CN116543854A - Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method - Google Patents
Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method Download PDFInfo
- Publication number
- CN116543854A CN116543854A CN202310362815.3A CN202310362815A CN116543854A CN 116543854 A CN116543854 A CN 116543854A CN 202310362815 A CN202310362815 A CN 202310362815A CN 116543854 A CN116543854 A CN 116543854A
- Authority
- CN
- China
- Prior art keywords
- concentration
- urea
- control unit
- denitration
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- XSQUKJJJFZCRTK-UHFFFAOYSA-N Urea Chemical compound NC(N)=O XSQUKJJJFZCRTK-UHFFFAOYSA-N 0.000 claims abstract description 50
- 239000004202 carbamide Substances 0.000 claims abstract description 50
- 238000013528 artificial neural network Methods 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 4
- 239000007921 spray Substances 0.000 claims description 31
- 239000000243 solution Substances 0.000 claims description 27
- 238000003860 storage Methods 0.000 claims description 21
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 claims description 20
- 239000003546 flue gas Substances 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 13
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 13
- 239000011259 mixed solution Substances 0.000 claims description 11
- 238000002156 mixing Methods 0.000 claims description 11
- 239000007788 liquid Substances 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 239000003638 chemical reducing agent Substances 0.000 claims description 8
- 239000007789 gas Substances 0.000 claims description 7
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 6
- 239000001301 oxygen Substances 0.000 claims description 6
- 229910052760 oxygen Inorganic materials 0.000 claims description 6
- 239000012895 dilution Substances 0.000 claims description 5
- 238000010790 dilution Methods 0.000 claims description 5
- 239000000446 fuel Substances 0.000 claims description 5
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000007865 diluting Methods 0.000 claims description 2
- 229910052757 nitrogen Inorganic materials 0.000 claims description 2
- 238000002360 preparation method Methods 0.000 claims description 2
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 8
- 230000001276 controlling effect Effects 0.000 description 5
- 230000033228 biological regulation Effects 0.000 description 4
- 229910021529 ammonia Inorganic materials 0.000 description 3
- 238000002485 combustion reaction Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005338 heat storage Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005507 spraying Methods 0.000 description 2
- 229910002651 NO3 Inorganic materials 0.000 description 1
- NHNBFGGVMKEFGY-UHFFFAOYSA-N Nitrate Chemical compound [O-][N+]([O-])=O NHNBFGGVMKEFGY-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000005243 fluidization Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000001802 infusion Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C10/00—Fluidised bed combustion apparatus
- F23C10/18—Details; Accessories
- F23C10/28—Control devices specially adapted for fluidised bed, combustion apparatus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mechanical Engineering (AREA)
- Combustion & Propulsion (AREA)
- Crystallography & Structural Chemistry (AREA)
- Bioinformatics & Computational Biology (AREA)
- Treating Waste Gases (AREA)
Abstract
Digital twin system NO of circulating fluidized bed boiler x An emission prediction and control method comprising: relevant real-time parameters of data acquired from a boiler DCS system and entering into an SIS system database are utilized, and a NO-based algorithm is built through a BP neural network nonlinear algorithm x The concentration prediction digital processing platform takes real-time parameters as input values to obtain prediction results and compares the prediction results with historical experience values under the same working condition, and performs feedback correction on an algorithm to improve prediction accuracy; constructing a DCS control-SIS combined system to communicate with a digital twin system; constructing a denitration automatic prediction control system, and predicting NO given by a platform by utilizing a digital twin system x Concentration and concentration ofThe urea solution flow solves the simplified control model, and transmits a control model signal to the control unit to perform predictive control of the denitration system, so that the urea solution flow can be accurately controlled, the urea consumption can be reduced, and the requirement of ultra-low emission can be met.
Description
Technical Field
The utility model relates to a digital twin system NO of a circulating fluidized bed boiler x The field of emission prediction and control, in particular to a digital twin system NO built by a circulating fluidized bed generator set x An emission concentration prediction method.
