows: the oxidation fan that the 6kV motor that has the frequency conversion drove is adopted to some power plants, and control the frequency conversion according to fixed excess air factor, in order to reach the effect that reduces oxidation wind system power consumption, this mode also has the problem that investment cost is high earlier stage, and to most power plants, its oxidation wind system fan is roots's fan, if carry out equipment transformation, investment cost will further increase, in addition, adopt the mode of fixed excess air factor to calculate the oxidation amount of wind, for guaranteeing that oxidation wind energy can satisfy the full operating mode requirement, need select great excess air factor, can cause great energy waste equally.
Disclosure of Invention
The invention aims to provide a desulfurization oxidation air system control method based on a data-driven multi-model, which solves the problem of reducing the energy consumption of an oxidation air system by designing an intelligent start-stop scheme of an oxidation fan of a thermal power plant; the technical effects that can be produced by the preferred technical scheme in the technical schemes provided by the invention are described in detail in the following.
In order to achieve the purpose, the invention provides the following technical scheme:
the invention provides a desulfurization oxidation air system control method based on a data-driven multi-model, which comprises the following steps,
step S1: screening modeling data by using historical operating data of unit and CaSO 3 Assay concentration data is used as modeling original data;
step S2: modeling, namely taking the number of running oxidation fans including the condition that all oxidation fans are stopped as a modeling division basis, taking historical running data of a unit as an input variable of a modeling process, and taking a CaSO (catalyst absorption optical absorption System) of an absorption tower 3 Assay concentrationThe degree data is used as system output, modeling is carried out by adopting a modeling mode based on an artificial neural network driven by historical data, partial least square, support vector machine, multiple linear regression and polynomial fitting, and the CaSO of the absorption tower when different oxidation fans operate is obtained 3 A concentration prediction model;
step S3: to real-time forecast obtained absorption tower CaSO 3 The concentration is calculated by adopting a sliding average method to obtain the CaSO of the absorption tower 3 Average concentration;
step S4: controlling the start and the stop of the oxidation fan, namely setting the CaSO of the absorption tower when the running number n of the oxidation fans is more than or equal to 1 3 Mean limit of concentration of
When the absorption tower CaSO obtained in the step S3 3 Average concentration exceeds
When the time is up, an oxidation fan is started, and at the same time, the current time k is set to 1 and then the timer is restarted until the time k is up>When m is reached, recalculating the absorption tower CaSO in step S3 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower CaSO obtained in the step S3 3 Average concentration of less than
Model calculation absorption tower CaSO when n oxidation fans are applied to operation 3 While averaging the concentration, synchronously adding a model in which n-1 oxidation fans operate to carry out CaSO on the absorption tower 3 Average concentration estimation, namely applying the system input from the (k-m +1) moment to the k moment and the model when the (n-1) oxidation fans operate to carry out CaSO on the absorption tower when the k oxidation fans operate 3 Estimating the average concentration, namely estimating the CaSO of the absorption tower by using two models corresponding to the working conditions of n oxidation fans and (n-1) oxidation fans 3 Average concentration is less than
While stopping one oxidationSetting current time k to 1, and timing again until k is reached>When m is reached, recalculating the absorption tower CaSO in step S3 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 And judging the start-stop operation of the oxidation fan by the average concentration.
Further, step S3 calculates and obtains the absorption tower CaSO 3 The average concentration process is that the real-time rolling prediction absorption tower CaSO is set 3 The time span of the concentration mean value is m, after the automatic control system of the oxidizing air is put into operation and m time passes, the CaSO of the absorption tower is started 3 The concentration is predicted, the current time is k, the number of running oxidation fans is n, and the CaSO of the absorption tower is at the time 3 Real-time prediction of concentration as rho n (k) Then there is k moment absorption tower CaSO 3 The average concentration was calculated as:
further, in step S4, when the number n of oxidation fans operating is 0, when the absorption tower CaSO obtained in step S3 is used 3 Average concentration exceeds
When the time is up, an oxidation fan is started, and at the same time, the current time k is set to 1 and then the timer is restarted until the time k is up>When m is reached, recalculating the absorption tower CaSO in step S3 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower CaSO obtained in the step S3 3 Average concentration of less than
The existing operating state is maintained.