Background
It is known that the circulating fluidized bed generator set in the novel electric power system bears the special working condition of near zero depth peak regulation, and utilizes the characteristic of large heat accumulation of the circulating fluidized bed generator set to cut off the fuel and the air smoke system in a certain period of time, and the original SNCR system is also utilized under the condition that the steam turbine is not switched off and the generator is not disconnected, so that the novel electric power system is also an operation working condition. The circulating fluidized bed boiler has the advantages of good coal adaptability, wide load adjusting range, strong continuous operation capability, strong heat storage capability, stable combustion under ultralow load and the like. Under normal operating conditions, the original NO x The production amount is low. SNCR is a common denitration method for power plants, mainly comprising the steps of spraying urea solution at the inlet of a cyclone separator, heating by using high-temperature flue gas to generate ammonia gas, uniformly mixing the ammonia gas and the flue gas as far as possible, and performing chemical reaction to generate N 2 And H 2 O, thereby reducing NO in flue gas x Is contained in the composition. In order to respond to the low-carbon emission reduction call, the electric power system starts to transition to a novel electric power system mainly comprising new energy. However, the new energy has fluctuation and intermittence, and after being integrated into the power grid on a large scale, the stability of the power grid can be obviously affected, and the thermal power is used as a power supply with the best regulation performance, so that deep peak shaving is needed, and the flexibility of the unit is improved. In the peak regulation process, the advantage of strong heat storage capacity of the circulating fluidized bed boiler can be utilized to develop near zero depth peak regulation, and the operation can be inevitably performed under the condition of rapid depth load change in the operation process. When the circulating fluidized bed boiler runs under low load and even ultra-low load, the requirements of boiler fluidization are met, unbalance of the primary air quantity, the secondary air quantity and the fuel ratio occurs, the oxygen content of tail flue gas is increased, and NO is promoted x Generates and the temperature of the hearth is lower under the low-load operation condition, thereby reducing NO represented by urea x The reduction efficiency of the reducing agent causes a decrease in denitration efficiency and a decrease in ammonia utilization rate. If the control index is difficult to meet by the existing control system, namely, 50mg/Nm is not met 3 Or cause a great waste of reducing agents and even accelerate the corrosion rate of the thermodynamic equipment. For this reason, some circulating fluidized bed coal-fired power plants in China already adopt an automatic control mode to participate in the strippingAnd (5) a nitrate process. The utility model of the patent CN217795433U discloses automatic flue gas denitration equipment, and the denitration efficiency is improved by uniformly conveying gas; the patent CN113433911A discloses an ammonia spraying accurate control system of a denitration device based on accurate concentration prediction; patent 1055975537B proposes a denitration control method based on predictive control technique, according to NO of a denitration reactor in each flue region x And comparing the concentration deviation with a preset value, and further judging the denitration dosage. The above patent generally uses automation, high denitration efficiency and conventional load as starting points to perform denitration control, and fails to meet the use efficiency of the denitration agent by near zero depth peak shaving and NO in the case of extremely wide load change x The emission concentration meets ultra-low emission requirements. Therefore, a new method is needed to solve the difficult problem that denitration control is difficult to effectively perform by using manual history experience under the wide load working condition.
With the rapid development of artificial intelligence, the actual industrial production technology is continuously innovated, and particularly, a digital twin system and a neural network are applied, so that the traditional industry of producing a large amount of operation data every day in a power plant brings the possibility of deep learning network modeling to the digital twin system. On the basis of analyzing the original PID control, we propose a digital twin denitration NO of a 300MW circulating fluidized bed boiler x Emission concentration prediction and control methods.
Disclosure of Invention
The utility model aims to provide a method for predicting and displaying and transmitting NOx emission concentration in real time before denitration of a circulating fluidized bed boiler based on a digital twin system.
Another object of the present utility model is to induce NO for fuel properties, secondary air volume and temperature under variable load conditions x The problem of difficult treatment of the concentration increase is that a human intervention or analog superposition mode is provided, namely, a digital twin system is constructed to NO x The method for pre-judging the concentration is an effective method before the DCS is not intelligent.
The utility model further aims to solve the denitration problem of the 3U circulating fluidized bed digital twin system based on Ultra low load (near zero) operation (Ultra-low load operation) -Ultra-long waiting (Ultra-long duration) -Ultra-fast recovery (Ultra-construction), solve the variable load denitration control problem and improve the control quality of the power plant denitration process.
In order to achieve the above object, in one aspect, the present utility model provides a method for predicting NOx emission concentration of a circulating fluidized bed boiler based on a digital twin system, comprising the steps of:
firstly, acquiring relevant parameters including fuel addition amount, unit load, hearth temperature, oxygen concentration, primary air quantity, secondary air quantity, urea solution concentration and SNCR reaction zone flue gas temperature from a boiler DCS control system in real time by utilizing an SIS system database, and embedding a predicted NO in a digital processing platform based on the acquired real-time relevant parameters x A BP neural network algorithm of concentration forms a digital processing prediction system;
step two, taking the relevant parameters acquired in real time as input values of an input layer of the BP neural network, and performing forward calculation of numerical values, namely transmitting the input parameters to an implicit layer and then transmitting the input parameters to an output layer by layer through an activation function;
step three, the back propagation of the error, the output result after the hidden layer calculation is obtained from the output layer, and the output result is compared with the history NO under the same working condition x Performing error calculation on the concentration experience value, reversely deriving an activation function, updating the connection weight and the threshold value of the neural network layer by layer, completing training of the network under the set iteration times, correcting future prediction trend, performing feedback correction, and finally realizing accurate prediction;
furthermore, the BP neural network can be used for denitrating NO before the denitration in the hearth under the working condition of wide load change based on a nonlinear algorithm x Accurately predicting the concentration;
in another aspect, the present utility model provides a reducing agent flow control method, comprising:
constructing a solution system of the flow of the urea solution after dilution and mixing, namely the flow of a variable frequency pump, in a digital platform based on a reducing agent consumption solution algorithm;
the formula of the reducing agent consumption solving algorithm is as follows:
wherein w is n The consumption of urea is kg/h; ρ is the molar concentration of NO, mg/Nm 3 ；
v q For the flow of flue gas at the reactor inlet, i.e. dry flue gas at actual oxygen content, nm 3 /h；
NSRN is the equivalent molar ratio of urea to nitrogen.
Further, the concentration of the mixed solution after urea dilution is set as mu,%;
according to the urea consumption w n The calculation formula for obtaining the average flow of the variable frequency pump is as follows:
f n the flow rate is kg/h of the variable frequency pump;
furthermore, all the algorithms can be used without changing field processes and equipment, and the corresponding data platform operates in a green operation mode;
furthermore, all the algorithms can be compatible with any windows platform, and the hardware investment is not required to be increased.
The real-time data of the control parameters are obtained from the boiler DCS control system, and the frequency is 0.2Hz, namely 5s each time.