Further, step S2 is implemented in the modeling process, using principal component variables obtained after principal component analysis and dimensionality reduction of unit historical operating data as input data of the modeling process, and using absorption tower CaSO 3 Assay concentration data is output as a system.
Further, the historical operation data of the unit in the step S2 includes: oxidation wind system related variables: load and FGD (flue gas desulfurization) inlet and outlet SO of unit 2 Concentration, inlet and outlet O 2 The method comprises the following steps of measuring, inlet and outlet flue gas quantity, inlet and outlet flue gas temperature, inlet and outlet flue gas pressure, inlet and outlet flue gas humidity, original flue gas dust concentration, absorption tower PH, absorption tower slurry density, absorption tower liquid level, gypsum discharge pump flow, oxidation fan inlet and outlet temperature, oxidation fan inlet and outlet pressure and oxidation fan inlet and outlet air quantity.
Further, the historical operation data of the unit in the step S1 is directly derived from DCS, and the absorption tower CaSO 3 The method comprises the steps of obtaining assay concentration data from daily assay records, screening historical operation data of a unit corresponding to assay time points and assay concentrations according to assay sampling moments, and obtaining original modeling data after the assay concentrations are combined.
Further, step S1 includes screening steady state operating condition data, including,
step 1.1: selecting the data acquired in the step S1, comprehensively considering the unit capacity and the desulfurization system arrangement mode, and selecting the minimum stable operation time threshold t s Each sampling instant and t before it are looked up s The load value of the running historical data unit in the time period is greater than L when the load fluctuation in the time period is greater than L max Deleting the group of data to obtain steady-state modeling data;
step 1.2: applying the steady-state modeling data obtained in step 1.1 to input variable data and CaSO 3 Performing Pearson correlation analysis on the concentration assay data, selecting a variable with a correlation coefficient larger than 0.2 as an input variable, and performing analysis on the correlation coefficient and CaSO 3 The assay concentrations together make up the final modeling data.
Further, the absorption tower CaSO 3 Mean limit of concentration
The invention provides a desulfurization oxidation air system control method based on a data-driven multi-model, which has the beneficial effects that:
according to the scheme, the historical data and the laboratory data stored by the unit for a long time are used as raw materials for scheme design and modeling, a modeling test is not required to be carried out on a system, and the safe operation of the unit is not influenced; according to the scheme, soft measurement of the concentration of the calcium sulfite is realized by adopting a neural network modeling mode and the like, new measuring equipment is not required to be added, the frequency conversion transformation of an original oxidizing air system is not required, only one industrial personal computer is required to be added, and the investment cost is low; the method applies long-time-span history and test data as modeling raw materials, and the modeling result can represent the wider operation condition of the unit.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Wherein:
FIG. 1 is a schematic block diagram of a modeling data screening process of the present invention;
FIG. 2 is a schematic block diagram of direct modeling of historical operational data of the unit of the present invention;
FIG. 3 is a schematic block diagram of modeling of historical operating data of the unit after principal component analysis;
FIG. 4 is a schematic block diagram of a calling mode of a unit historical operating data direct modeling model according to the present invention;
FIG. 5 is a schematic block diagram of a calling mode of a modeling model after principal component analysis of historical operating data of a unit;
FIG. 6 is a flow chart of start-stop control of the oxidation blower according to the present invention.
Detailed Description
Hereinafter, an embodiment of the data-driven multi-model-based desulfurization oxidation air system control method of the present invention will be described with reference to the accompanying drawings.