In a third aspect, to achieve the above object, the present utility model provides a method for constructing an automated denitration system, including:
step four, establishing NO x The control unit is automatically removed and is connected with the NOx concentration prediction digital processing platform;
further, NO predicted using a digitizing platform x The concentration is judged in a control unit in a concentration range, and the control unit is used for realizing the matching with the working quantity of the spray heads on the cyclone separator;
the concentration range is as follows:
ρ≤50mg/Nm 3 ；
50<ρ≤100mg/Nm 3 ；
100<ρ≤200mg/Nm 3 ；
ρ>200mg/Nm 3 ；
wherein ρ is predicted NO x Concentration due to NO emitted during boiler combustion x Mainly NO, at about 90% of the total, thus predicted NO in this patent x The concentration approximation is processed to be the concentration of NO;
the working number range of the spray heads on the cyclone separator is as follows:
2 spray heads work;
6 spray heads work;
10 spray heads work;
the 12 spray heads work;
14 spray heads are mounted on the cyclone separator in total, wherein 2 spray heads are reserved for standby;
controlling the flow of the variable frequency pump and the starting of the urea solution storage and distribution system by using the urea solution flow solving result in the digital processing platform in the control unit;
further, a urea preparation and storage combined integral module is established, which comprises a urea solution storage tank with higher concentration and is connected with a control unit;
a water storage tank is arranged above the liquid storage tank and is connected with the control unit;
a urea solution diluting and mixing tank is arranged and is respectively connected with the liquid storage tank and the water storage tank;
a variable frequency pump for outputting mixed liquid is arranged at the outlet of the mixing tank, and the variable frequency pump is connected with the spray head in the fourth step to control the total flow of the spray head;
step six, establishing NO after flue gas denitration x A concentration monitoring system;
the NO x The concentration monitoring system is characterized in that an NO is arranged at the tail outlet of the flue x A gas concentration sensor;
the concentration sensor is connected with the control unit, and the control unit judges and controls whether the denitration process is operated or not according to signals acquired by the sensor;
if rho is denitrated<50mg/Nm 3 The variable frequency pump is stopped by the control unit, the electromagnetic valve at the front end of the spray head is closed, and the denitration process is finished;
if rho is denitrated>50mg/Nm 3 The reserved spray head is started to work, and the control unit is used for controlling the variable frequency pump to increase the overall flow of the spray head;
when the automatic denitration system operates, the control unit firstly predicts NO according to accuracy x The concentration of the urea solution is controlled by controlling the liquid storage tank and the water storage tank to respectively convey a specified amount of high-concentration urea solution and water to the mixing tank, and the urea solution and the water are mixed in the mixing tank to form diluted urea mixed solution;
the control unit then generates a signal according to NO x The concentration prediction digital processing platform controls the variable frequency pump to quantitatively output the diluted urea mixed solution and mix the diluted urea mixed solution with the flue gas according to instructions of the diluted urea solution flow solving result, and NO in the flue gas are removed through the reaction of the urea mixed solution and the flue gas 2 Finally, the denitration of the circulating fluidized bed boiler is realized.
The beneficial effects are that:
the algorithm can be used without changing the field process and equipment, and the corresponding data platform operates in a green operation mode; all algorithms of the utility model can be compatible with any windows platform without increasing hardware investment. The method can effectively control the urea usage amount and save the enterprise cost by utilizing the scientific research of the predictive digital platform to predict the urea injection amount. Meanwhile, the ammonia escape amount after the denitration reaction is reduced, and a better removal effect is achieved.
Drawings
FIG. 1 is a digital twin system NO x Emission prediction and control method schematic diagram, wherein part (1) in the diagram is NO x A concentration prediction digital processing platform; in the figure, (2) is a part of a denitration automatic control system.
Fig. 2 is a flowchart of the BP neural network algorithm.
Detailed Description
The digital twin of the circulating fluidized bed boiler is described below with reference to the accompanying drawingsSystem NO x Specific embodiments of the prediction and control methods. As shown in part (1) of fig. 1, the BP neural network in the digital processing platform mainly consists of three layers, including an input layer of the BP neural network, an hidden layer of the BP neural network and an output layer of the BP neural network.
The BP neural network implementation principle is that the deviation between the predicted output and the actual output is calculated, then the deviation is reversely derived to update the connection weight and the threshold value of the neural network, and the connection weight and the threshold value are analyzed and compared with historical experience data, so that the model structure of the neural network is continuously optimized and adjusted to improve the fitting effect, and the concentration of NOx before denitration is accurately predicted.
The real-time related parameters of various information from the outside such as the patent are transmitted as the input layer of the BP neural network to enter the hidden layer thereof for network operation processing, and the final processing result is obtained through the output of the output layer. When the error between the output result of the output layer of the BP neural network and the preset input value is large, the back propagation stage of the BP neural network is entered, and the updating of the network weight is carried out until the error between the output result and the expected result meets a certain condition.
The circulating fluidized bed digital twin system comprises a NOx concentration prediction digital processing platform, a control unit and an automatic denitration system. The boiler DCS control-SIS combined system and the digital processing platform are communicated through PROFINET industrial Ethernet, PROFIBUS-DP, TCP/IP and MPI, and relative independence exists among the systems, so that the digital processing system fails, and the denitration process is not affected.
Part (2) of fig. 1, which is a denitration automation control system;
when the denitration system works, the opening of the water outlet regulating valve of the water storage tank is controlled by the control unit, meanwhile, the control unit can control the working frequency of the infusion pump at the outlet of the high-concentration urea liquid storage tank, and the urea dilution mixed solution with given concentration is accurately proportioned by controlling the amounts of water and the high-concentration urea solution.
The digital processing platform collects the combustion from the SIS system databaseThe real-time data of the material feeding amount, the unit load, the hearth temperature, the oxygen concentration, the primary air quantity, the secondary air quantity, the urea solution concentration and the SNCR reaction zone flue gas temperature. Then predicting NO by using BP neural network prediction system in the digital processing platform through the data x The concentration, the concentration information is transmitted to the control unit, the digital processing platform calculates the total flow of the variable frequency pump according to the concentration, and the control unit sets NO according to the set concentration x The concentration range and the target flow of the variable frequency pump are used for controlling the variable frequency pump and the spray head to work.
When the predicted concentration rho is less than or equal to 50mg/Nm 3 When the cyclone separator is in a closed state, the control unit opens 2 spray heads on the cyclone separator according to the preset condition;
when the concentration is predicted to be 50<ρ≤100mg/Nm 3 The control unit opens 6 spray heads on the cyclone separator;
when the predicted concentration is 100<ρ≤200mg/Nm 3 The control unit increases the working quantity of the spray heads on the cyclone separator to 10;
when predicting the concentration ρ>200mg/Nm 3 The number of heads was increased to 12.
The flow of the variable frequency pump is calculated by the digital processing platform, then the target flow is transmitted to the control unit, and the control unit controls the flow by adjusting the frequency of the variable frequency pump.
The gas concentration sensor arranged on the tail flue can monitor the concentration of NOx after denitration in real time, the gas concentration sensor transmits an electric signal to the control unit, the control unit controls the flow of the variable frequency pump and the working number of the spray heads according to the fed back concentration, and when the concentration rho displayed by the gas sensor is more than or equal to 50mg/Nm 3 When the control unit opens the reserved 2 spray heads; when the concentration rho of the gas sensor is less than or equal to 50mg/Nm 3 When the control unit controls the variable frequency pump, the spray head and the urea storage and distribution system to stop working, the denitration process is temporarily ended until the control unit collects the NOx concentration rho from the sensor to be more than or equal to 50mg/Nm 3 Repeating the above steps.
The frequency conversion pump frequency and the number of the work of the spray heads on the cyclone separator are controlled by the control unit, so that denitration can be reasonably and efficiently performed.
As shown in fig. 2, the chart is a BP neural network algorithm flow chart, and the forward propagation and the backward propagation of the signals in the BP neural network are repeatedly alternated until the error output by the network can reach the accuracy requirement or the preset training times. After the network training is completed, the BP neural network can process similar input signals by itself. The data is re-input into the trained neural network, so that accurate prediction results can be obtained, and NO can be performed by utilizing the principle x And (5) concentration prediction.
The examples described herein are merely illustrative of the preferred embodiments of the present utility model and are not intended to limit the utility model; various modifications and improvements of the technical scheme of the utility model will fall within the protection scope of the utility model.