The embodiments described herein are specific embodiments of the present invention, and are intended to be illustrative and exemplary of the concepts of the present invention and should not be construed as limiting the scope of the embodiments of the present invention. In addition to the embodiments described herein, those skilled in the art will be able to employ other technical solutions which are obvious based on the disclosure of the claims and the specification of the present application, and these technical solutions include technical solutions which make any obvious replacement or modification for the embodiments described herein.
The drawings in the present specification are schematic views to assist in explaining the concept of the present invention, and schematically show the shapes of respective portions and their mutual relationships. It is noted that the drawings are not necessarily to the same scale, emphasis instead being placed upon clearly illustrating the structure of various elements of embodiments of the present invention. Like reference numerals are used to denote like parts.
Example 1:
according to the method, when the number of operating different oxidation fans is respectively established, the related input of an oxidation air system and system absorption tower slurry CaSO 3 Model of relationship between concentrations, to the absorption column CaSO 3 The concentration is predicated, and the oxidation fan is started and stopped as required according to the predicated concentration, so that the intelligent dynamic start and stop of the oxidation fan are realized, and the purpose of greatly reducing the energy consumption of an oxidation air system is achieved, and the method comprises the following steps:
1. modeling data screening:
(1) the historical operation and test data of the unit are used as original modeling data, the historical operation data of the oxidation air system can be directly derived from DCS, and the absorption tower CaSO 3 The concentration data can be obtained from daily chemical examination records, then the historical operating data of the oxidation air system corresponding to the chemical examination time point and the chemical examination concentration is screened according to the chemical examination sampling time, and the original modeling data is obtained after the chemical examination concentration is combined.
(2) In order to avoid the influence of the transition process of the system dynamic response on the modeling precision, the steady-state working condition data needs to be further screened, and the screening mode is as follows: selecting the data obtained in the step (1), comprehensively considering the unit capacity and the desulfurization system arrangement mode, and selecting the minimum stable operation time threshold t s Each sampling instant and t before it are looked up s TimeThe load value of the operation historical data unit of the section is calculated, and when the load fluctuation in the time section is more than L max Deleting the group of data to obtain final modeling data; the screening process is shown in FIG. 1.
2. Modeling:
the number of the running oxidation fans including the condition that all the oxidation fans are stopped is used as a modeling division basis, and relevant variables of an oxidation air system are as follows: load and FGD (flue gas desulfurization) inlet and outlet SO of unit 2 Concentration, inlet and outlet O 2 The method comprises the steps of obtaining data such as quantity, inlet and outlet flue gas temperature, inlet and outlet flue gas pressure, inlet and outlet flue gas humidity, original flue gas dust concentration, absorption tower PH, absorption tower slurry density, absorption tower liquid level, gypsum discharge pump flow, oxidation fan inlet and outlet temperature, oxidation fan inlet and outlet pressure, oxidation fan inlet and outlet air quantity and the like, or using principal component variables obtained after subjecting the data to principal component analysis and dimensionality reduction as modeling process input data, and using CaSO of the absorption tower 3 And (3) taking the assay concentration data as system output, and adopting a historical data-driven modeling mode, such as: modeling by adopting data driving modeling modes such as artificial neural network, partial least square, support vector machine, multiple linear regression, polynomial fitting and the like to obtain CaSO of the absorption tower when different oxidation fans operate 3 A concentration model; the modeling process is shown in fig. 2 and 3.
3. Controlling the start and the stop of the oxidation fan:
by unit load, FGD inlet and outlet SO 2 Concentration, inlet and outlet O 2 The data such as the amount, the inlet and outlet flue gas temperature, the inlet and outlet flue gas pressure, the inlet and outlet flue gas humidity, the original flue gas dust concentration, the PH of the absorption tower, the slurry density of the absorption tower, the liquid level of the absorption tower, the gypsum discharge pump flow, the temperature of the inlet and outlet of the oxidation fan, the pressure of the inlet and outlet of the oxidation fan, the air quantity of the inlet and outlet of the oxidation fan and the like, or the principal component variables formed by weighting and combining the data are used as input data, and the CaSO of the absorption tower is predicted by applying the model built in the 2 3 The model calling method is shown in fig. 4 and 5.
And taking the predicted concentration as a judgment basis for increasing or decreasing the number of the running oxidation fans, wherein the judgment process is as follows:
(1) calculating CaSO of absorption tower 3 Average concentration:
to real-time forecast obtained absorption tower CaSO 3 The concentration adopts a sliding average mode to obtain CaSO of the absorption tower 3 Average concentration, real-time rolling prediction of absorption tower CaSO 3 The time span of the concentration mean value is m, after the automatic control system of the oxidizing air is put into operation and m time passes, the CaSO of the absorption tower is started 3 The concentration is predicted, the current time is k, the number of running oxidation fans is n, and the CaSO of the absorption tower is at the time 3 Real-time prediction of concentration as rho n (k) Then there is k moment absorption tower CaSO 3 The average concentration was calculated as:
(2) oxidation fan operation number control
1) When the number n of the operating oxidation fans is more than or equal to 1, the CaSO of the absorption tower 3 Average concentration limit of
General selection
When the absorption tower obtained in 1) is CaSO 3 Average concentration exceeds
When the time is up, an oxidation fan is started, and at the same time, the current time k is set to 1 and then the timer is restarted until the time k is up>When m is reached, recalculating the absorption tower CaSO according to 1) 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower obtained in 1) is CaSO 3 Average concentration of less than
Model calculation absorption tower CaSO when n oxidation fans are applied to operation 3 While the average concentration is added, the operation of n-1 oxidation fans is synchronously appliedCaSO of absorption tower by model 3 Estimating average concentration, namely applying the system input from the time (k-m +1) to the time k and the model when (n-1) oxidation fans operate to perform CaSO (absorbed SO) of the absorption tower when the oxidation fans operate at the time k 3 Estimating the average concentration, namely estimating the CaSO of the absorption tower by using two models corresponding to the working conditions of n oxidation fans and (n-1) oxidation fans 3 Average concentration is less than
When the current time k is set to 1, the time is counted again until k is reached>When m is reached, recalculating the absorption tower CaSO according to 1) 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 And judging the start-stop operation of the oxidation fan by the average concentration.
2) When the number n of the oxidation fans is equal to 0, the absorption tower CaSO obtained in the step 1) is used 3 Average concentration exceeds
When the current time k is set to be 1, the time is counted again until k is reached>When m is reached, recalculating the absorption tower CaSO according to 1) 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower obtained in 1) is CaSO 3 Average concentration of less than
Keeping the existing running state; the flow chart of the start-stop control of the oxidation fan is shown in fig. 6.
Example 2:
the method for controlling the oxidizing air system based on the data model and the mechanism operation can be further improved as follows, when the steady-state working condition is further screened from the modeling data, the stable running time of the load working condition before and after the sampling time is not taken as a steady-state judgment basis, and the stable running time of variables such as the flue gas flow, the FGD inlet sulfur content or the flue gas flow multiplied by the FGD inlet sulfur content before and after the sampling time is taken as a steady-state judgment basis; the method establishes the related variables of the absorption towerCaSO in absorption tower slurry 3 Model between concentrations, due to elimination of assay absorption tower CaSO in the actual assay process 3 In addition to the concentration, the CaSO in the dehydrated gypsum can be treated 3 The concentration is tested, so that the related variables of the absorption tower and CaSO in the dehydrated gypsum can be established 3 And (4) carrying out approximate replacement on the original model by using the model between concentrations.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (8)
1. A desulfurization oxidation air system control method based on data-driven multiple models is characterized by comprising the following steps,
step S1: screening modeling data by using historical operating data of unit and CaSO 3 Assay concentration data is used as modeling original data;
step S2: modeling, namely taking the number of running oxidation fans including the condition that all oxidation fans are stopped as a modeling division basis, taking historical running data of a unit as an input variable of a modeling process, and taking a CaSO (catalyst absorption optical absorption System) of an absorption tower 3 The test concentration data is used as system output, modeling is carried out by adopting a modeling mode based on an artificial neural network driven by historical data, partial least square, support vector machine, multiple linear regression and polynomial fitting, and the CaSO of the absorption tower when different oxidation fans operate is obtained 3 A concentration prediction model;
step S3: for absorption tower CaSO obtained by real-time prediction 3 The concentration is calculated by adopting a sliding average method to obtain the CaSO of the absorption tower 3 Average concentration, real-time rolling prediction of absorption tower CaSO 3 The concentration mean time span is m;
step S4: controlling the start and the stop of the oxidation fan, namely setting the CaSO of the absorption tower when the running number n of the oxidation fans is more than or equal to 1 3 Mean limit of concentration of
When the absorption tower CaSO obtained in the step S3 3 Average concentration exceeds
When the time is up, an oxidation fan is started, and at the same time, the current time k is set to 1 and then the timer is restarted until the time k is up>When m is reached, recalculating absorption tower CaSO in step S3 3 Average concentration, again according to the obtained CaSO of the absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower CaSO obtained in the step S3 3 Average concentration of less than
Model calculation absorption tower CaSO when n oxidation fans are applied to operation 3 Synchronously adding a model for carrying out the operation of n-1 oxidation fans to the absorption tower CaSO while averaging the concentration 3 Average concentration estimation, namely applying the system input from the (k-m +1) moment to the k moment and the model when the (n-1) oxidation fans operate to carry out CaSO on the absorption tower when the k oxidation fans operate 3 Estimating average concentration, and estimating CaSO of the absorption tower by using two models corresponding to the working conditions of n oxidation fans and (n-1) oxidation fans 3 Average concentration is less than
Stopping an oxidation fan, setting the current time k to 1, timing again, and waiting for k>When m is reached, recalculating absorption tower CaSO in step S3 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 And judging the start-stop operation of the oxidation fan by the average concentration.
2. The method for controlling the desulfurization oxidation air system based on the data-driven multi-model as claimed in claim 1, wherein the step S3 is implemented by calculating and obtaining CaSO of the absorption tower 3 The average concentration process is that the real-time rolling prediction is set to absorb the CaSO of the tower 3 The time span of the concentration mean value is m, and the oxidation wind is automaticAfter the control system is put into operation and m time later, the absorption tower CaSO is started to be carried out 3 The concentration is predicted, the present time is set as k, the number of the operating oxidation fans is set as n, and the CaSO of the absorption tower is set at the time 3 Real-time prediction of concentration as rho n (k) Then there is k moment absorption tower CaSO 3 The average concentration was calculated as:
3. the method as claimed in claim 1, wherein in step S4, when the number n of the oxidation fans is 0, the absorption tower CaSO obtained in step S3 is used as the absorption tower CaSO 3 Average concentration exceeds
When the current time k is set to be 1, the time is counted again until k is reached>When m is reached, recalculating the absorption tower CaSO in step S3 3 Average concentration, again according to the CaSO of the obtained absorption tower 3 Judging the start-stop operation of the oxidation fan according to the average concentration; when the absorption tower CaSO obtained in the step S3 3 Average concentration of less than
The existing operating state is maintained.
4. The method for controlling the desulfurization and oxidation air system based on the data-driven multiple models as claimed in claim 1, wherein step S2 is implemented in a modeling process, a principal component variable obtained after principal component analysis and dimensionality reduction are implemented on historical unit operation data is used as input data of the modeling process, and an absorption tower CaSO is used as input data of the modeling process 3 Assay concentration data is output as a system.
5. The data-driven multi-model-based desulfurization oxidation air system control method according to claim 1The method is characterized in that the historical operation data of the unit in the step S2 includes: oxidation wind system related variables: load and FGD (flue gas desulfurization) inlet and outlet SO of unit 2 Concentration, inlet and outlet O 2 The device comprises the following components of quantity, inlet and outlet flue gas temperature, inlet and outlet flue gas pressure, inlet and outlet flue gas humidity, original flue gas dust concentration, absorption tower PH, absorption tower slurry density, absorption tower liquid level, gypsum discharge pump flow, oxidation fan inlet and outlet temperature, oxidation fan inlet and outlet pressure and oxidation fan inlet and outlet air quantity.
6. The method for controlling a desulfurization oxidation air system based on data driving multiple models according to claim 1, wherein the historical operating data of the unit in the step S1 is directly derived from DCS, and CaSO of the absorption tower 3 The method comprises the steps of obtaining assay concentration data from daily assay records, screening historical operation data of a unit corresponding to assay time points and assay concentrations according to assay sampling moments, and obtaining original modeling data after the assay concentrations are combined.
7. The method for controlling a desulfurization oxidizing air system based on data-driven multiple models according to claim 1, wherein the step S1 further comprises screening steady-state condition data including,
step 1.1: selecting the data obtained in the step S1, comprehensively considering the unit capacity and the desulfurization system arrangement mode, and selecting the minimum stable operation time threshold t s Each sampling instant and t before it are looked up s The load value of the running historical data unit in the time period is greater than L when the load fluctuation in the time period is greater than L max The set of data is deleted and the steady state modeling data, L, is obtained max Is the maximum constant value of the load fluctuation;
step 1.2: applying the steady-state modeling data obtained in step 1.1 to input variable data and CaSO 3 Performing Pearson correlation analysis on the concentration assay data, selecting a variable with a correlation coefficient larger than 0.2 as an input variable, and performing analysis on the correlation coefficient and CaSO 3 The assay concentrations together make up the final modeling data.
8. The data-driven multi-model-based desulfurization and oxidation air system control method according to claim 1, wherein the absorption tower CaSO 3 Mean limit of concentration
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010828597.4A CN112044270B (en) | 2020-08-18 | 2020-08-18 | Desulfurization oxidation air system control method based on data-driven multiple models |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010828597.4A CN112044270B (en) | 2020-08-18 | 2020-08-18 | Desulfurization oxidation air system control method based on data-driven multiple models |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112044270A CN112044270A (en) | 2020-12-08 |
| CN112044270B true CN112044270B (en) | 2022-09-02 |
Family
ID=73600697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010828597.4A Active CN112044270B (en) | 2020-08-18 | 2020-08-18 | Desulfurization oxidation air system control method based on data-driven multiple models |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112044270B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113082954B (en) * | 2021-04-07 | 2022-04-01 | 浙江大学 | Whole-process intelligent operation regulation and control system of wet desulphurization device |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5283737B2 (en) * | 2011-09-08 | 2013-09-04 | 中国電力株式会社 | Gas mixing ventilator operation information providing device and method |
| CN105138718B (en) * | 2015-07-10 | 2018-09-28 | 广东电网有限责任公司电力科学研究院 | A kind of projectional technique and its subsidiary engine load adjustment method of combination UDF Novel fires desulfurizing tower of electric power plant desulfuration efficiency |
| CN105467954B (en) * | 2015-11-20 | 2018-09-18 | 华北电力大学（保定） | A kind of method and device of desulfurization method of limestone-gypsum ORP and pH double controls |
| CN108317100A (en) * | 2018-04-04 | 2018-07-24 | 华润电力（沧州运东）有限公司 | A kind of air compressor control method, equipment and system |
| CN208229629U (en) * | 2018-04-10 | 2018-12-14 | 安徽科力德电力科技有限公司 | A kind of wet flue gas desulfurization oxidation energy-saving control device of blower |
| CN109085856A (en) * | 2018-09-10 | 2018-12-25 | 厦门邑通软件科技有限公司 | A kind of desulfuration absorbing tower oxidation fan consumption-reducing method and system |
| CN109636115B (en) * | 2018-11-13 | 2022-03-15 | 国能龙源环保有限公司 | Absorption tower slurry calcium sulfite concentration fitting method and system based on big data |
-
2020
- 2020-08-18 CN CN202010828597.4A patent/CN112044270B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112044270A (en) | 2020-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112506162B (en) | Oxidation air system control method based on data model and mechanism operation | |
| CN111340199B (en) | Material balance and deep learning-based energy-saving method for desulfurization system | |
| CN101477367A (en) | Energy-saving consumption reduction oriented process efficiency optimization control method and use thereof | |
| CN111897373B (en) | Model prediction-based ammonia injection flow adjusting method for SCR denitration device | |
| CN105498497A (en) | Flue gas desulfurization and denitration integrated equipment controlled through multiple variables and control method thereof | |
| CN112465239B (en) | Desulfurization system operation optimization method based on improved PSO-FCM algorithm | |
| CN110992209A (en) | Flow prediction method | |
| CN112044270B (en) | Desulfurization oxidation air system control method based on data-driven multiple models | |
| CN111598308B (en) | Method for solving combination optimization of slurry circulating pump based on regression and double PSO algorithm | |
| CN113530793B (en) | Intelligent adjusting system for air compression station | |
| CN112364562A (en) | Cooperative control method and system for flue gas environmental protection island | |
| CN115309603A (en) | Data center energy consumption prediction optimization method, system, medium and computing device | |
| CN113112060A (en) | Lifting pump station scheduling method and device based on soft flow measurement and computer equipment | |
| CN113440990A (en) | EMD-LSTM based outlet SO2Concentration prediction method | |
| CN112742187A (en) | Method and device for controlling pH value in desulfurization system | |
| US7155367B1 (en) | Method for evaluating relative efficiency of equipment | |
| KR20060092660A (en) | Automatic control device and method for wastewater treatment using fuzzy control | |
| CN112023650A (en) | Full-working-condition self-adaptive pH adjusting system | |
| CN112415148B (en) | Wet flue gas desulfurization system CaSO based on online learning 3 Soft measurement method | |
| CN113468813B (en) | Desulfurization system inlet SO2Concentration prediction method and device and electronic equipment | |
| CN114444391A (en) | Oxidation fan operation optimization algorithm based on particle swarm optimization | |
| CN114880944A (en) | Desulfurization system operation optimization method and system based on Lasso algorithm | |
| CN111111392B (en) | Online automatic control method for SO2 concentration during blowback of CEMS table of power plant | |
| CN115869745B (en) | Desulfurization treatment method and device | |
| CN113671830B (en) | Cold end optimization closed-loop control method for thermal power generating unit based on intelligent scoring |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100039 room 901, 9 / F, building 1, yard 16, West Fourth Ring Middle Road, Haidian District, Beijing Applicant after: Guoneng Longyuan environmental protection Co.,Ltd. Address before: 100039 room 911, 1 building, 16 West Fourth Ring Road, Haidian District, Beijing. Applicant before: BEIJING GUODIAN LONGYUAN ENVIRONMENTAL ENGINEERING Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221025 Address after: 100039 room 901, 9 / F, building 1, yard 16, West Fourth Ring Middle Road, Haidian District, Beijing Patentee after: Guoneng Longyuan environmental protection Co.,Ltd. Patentee after: Guoneng (Shandong) energy environment Co.,Ltd. Address before: 100039 room 901, 9 / F, building 1, yard 16, West Fourth Ring Middle Road, Haidian District, Beijing Patentee before: Guoneng Longyuan environmental protection Co.,Ltd. |