Claims (10)
1. Circulating fluidized bed boiler NO based on digital twin system x The emission concentration prediction method is characterized in that: the method comprises the following steps:
firstly, acquiring relevant parameters including fuel addition amount, unit load, hearth temperature, oxygen concentration, primary air quantity, secondary air quantity, urea solution concentration and SNCR reaction zone flue gas temperature from a boiler DCS control system in real time by utilizing an SIS system database, and embedding a predicted NO in a digital processing platform based on the acquired real-time relevant parameters x A BP neural network algorithm of concentration forms a digital processing prediction system;
step two, taking the relevant parameters acquired in real time as input values of an input layer of the BP neural network, and performing forward calculation of numerical values, namely transmitting the input parameters to an implicit layer and then transmitting the input parameters to an output layer by layer through an activation function;
step three, the back propagation of the error, the output result after the hidden layer calculation is obtained from the output layer, and the output result is compared with the history NO under the same working condition x Error calculation is carried out on the concentration experience value, the connection weight and the threshold value of the neural network are updated layer by reversely deriving the activation function, the training of the network is completed under the set iteration times, and the correction is carried outAnd predicting trend in the future, carrying out feedback correction, and finally realizing accurate prediction.
2. The method according to claim 1, wherein: step one, the BP neural network is based on a nonlinear algorithm, and NO before denitration in a hearth is performed under a wide load change working condition x And (5) accurately predicting the concentration.
3. A reducing agent flow control method based on claim 1, characterized by: comprising the following steps:
constructing a solution system of the flow of the urea solution after dilution and mixing, namely the flow of a variable frequency pump, in a digital platform based on a reducing agent consumption solution algorithm;
the formula of the reducing agent consumption solving algorithm is as follows:
wherein w is n The consumption of urea is kg/h; ρ is the molar concentration of NO, mg/Nm 3 ；
v q For the flow of flue gas at the reactor inlet, i.e. dry flue gas at actual oxygen content, nm 3 /h；
NSRN is the equivalent molar ratio of urea to nitrogen.
4. A method according to claim 3, characterized in that: setting the concentration of the mixed solution after urea dilution as mu,%;
according to the urea consumption w n The calculation formula for obtaining the average flow of the variable frequency pump is as follows:
f n is the flow rate of the variable frequency pump, kg/h.
The real-time data of the control parameters are obtained from the boiler DCS control system, and the frequency is 0.2Hz, namely 5s each time.
5. The method according to claim 1, wherein: further comprises:
step four, constructing an automatic denitration system to establish NO x A control unit for automatic removal, said control unit being to be associated with NO x The concentration prediction digital processing platforms are interconnected;
controlling the flow of the variable frequency pump and the starting of the urea solution storage and distribution system by using the urea solution flow solving result in the digital processing platform in the control unit;
step six, establishing NO after flue gas denitration x A concentration monitoring system.
6. The method according to claim 5, wherein: NO predicted using a digitizing platform x The concentration is judged in a control unit in a concentration range, and the control unit is used for realizing the matching with the working quantity of the spray heads on the cyclone separator;
the concentration range is as follows:
ρ≤50mg/Nm 3 ；
50<ρ≤100mg/Nm 3 ；
100<ρ≤200mg/Nm 3 ；
ρ>200mg/Nm 3 ；
wherein ρ is predicted NO x Concentration;
the working number range of the spray heads on the cyclone separator is as follows: 2. 6, 10 and 12 spray heads work;
14 spray heads are mounted on the cyclone separator in total, wherein 2 spray heads are reserved for standby.
7. The method according to claim 5, wherein: the urea preparation and storage combined integral module is established and specifically comprises the following steps: a urea solution storage tank with higher concentration is arranged and connected with the control unit;
a water storage tank is arranged above the liquid storage tank and is connected with the control unit;
a urea solution diluting and mixing tank is arranged and is respectively connected with the liquid storage tank and the water storage tank;
and (3) arranging a variable frequency pump for outputting mixed liquid at the outlet of the mixing tank, wherein the variable frequency pump is connected with the spray head in the step four, and plays a role in controlling the total flow of the spray head.
8. The method according to claim 5, wherein: the NO x The concentration monitoring system is characterized in that an NO is arranged at the tail outlet of the flue x A gas concentration sensor;
the concentration sensor is connected with the control unit, and the control unit judges and controls whether the denitration process is operated or not according to signals acquired by the sensor.
9. The method according to claim 8, wherein: if rho is denitrated<50mg/Nm 3 The variable frequency pump is stopped by the control unit, the electromagnetic valve at the front end of the spray head is closed, and the denitration process is finished; if rho is denitrated>50mg/Nm 3 The reserved spray head is started to work, and the control unit is used for controlling the variable frequency pump to increase the overall flow of the spray head.
10. The method according to claim 9, wherein: when the automatic denitration system operates, the control unit firstly predicts NO according to accuracy x The concentration of the urea solution is controlled by controlling the liquid storage tank and the water storage tank to respectively convey a specified amount of high-concentration urea solution and water to the mixing tank, and the urea solution and the water are mixed in the mixing tank to form diluted urea mixed solution;
the control unit then generates a signal according to NO x The concentration prediction digital processing platform controls the variable frequency pump to quantitatively output the diluted urea mixed solution and mix the diluted urea mixed solution with the flue gas according to instructions of the diluted urea solution flow solving result, and NO in the flue gas are removed through the reaction of the urea mixed solution and the flue gas 2 Finally, the denitration of the circulating fluidized bed boiler is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310362815.3A CN116543854A (en) | 2023-04-06 | 2023-04-06 | Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310362815.3A CN116543854A (en) | 2023-04-06 | 2023-04-06 | Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116543854A true CN116543854A (en) | 2023-08-04 |
Family
ID=87449665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310362815.3A Pending CN116543854A (en) | 2023-04-06 | 2023-04-06 | Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116543854A (en) |
-
2023
- 2023-04-06 CN CN202310362815.3A patent/CN116543854A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020088485A1 (en) | Intelligent multi-pollutant ultra-low emission system and global optimization method | |
| CN107243257B (en) | It is suitble to the intelligence spray ammonia control system of full load | |
| CN107526292B (en) | A method of the regulation ammonia spraying amount based on inlet NOx concentration prediction | |
| CN105597537B (en) | Denitration control method based on Prediction and Control Technology | |
| CN112580250A (en) | Thermal power generating unit denitration system based on deep learning and optimization control method | |
| CN108664006A (en) | It is a kind of that ammonia optimization method and system are intelligently sprayed based on the SCR denitration of zonal control and Dynamic matrix control | |
| CN109062053A (en) | A kind of denitration spray ammonia control method based on multivariate calibration | |
| CN108628177A (en) | A kind of SCR denitration intelligence spray ammonia optimization method and system based on model adaptation PID | |
| CN108837698A (en) | Based on advanced measuring instrumentss and the SCR denitration of advanced control algorithm spray ammonia optimization method and system | |
| CN111624876B (en) | Intelligent ammonia injection optimization control system | |
| CN110263452B (en) | Flue gas time distribution characteristic analysis method and system in flue and denitration system | |
| CN108803309A (en) | It is a kind of that ammonia optimization method and system are intelligently sprayed based on the SCR denitration of hard measurement and model adaptation | |
| CN110038426A (en) | A kind of frequency modulation type is intelligent to control ammonia system and method | |
| CN111522290A (en) | Denitration control method and system based on deep learning method | |
| CN112506162B (en) | Oxidation air system control method based on data model and mechanism operation | |
| CN111968708B (en) | SCR denitration ammonia injection amount prediction method based on random forest and LSTM neural network | |
| CN111460726A (en) | Optimization method for ammonia escape of coal slime fluidized bed boiler denitration system | |
| CN115111601A (en) | Multi-target boiler combustion optimization control method with embedded algorithm fusion under variable loads | |
| CN109833773B (en) | Efficient denitration ammonia flow accurate control method | |
| CN111111430A (en) | SCR denitration regulation control method for urea pyrolysis furnace | |
| CN115524976A (en) | SCR system ammonia injection amount adjusting method considering boiler combustion state | |
| CN109046021B (en) | SCR system accurate ammonia injection control method with strong self-adaptive capacity | |
| CN110652856A (en) | SNCR control system based on model | |
| CN116543854A (en) | Digital twin system NO of circulating fluidized bed boiler x Emission prediction and control method | |
| CN116712856A (en) | Intelligent ammonia spraying control method and system for SCR denitration system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |