A dust removal device and dust removal method for a flue gas electric heater

Abstract

The present invention discloses a dust removal device and method for a flue gas electric heater. The device comprises: a flue gas inlet pipe, an atomizing box, a dust pre-agglomeration box, an ultrasonic electric field composite dust collector, and a flue gas exhaust pipe, which are sequentially connected along the airflow direction; a first ultrasonic transmitter is provided in the dust pre-agglomeration box; the ultrasonic electric field composite dust collector comprises an electrostatic precipitator and a second ultrasonic transmitter provided in a dust removal chamber of the electrostatic precipitator; a gas flow controller, a humidity sensor, a temperature sensor, and an intake dust sensor are provided on the flue gas inlet pipe, and an exhaust dust sensor is provided on the flue gas exhaust pipe. On the one hand, the present invention can intelligently adjust and optimize relevant process parameters based on the properties of the flue gas; on the other hand, the present invention can determine whether dust cleaning is necessary based on characteristic indicators that directly reflect the dust accumulation condition of the electrostatic precipitator, thereby enabling timely dust cleaning and avoiding a decrease in dust removal effect due to excessive dust accumulation.

Description

Technical Field

[0001] The present invention relates to the field of dust removal equipment, and in particular to a dust removal device and a dust removal method for a flue gas electric heater. Background Art

[0002] Flue gas, a widely generated waste gas in industrial production, typically contains large amounts of dust and other pollutants and requires purification before discharge. During flue gas purification processes, much of the flue gas needs to be heated to raise its temperature for subsequent processing. Electric flue gas heaters are used to heat the flue gas to the required temperature. Before heating, the flue gas must be dust-removed to minimize any adverse effects on the heater, such as increased energy consumption and damage to equipment.

[0003] The electrostatic precipitator (ESP) is a widely used dust removal device. As flue gas passes through the flue duct in front of the ESP's main structure, it becomes positively charged. The flue gas then enters the ESP's channel, which is equipped with multiple layers of cathode plates. Due to the mutual attraction between the positively charged dust and the cathode plates, particles in the flue gas are attracted to the cathodes. Timed impacts with the cathode plates cause a certain thickness of dust to fall into the hopper below the ESP structure under the combined effects of its own weight and vibration, thereby removing dust from the flue gas.

[0004] Traditional electrostatic precipitators (ESPs) have the following shortcomings: Small dust particles in flue gas are lightweight and have a low charge in the electric field, making them easily carried away by the airflow and difficult to collect, resulting in low removal rates. Furthermore, mechanical vibration during dust removal can easily damage the equipment. Furthermore, fixed cleaning times are often used, which don't accurately determine the timing based on the actual dust accumulation in the ESP, leading to untimely or unnecessary cleaning.

[0005] Ultrasonic dust removal has already been applied. For example, patent CN108939787B discloses a flue gas ultrasonic wet-electrostatic targeted dust removal method and system. This method, by introducing a targeted polymer carrier, utilizes the acoustic agglomeration process to directionally agglomerate tiny particles in the flue gas, particularly those 0.3-0.6 μm in size, and then removes them, achieving complete flue gas purification and emission. Patent CN114432801A also discloses a flue gas dust removal device, flue gas emission purification system and method, and patent CN115006952B discloses an ultrasonic dust removal device. However, these solutions fail to consider the impact of ultrasonic parameters and flue gas properties on the process, and cannot intelligently adjust and optimize relevant process parameters based on flue gas properties.

[0006] Patent CN214811731U discloses an electrostatic precipitator with ultrasonic dust removal, which uses ultrasonic cleaning to effectively avoid damage to the equipment. However, like traditional solutions, the cleaning time is also fixed, which still makes it easy to clean the dust untimely or unnecessarily.

[0007] Therefore, it is necessary to improve the existing technology to provide a more reliable solution. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a dust removal device and a dust removal method for a flue gas electric heater in view of the above-mentioned deficiencies in the prior art.

[0009] To solve the above technical problems, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a dust removal device for a flue gas electric heater, comprising: a flue gas inlet pipe, an atomizing box, a dust pre-agglomeration box, an ultrasonic electric field composite dust collector, and a flue gas exhaust pipe, which are sequentially connected along the airflow direction;

[0010] A first ultrasonic transmitter is provided in the dust pre-agglomeration box, and the ultrasonic electric field composite dust collector includes an electrostatic precipitator and a second ultrasonic transmitter provided in a dust removal chamber of the electrostatic precipitator;

[0011] The flue gas inlet pipe is provided with a gas flow controller, a humidity sensor, a temperature sensor and an intake dust sensor, and the flue gas exhaust pipe is provided with an exhaust dust sensor.

[0012] Preferably, an atomizing nozzle is provided in the atomizing box, and the atomizing nozzle is connected to a water tank provided on the atomizing box through a water pipe for spraying atomized droplets into the atomizing box, and a water flow controller is provided on the water pipe.

[0013] Preferably, the electrostatic precipitator includes an electrostatic precipitator main unit, a discharge electrode and a dust collecting electrode arranged in the dust removal chamber, and a dust collecting hopper located below the electrostatic precipitator chamber, and the second ultrasonic transmitter is arranged in the electrostatic precipitator chamber.

[0014] Preferably, a plurality of guide channels are formed inside the dust pre-agglomeration box by a plurality of guide plates which are parallel to the axial direction of the dust pre-agglomeration box and spaced apart from each other, and both ends of the guide plates have tapered tips.

[0015] Preferably, the dust removal device further comprises a controller connected to the gas flow controller, the humidity sensor, the temperature sensor, the intake dust sensor, the water flow controller, the first ultrasonic transmitter, the second ultrasonic transmitter, the electrostatic precipitator host and the exhaust dust sensor;

[0016] The controller includes a parameter acquisition module, a parameter optimization model, a dust cleaning judgment module and a control module. The parameter optimization model is based on the characteristic parameter Fx of the flue gas introduced into the flue gas inlet pipe and the preset dust concentration target value C in the exhaust gas discharged from the flue gas exhaust pipe. m , using a machine learning algorithm to optimize the process parameter Gy of the dust removal device to obtain an optimized process parameter Gy', and the control module controls the water flow controller, the first ultrasonic transmitter, the second ultrasonic transmitter, and the electrostatic precipitator host according to Gy';

[0017] The characteristic parameters Fx of the flue gas include the flue gas flow rate Qw entering the dust removal device, the relative humidity RH of the flue gas, the temperature T, the dust concentration C in the flue gas, and the dust particle size distribution parameter Lc, which can be expressed as Fx = (Qw, RH, T, C, Lc); wherein the dust particle size distribution parameter Lc includes the concentration C of PM1.0 in the dust. 1.0 、PM2.5 concentration C 2.5 and PM10.0 concentration C 10.0 ;

[0018] The process parameters Gy of the dust removal device include the atomized droplet ratio coefficient K, the power P1 and frequency f1 of the first ultrasonic transmitter, the power P2 and frequency f2 of the second ultrasonic transmitter, and the operating voltage U of the electrostatic precipitator host, denoted as Gy = (K, P1, f1, P2, f2, U);

[0019] The atomized droplet ratio coefficient K is the ratio of the water flow Qs flowing into the water pipe to the flue gas flow Qw flowing into the flue gas inlet pipe, that is, K=Qs / Qw.

[0020] Preferably, the parameter optimization model optimizes the process parameters periodically, and the optimized process parameters are the characteristic parameter Fx of the currently introduced flue gas and the target value C of the dust concentration. m Under the condition of max When , the total power consumption W reaches the minimum;

[0021] The total power consumption W is the sum of the power consumption W1 of the first ultrasonic transmitter, the power consumption W2 of the second ultrasonic transmitter and the power consumption W3 of the electrostatic precipitator host, that is, W=W1+W2+W3.

[0022] Preferably, the cleaning judgment module is used to judge whether the discharge electrode and the dust collecting electrode of the electrostatic precipitator need to be cleaned. When cleaning is required, the control module controls the second ultrasonic transmitter to emit ultrasonic waves to clean the discharge electrode and the dust collecting electrode.

[0023] A second aspect of the present invention provides a dust removal method for the dust removal device for the flue gas electric heater as described above, comprising the following steps:

[0024] S1. During the operation of the dust removal device, the parameter optimization model performs process parameter optimization on a periodic basis: First, the parameter acquisition module obtains the following parameters of the flue gas currently entering the flue gas inlet pipe:

[0025] The flue gas flow rate Q detected by the gas flow controller t , relative humidity RH detected by the humidity sensor t , the temperature T detected by the temperature sensor t And the dust concentration C detected by the intake dust sensor t and dust particle size distribution parameter Lc t , these parameters are used as the characteristic parameters Fx of the flue gas in the current cycle t t , denoted as Fx t =(Q t RH t 、T t 、C t 、Lc t );

[0026] The parameter acquisition module then obtains the preset dust concentration target value C m ;

[0027] S2, the parameter acquisition module collects the characteristic parameters Fx of the flue gas t , dust concentration target value C m Enter the parameter optimization model to obtain the optimized process parameters Gy' in the current cycle t t , Gy' t =(K' t 、P1' t 、f1' t 、P2' t 、f2' t 、U' t );

[0028] K' t Represents the optimized atomized droplet ratio coefficient, P1' t 、f1' t Respectively represent the power and frequency of the first ultrasonic transmitter obtained by optimization, P2' t 、f2' t Respectively represent the power and frequency of the optimized second ultrasonic transmitter, U' t Indicates the optimized working voltage of the electrostatic precipitator host;

[0029] S3, the control module obtains the optimized process parameter Gy' according to the parameter optimization model t , control so that in the current cycle t:

[0030] The ratio of the water flow rate in the water pipe to the flue gas flow rate in the flue gas inlet pipe is K' t , the power of the first ultrasonic transmitter is P1' t , frequency is f1' t , the power of the second ultrasonic transmitter is P2' t , frequency is f2' t , the working voltage of the electrostatic precipitator host is U' t ;

[0031] S4. In the next cycle t+1, repeat steps S1-S3.

[0032] Preferably, the parameter optimization model is constructed by the following method:

[0033] S1-1. Construct training dataset Z:

[0034] S1-1-1. The flue gas is passed into the dust removal device for treatment. First, the flue gas flow rate Qw and the target dust concentration value C in the exhaust gas are fixed. m ;

[0035] Get different relative humidity RH i , temperature T i , dust concentration C i And the dust particle size distribution parameter Lc i Under this condition, the actual dust concentration in the exhaust gas (detected by the exhaust dust sensor) is not higher than the dust concentration target value C m When the total power consumption W reaches the minimum, the process parameter Gy i , the total power consumption at this time is W i ;

[0036] Among them, the process parameter Gy i Including the atomized droplet ratio coefficient K i , the power P1 of the first ultrasonic transmitter i and frequency f1 i , the power of the second ultrasonic transmitter P2 i and frequency f2 i , Working voltage U of electrostatic precipitator host i ;

[0037] Qw、C m RH i 、T i 、C i 、Lc i and the corresponding Gy i 、W i Combine to get training data z i, thereby obtaining the corresponding flue gas flow Qw and dust concentration target value C m The n training data under, i = 1, 2, ..., n, n = 10-100;

[0038] S1-1-2. Change the value of the incoming flue gas flow Qw and the target value of the dust concentration C m , use the same method as step S1-1-1 to obtain m training data z j ,i=1,2,...,m,m=50-5000;

[0039] S1-1-3, combine all the training data to obtain the training data set Z;

[0040] S1-2. Model training:

[0041] Using the training data set Z, Qw, C m RH i 、T i 、C i 、Lc i For access, Gy i The CNN convolutional neural network is trained for the target output, and the parameter optimization model is obtained after the training is completed.

[0042] Preferably, during the operation of the dust removal device, in each cycle t0 of the process parameter optimization model, the exhaust dust sensor collects N dust concentration values ​​in the exhaust gas, which are recorded as the actual dust concentration C S , N = 5-200; the parameter acquisition module obtains the detection results of the exhaust dust sensor and transmits them to the cleaning judgment module. The cleaning judgment module determines whether cleaning is required in each cycle t0 according to the following method:

[0043] 1) Calculate the actual dust concentration C within the period t0 S Greater than the dust concentration target value C m The number of dust concentration data values ​​N1, if N1 / N ≥ 0.01 ~ 0.05, go to step 2), otherwise it is determined that dust cleaning is not required;

[0044] 2) Calculate the total time of each cycle t0 as △t, and divide one cycle into four time periods: 0~0.25△t, 0.25△t~0.5△t, 0.5△t~0.75△t, 0.75△t~△t, which are recorded as time period 1, time period 2, time period 3, and time period 4 respectively. Calculate the actual dust concentration C in each time period S The average value of If satisfied If it is not, it is judged that cleaning is required; otherwise, cleaning is not required;

[0045] 3) When the cleaning judgment module determines that cleaning is required, after the current cycle ends, the control module controls the cleaning operation in the following manner: controlling the gas flow controller to be closed, stopping the introduction of flue gas, controlling the water flow controller to be closed, stopping the spraying of atomized droplets, controlling the first ultrasonic transmitter to stop working, controlling the electrostatic precipitator host to stop working, and controlling the second ultrasonic transmitter to work, emitting ultrasonic waves to clean the discharge electrode and the dust collecting electrode;

[0046] After the dust cleaning is completed, the first ultrasonic transmitter, the second ultrasonic transmitter and the electrostatic precipitator host are controlled to work according to the optimized process parameter Gy' obtained in the previous cycle, and the water flow controller is controlled to open, and finally the gas flow controller is controlled to open to re-introduce the flue gas.

[0047] The beneficial effects of the present invention are:

[0048] The present invention provides a dust removal device and dust removal method for a flue gas electric heater. On the one hand, the present invention can intelligently adjust and optimize relevant process parameters based on the properties of the flue gas. On the other hand, the present invention can determine whether dust cleaning is required based on characteristic indicators that directly reflect the dust accumulation of the electrostatic precipitator, thereby enabling timely dust cleaning and avoiding a decrease in dust removal effect due to excessive dust accumulation.

[0049] The present invention adopts a CNN convolutional neural network as the basic network model, and constructs a parameter optimization model by training a training data set composed of the optimized process parameter data corresponding to the characteristic parameters of different flue gases and the target dust concentration values ​​obtained in advance. The model can analyze the characteristic parameters of the flue gas and the target dust concentration values ​​to obtain the process parameters with the relatively lowest energy consumption while meeting the dust removal requirements, that is, the optimized process parameters. The optimized process parameters are used to control the dust removal operation of the dust removal device, thereby achieving efficient and low-energy dust removal.

[0050] The present invention uses a cleaning judgment module to determine the timing of cleaning, enabling timely cleaning and avoiding a significant decrease in dust removal efficiency due to excessive dust accumulation. Furthermore, when cleaning is required, the control module controls the second ultrasonic transmitter to emit ultrasonic waves to clean the discharge and dust collection electrodes, avoiding the drawback of conventional mechanical vibration cleaning that easily damages the electrodes and achieving better cleaning results.

[0051] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and to implement it according to the contents of the description, the following preferred embodiments of the present invention are described in detail with reference to the accompanying drawings. The specific implementation methods of the present invention are given in detail by the following embodiments and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1This is a schematic structural diagram of the dust removal device for the flue gas electric heater in Example 1;

[0053] Figure 2 This is a functional block diagram of the controller in Example 1;

[0054] Figure 3 This is a flow chart of the dust removal method of the dust removal device for the flue gas electric heater in Example 2;

[0055] Figure 4 This is a flow chart for constructing the parameter optimization model in Example 2;

[0056] Figure 5 This is a flowchart of constructing the training data set Z in Example 2;

[0057] Figure 6 This is a flow chart of the dust cleaning judgment module in Example 2 for judging whether dust cleaning is required;

[0058] Figure 7 The actual dust concentration test results of the control case and the test case in the application test;

[0059] Figure 8 is the test result of the power consumption reference value η of the test case in the application test.

[0060] Description of reference numerals:

[0061] 1—Flue gas inlet pipe; 2—Atomization box; 3—Dust pre-agglomeration box; 4—Ultrasonic electric field composite dust collector; 5—Flue gas exhaust pipe;

[0062] 11—gas flow controller; 12—humidity sensor; 13—temperature sensor; 14—intake dust sensor;

[0063] 21—atomizing nozzle; 22—water tank; 23—water pipe; 24—water flow controller;

[0064] 31 - first ultrasonic transmitter; 32 - guide vane; 33 - guide channel;

[0065] 41—electrostatic precipitator; 42—dust removal chamber; 43—electrostatic precipitator main unit; 44—discharge electrode; 45—dust collecting electrode; 46—dust collecting hopper; 47—second ultrasonic transmitter;

[0066] 51—Exhaust dust sensor;

[0067] 6—controller; 61—parameter acquisition module; 62—parameter optimization model; 63—ash cleaning judgment module; 64—control module. DETAILED DESCRIPTION

[0068] The present invention is further described in detail below with reference to the embodiments so that those skilled in the art can implement the invention with reference to the description.

[0069] It should be understood that terms such as “having”, “including” and “comprising” used herein do not preclude the existence or addition of one or more other elements or combinations thereof.

[0070] Example 1

[0071] Reference Figure 1-2 This embodiment provides a dust removal device for a flue gas electric heater, comprising: a flue gas inlet pipe 1, an atomizing box 2, a dust pre-agglomeration box 3, an ultrasonic electric field composite dust collector 4, and a flue gas exhaust pipe 5, which are sequentially connected along the airflow direction;

[0072] A first ultrasonic transmitter 31 is provided in the dust pre-agglomeration box 3, and the ultrasonic electric field composite dust collector 4 includes an electrostatic precipitator 41 and a second ultrasonic transmitter 47 provided in a dust removal chamber 42 of the electrostatic precipitator 41;

[0073] The flue gas inlet pipe 1 is provided with a gas flow controller 11 , a humidity sensor 12 , a temperature sensor 13 and an intake dust sensor 14 , and the flue gas exhaust pipe 5 is provided with an exhaust dust sensor 51 .

[0074] In this embodiment, a plurality of atomizing nozzles 21 are provided in the atomizing box 2, and all the atomizing nozzles 21 are connected to a water tank 22 provided on the atomizing box 2 through a water pipe 23, so as to spray atomized droplets into the atomizing box 2. A water flow controller 1124 is provided on the water pipe 23 for controlling the flow of water entering all the atomizing nozzles 21, which can realize automatic control.

[0075] A plurality of guide channels 33 are formed inside the dust pre-agglomeration box 3 by a plurality of guide vanes 32 which are arranged parallel to the axial direction of the dust pre-agglomeration box 3 and spaced apart from each other. Both ends of the guide vanes 32 have tapered tips.

[0076] The electrostatic precipitator 41 includes an electrostatic precipitator main unit 43 , a discharge electrode 44 and a dust collecting electrode 45 arranged in a dust removal chamber 42 , a dust collecting hopper 46 located below the electrostatic precipitator chamber 42 , and a second ultrasonic transmitter 47 arranged in the electrostatic precipitator chamber 42 .

[0077] The flue gas entering the dust removal device enters from the flue gas inlet pipe 1, and then passes through the atomization box 2, the dust pre-agglomeration box 3, and the ultrasonic electric field composite dust collector 4 in sequence. After dust removal, it is discharged from the flue gas exhaust pipe 5 and enters subsequent treatment.

[0078] In the atomizing box 2, the purpose of spraying atomized droplets through the atomizing nozzle 21 is to use the atomized droplets as a carrier to promote the agglomeration of dust in the flue gas under the subsequent ultrasonic action. Acoustic agglomeration is to emit sound waves to make the dust in the gas vibrate within a certain range, so that it can collide with surrounding particles and even merge and agglomerate. By introducing atomized droplets, the atomized droplets and dust particles agglomerate or collide with each other, causing liquid water to be contained in the dust, or dust particles to be contained in the droplets. This prepares for subsequent dust removal.

[0079] In the dust pre-agglomeration box 3, the dust in the flue gas is initially agglomerated by the action of ultrasound before entering the ultrasonic electric field composite dust collector 4, which can improve the subsequent dust removal efficiency in the ultrasonic electric field composite dust collector 4. The multiple guide channels 33 in the dust pre-agglomeration box 3 can help to evenly distribute the flue gas entering the ultrasonic electric field composite dust collector 4 and can also help promote the pre-agglomeration of dust in the dust pre-agglomeration box 3.

[0080] In the ultrasonic electric field composite dust collector 4, the applied ultrasonic wave further promotes the agglomeration of dust in the flue gas, especially promotes the full agglomeration of fine dust in the flue gas, and forms larger aggregate particles through two ultrasonic waves, which can facilitate the efficient removal of dust by the electrostatic precipitator 41 (dust with small particle size is light in weight and has less charge under the electric field, so it is easily carried away by the airflow and is not easily collected by the electric field, resulting in a low removal rate).

[0081] Therefore, in the present invention, the efficiency of removing dust in the flue gas by the electrostatic precipitator 41 can be significantly improved by combining the carrier effect of the atomized droplets and the auxiliary effect of the two-stage ultrasonic waves.

[0082] In this embodiment, the dust removal device further includes a controller 6 connected to the gas flow controller 11, the humidity sensor 12, the temperature sensor 13, the intake dust sensor 14, the water flow controller 1124, the first ultrasonic transmitter 31, the second ultrasonic transmitter 47, the electrostatic precipitator host 43, and the exhaust dust sensor 51;

[0083] The controller 6 includes a parameter acquisition module 61, a parameter optimization model 62, a dust cleaning judgment module 63 and a control module 64. The parameter optimization model 62 is based on the characteristic parameter Fx of the flue gas introduced into the flue gas inlet pipe 1 and the preset dust concentration target value C in the exhaust gas discharged from the flue gas exhaust pipe 5. m , a machine learning algorithm is used to optimize the process parameter Gy of the dust removal device to obtain the optimized process parameter Gy'. The control module 64 controls the water flow controller 1124, the first ultrasonic transmitter 31, the second ultrasonic transmitter 47, and the electrostatic precipitator host 43 according to Gy';

[0084] The characteristic parameters Fx of the flue gas include the flue gas flow rate Qw entering the dust removal device, the relative humidity RH of the flue gas, the temperature T, the dust concentration C in the flue gas, and the dust particle size distribution parameter Lc, which can be expressed as Fx = (Qw, RH, T, C, Lc); wherein the dust particle size distribution parameter Lc includes the concentration C of PM1.0 in the dust. 1.0 、PM2.5 concentration C 2.5 and PM10.0 concentration C 10.0 ;

[0085] The process parameters Gy of the dust removal device include the atomized droplet ratio coefficient K, the power P1 and frequency f1 of the first ultrasonic transmitter 31, the power P2 and frequency f2 of the second ultrasonic transmitter 47, and the operating voltage U of the electrostatic precipitator host 43, and are expressed as Gy = (K, P1, f1, P2, f2, U);

[0086] The atomized droplet ratio coefficient K is the ratio of the water flow Qs flowing into the water pipe 23 to the flue gas flow Qw flowing into the flue gas inlet pipe 1, that is, K=Qs / Qw.

[0087] In this embodiment, the parameter optimization model 62 optimizes the process parameters in a periodic manner. The optimized process parameters are the characteristic parameter Fx of the currently introduced flue gas and the target value C of the dust concentration. m Under the condition of max When , the total power consumption W reaches the minimum;

[0088] The total power consumption W is the sum of the power consumption W1 of the first ultrasonic transmitter 31, the power consumption W2 of the second ultrasonic transmitter 47 and the power consumption W3 of the electrostatic precipitator main unit 43, that is, W=W1+W2+W3.

[0089] The dust removal device of the present invention is used for removing dust from cold flue gas at the front end of the flue gas electric heater. The flue gas after dust removal is then heated and then enters the subsequent purification process. In other words, the processing object of the present invention is cold flue gas that needs to be reheated later. The temperature fluctuation range will be relatively large, mainly depending on the previous processing operation. For example, in some embodiments, the temperature range is 50-150°C. The temperature of the flue gas will have a great influence on the efficiency of the electrostatic precipitator. For example, when the flue gas temperature is too low, in the electrostatic precipitator 41, due to insufficient energy, it is difficult for the particulate matter to obtain sufficient charge, resulting in the particulate matter being difficult to be captured and the dust removal efficiency will be reduced.

[0090] Flue gas humidity also affects the efficiency of electrostatic precipitators. When the humidity is too low, dust agglomeration is reduced, hindering its capture by the electrodes. When the humidity is too high, water vapor becomes a conductor in the air, increasing the conductivity of the air, which reduces the electric field strength and thus affects the overall performance of the electrostatic precipitator 41. In the present invention, the atomized droplet ratio coefficient K represents the amount of atomized droplets introduced into the flue gas. It can regulate the flue gas humidity and directly affect the amount of atomized droplet carriers in the flue gas, thus significantly affecting the dust removal efficiency.

[0091] In the two-stage ultrasonic operation, the main operating parameters of the first ultrasonic transmitter 31 and the second ultrasonic transmitter 47: ultrasonic power and frequency (i.e., the frequency of the generated sound waves) are the main factors affecting dust removal efficiency. In the electrostatic precipitator 41, its operating voltage is the main parameter affecting dust removal efficiency.

[0092] From the above analysis, it can be seen that the atomized droplet proportion coefficient K, the power P1 and frequency f1 of the first ultrasonic transmitter 31, the power P2 and frequency f2 of the second ultrasonic transmitter 47, and the working voltage U of the electrostatic precipitator host 43 will have a great influence on the dust removal efficiency of the dust removal device of the present invention. By optimizing these process parameters, the dust removal efficiency can be effectively improved, and energy consumption can be reduced while meeting the dust removal requirements, thereby reducing costs.

[0093] However, the effects of these process parameters on dust removal efficiency are nonlinear and mutually superimposed. The relationship between process parameters and dust removal efficiency is relatively complex and difficult to accurately express through specific mathematical formulas. In the present invention, a parameter optimization model 62 based on a machine learning algorithm is adopted. The relationship between process parameters and dust removal efficiency is analyzed and accumulated through experience using previous training data. The model can be used to calculate the dust removal efficiency based on the characteristic parameter Fx of the flue gas and the target value C of the dust concentration. m A more accurate analysis can obtain the process parameters with the lowest energy consumption while meeting the dust removal requirements. These parameters are output as optimized process parameters, and the dust removal device is controlled to achieve high-efficiency and low-energy dust removal.

[0094] In this embodiment, the cleaning determination module 63 is used to determine whether the discharge electrode 44 and the dust collecting electrode 45 of the electrostatic precipitator 41 need to be cleaned. When cleaning is required, the control module 64 controls the second ultrasonic transmitter 47 to emit ultrasonic waves to clean the discharge electrode 44 and the dust collecting electrode 45. By determining the timing of cleaning with the cleaning determination module 63, timely cleaning can be achieved, avoiding a significant decrease in dust removal efficiency due to excessive dust accumulation. This method will be described in detail in the following embodiments.

[0095] Example 2

[0096] Reference Figure 3-6This embodiment provides a dust removal method for a dust removal device for a flue gas electric heater according to embodiment 1, comprising the following steps:

[0097] S1. During the operation of the dust removal device, the parameter optimization model 62 performs process parameter optimization on a periodic basis: First, the parameter acquisition module 61 obtains the following parameters of the flue gas currently entering the flue gas inlet pipe 1:

[0098] The flue gas flow rate Q detected by the gas flow controller 11 t , relative humidity RH detected by humidity sensor 12 t , the temperature T detected by the temperature sensor 13 t and the dust concentration C detected by the intake dust sensor 14 t and dust particle size distribution parameter Lc t , these parameters are used as the characteristic parameters Fx of the flue gas in the current cycle t t , denoted as Fx t =(Q t RH t 、T t 、C t 、Lc t );

[0099] The parameter acquisition module 61 then obtains the preset dust concentration target value C m ;

[0100] S2, the parameter acquisition module 61 collects the characteristic parameters Fx of the flue gas t , dust concentration target value C m Enter the parameter optimization model 62 to obtain the optimized process parameters Gy' in the current cycle t t , Gy' t =(K' t 、P1' t 、f1' t 、P2' t 、f2' t 、U' t );

[0101] K' t Represents the optimized atomized droplet ratio coefficient, P1' t 、f1' t and P2′ represent the optimized power and frequency of the first ultrasonic transmitter 31, respectively. t 、f2' t Respectively represent the optimized power and frequency of the second ultrasonic transmitter 47, U' t represents the optimized working voltage of the electrostatic precipitator host 43;

[0102] S3, the control module 64 obtains the optimized process parameter Gy' according to the parameter optimization model 62 t , control so that in the current cycle t:

[0103] The ratio of the flow rate of water entering the water pipe 23 to the flow rate of flue gas entering the flue gas inlet pipe 1 is K' t , the power of the first ultrasonic transmitter 31 is P1' t , frequency is f1' t , the power of the second ultrasonic transmitter 47 is P2' t , frequency is f2' t , the working voltage of the electrostatic precipitator host 43 is U' t ;

[0104] S4. In the next cycle t+1, repeat steps S1-S3.

[0105] The parameter optimization model 62 optimizes the process parameters, that is, within a time period, the optimized process parameters obtained in the period are controlled until the next time period is optimized and updated again. The time period t is conventionally selected according to actual conditions; for example, when the properties of the flue gas entering the dust removal device (including characteristic parameters such as flow rate, humidity, temperature, dust concentration, dust particle size distribution parameters, etc.) fluctuate slightly, a larger value can be selected; when the fluctuation is large, a smaller value is selected for the period, for example, the period is 5-240 minutes. Furthermore, in one embodiment, the period is 30 minutes; that is, the process can be optimized in time to ensure the dust removal effect, and the data processing volume of the parameter optimization model 62 is too large to cause waste.

[0106] In this embodiment, the parameter optimization model 62 is constructed by the following method:

[0107] S1-1. Construct training dataset Z:

[0108] S1-1-1. The flue gas is passed into the dust removal device for treatment. First, the flue gas flow rate Qw and the target dust concentration value C in the exhaust gas are fixed. m ;

[0109] Get different relative humidity RH i , temperature T i , dust concentration C i And the dust particle size distribution parameter Lc i Under this condition, the actual dust concentration in the exhaust gas (detected by the exhaust dust sensor 51) is not higher than the dust concentration target value C m When the total power consumption W reaches the minimum, the process parameter Gy i , the total power consumption at this time is W iThe total power consumption W is the sum of the power consumption W1 of the first ultrasonic transmitter 31, the power consumption W2 of the second ultrasonic transmitter 47 and the power consumption W3 of the electrostatic precipitator host 43, that is, W = W1 + W2 + W 3；

[0110] Among them, the process parameter Gy i Including the atomized droplet ratio coefficient K i , the power P1 of the first ultrasonic transmitter 31 i and frequency f1 i , the power P2 of the second ultrasonic transmitter 47 i and frequency f2 i , Working voltage U of electrostatic precipitator host 43 i ;

[0111] Qw、C m RH i 、T i 、C i 、Lc i and the corresponding Gy i 、W i Combine to get training data z i , thereby obtaining the corresponding flue gas flow Qw and dust concentration target value C m n training data under, i=1,2,...,n, n=10-100; Lc i The concentration of PM1.0 in dust is included in C 1.0i 、PM2.5 concentration C 2.5i and PM10.0 concentration C 10.0i ;

[0112] In this embodiment, RH i The value is taken in the range of 5%-95%, T i The value is taken in the range of 50-150℃, C i 20-500 mg / m 3 Take values ​​within the range of

[0113] In this embodiment, the atomized droplet ratio coefficient K i The value is taken in the range of 0.01-0.2; P1 i The value is selected in the range of 100-1000W, f1 i The value is selected in the range of 10-100KHz; P2 i The value is selected in the range of 200-1500W, f2 i The value is selected within the range of 25-150KHz; U i (Operating voltage) is within the range of 3000V-15000V.

[0114] In this embodiment, a total of 80 pieces of training data are obtained within the above parameter range, that is, n=80.

[0115] S1-1-2. Change the value of the incoming flue gas flow Qw and the target value of the dust concentration C m , use the same method as step S1-1-1 to obtain m training data z j ,i=1,2,...,m,m=50-5000;

[0116] In this embodiment, the flue gas flow rate Qw is 5000Nm 3 / h-30000Nm 3 / h range, C m 1-20 mg / m 3 The value of m is 800.

[0117] S1-1-3, combine all the training data to obtain the training data set Z;

[0118] S1-2. Model training:

[0119] Using the training data set Z, Qw, C m RH i 、T i 、C i 、Lc i For access, Gy i The CNN convolutional neural network is trained for the target output, and after the training is completed, a parameter optimization model 62 is obtained.

[0120] The present invention adopts a CNN convolutional neural network as a basic network model, and constructs a parameter optimization model 62 by training a training data set composed of optimized process parameter data corresponding to different flue gas characteristic parameters and dust concentration target values ​​obtained in advance. The parameter optimization model 62 can analyze the characteristic parameters of the flue gas and the dust concentration target values ​​to obtain the process parameters with relatively lowest energy consumption while meeting the dust removal requirements, that is, the optimized process parameters. The optimized process parameters are used to control the dust removal operation of the dust removal device, so as to achieve high-efficiency and low-energy dust removal.

[0121] In this embodiment, during the operation of the dust removal device, in each cycle t0 of the process parameter optimization model 62, the exhaust dust sensor 51 collects N dust concentration values ​​in the exhaust gas, which are recorded as the actual dust concentration C S, N = 5-200, the value of N is selected according to the size of t0. For example, in one embodiment, t0 = 30 min and N = 100. The parameter acquisition module 61 obtains the detection result of the exhaust dust sensor 51 and transmits it to the cleaning judgment module 63. The cleaning judgment module 63 judges whether the cleaning process is required in each cycle t0 according to the following method:

[0122] 1) Calculate the actual dust concentration C within the period t0 S Greater than the dust concentration target value C m The number of dust concentration data values ​​N1, if N1 / N ≥ 0.05, go to step 2), otherwise it is determined that cleaning is not required;

[0123] When the actual dust concentration C S There are a few values ​​greater than the dust concentration target value C m When the concentration is high, it can be considered that it is an occasional high concentration caused by normal fluctuations, which does not mean that there is too much dust accumulation. However, if a large number (for example, more than 5) of the actual dust concentration values ​​are higher than the dust concentration target value C m If the dust concentration in the flue gas after the electrostatic precipitator 41 is too high, it should be considered that the dust accumulation on the electrodes (discharge electrode 44 and dust collecting electrode 45) of the electrostatic precipitator 41 may be too much, resulting in a significant decrease in dust removal efficiency and a high dust concentration in the flue gas after the electrostatic precipitator 41 is dusted. Further judgment needs to be made through step 2).

[0124] 2) Calculate the total time of each cycle t0 as △t, and divide one cycle into four time periods: 0~0.25△t, 0.25△t~0.5△t, 0.5△t~0.75△t, 0.75△t~△t, which are recorded as time period 1, time period 2, time period 3, and time period 4 respectively. Calculate the actual dust concentration C in each time period S The average value of If satisfied If it is not, it is judged that cleaning is required; otherwise, cleaning is not required;

[0125] When the actual dust concentration value is higher than the dust concentration target value C in step 1) mWhen there is too much data, this step is used to confirm whether there is too much dust accumulation. In this step, a cycle t0 is divided into 4 segments in chronological order, and the actual dust concentration average value within a time period is determined. If the average value shows a trend of continuous increase, it means that the actual dust concentration value as a whole shows a trend of gradually increasing with time. It can be considered that the electrodes (discharge electrode 44, dust collecting electrode 45) of the electrostatic precipitator 41 have too much dust accumulation, and the dust accumulation is constantly increasing. The dust removal efficiency is significantly affected and needs to be cleaned. Otherwise, it is considered that the actual dust concentration value is in a normal fluctuation state and no cleaning is currently required. Combining step 1) with step 2) to determine whether cleaning is needed can prevent misjudgment and unnecessary cleaning, and can ensure timely cleaning.

[0126] 3) When the cleaning judgment module 63 determines that cleaning is required, after the current cycle ends, the control module 64 controls the cleaning operation in the following manner: controls the gas flow controller 11 to be closed, stops the introduction of flue gas, controls the water flow controller 1124 to be closed, stops the injection of atomized droplets, controls the first ultrasonic transmitter 31 to stop working, controls the electrostatic precipitator host 43 to stop working, controls the second ultrasonic transmitter 47 to work, and emits ultrasonic waves to clean the discharge electrode 44 and the dust collecting electrode 45; at this time, the power and frequency of the second ultrasonic transmitter 47 will be greater than their values ​​during normal dust removal operation. For example, in this embodiment, the power of the second ultrasonic transmitter 47 in the cleaning stage is 1200W and the frequency is 180KHz.

[0127] After the dust cleaning is completed, the first ultrasonic transmitter 31, the second ultrasonic transmitter 47 and the electrostatic precipitator host 43 are controlled to work according to the optimized process parameter Gy' obtained in the previous cycle, and the water flow controller 1124 is controlled to open, and finally the gas flow controller 11 is controlled to open to re-introduce the flue gas.

[0128] That is, after each dust removal cycle, the dust cleaning judgment module 63 determines whether dust cleaning is required. If not, the dust removal device continues to operate normally and directly enters the next dust removal cycle. If the dust cleaning judgment module 63 determines that dust cleaning is required, the second ultrasonic transmitter 47 emits ultrasonic waves, which use the vibration generated to cause the dust accumulated on the discharge electrode 44 and the dust collecting electrode 45 to fall off and be collected and discharged through the dust hopper 46. This method uses parameters that directly reflect the dust removal effect of the electrostatic precipitator 41 to determine the dust accumulation situation, allowing for timely dust cleaning of the discharge electrode 44 and the dust collecting electrode 45, avoiding a significant decrease in dust removal efficiency due to excessive dust accumulation. Ultrasonic dust cleaning also avoids the defect of conventional mechanical vibration dust cleaning that easily damages the electrodes, resulting in better dust cleaning results.

[0129] Application Testing

[0130] The flue gas is cold asphalt flue gas from a company's aluminum forming workshop. The flue gas properties are as follows: The flue gas flow rate is 10000-12000Nm 3 / h, and the dust concentration in the flue gas is between 45-55mg / m 3 The temperature fluctuates within the range of 120-145℃ and the relative humidity fluctuates within the range of 10-25%.

[0131] The method of Example 2 is used to remove dust from the flue gas, and the dust concentration target value C is set. m =3mg / m 3 The main process parameters Gy at the initial stage are set as follows: atomized droplet proportion coefficient K = 0.025, the power P1 = 350 W and the frequency f1 = 45 kHz of the first ultrasonic transmitter 31, the power P2 = 900 W and the frequency f2 = 80 kHz of the second ultrasonic transmitter 47, and the operating voltage U = 11000 V of the electrostatic precipitator 41; the power P2 = 1200 W and the frequency f2 = 180 kHz of the second ultrasonic transmitter 47 during the cleaning stage.

[0132] A test example and a control example were set up. The test example completely followed the method of Example 2, using the above-mentioned process parameter Gy as the initial parameter. During the operation, the process parameter Gy was continuously optimized by the parameter optimization model 62 (t0=30min); in the control example, the above-mentioned process parameter Gy was kept unchanged; after running for 12 hours, a sampling index test was performed every 6 hours, and the average value of the actual dust concentration in the exhaust gas within 1 hour and the total power consumption W (the total power consumption W is the power consumption W1 of the first ultrasonic transmitter 31, the power consumption W2 of the second ultrasonic transmitter 47) were calculated. The power consumption W2 of the control example and the power consumption W3 of the electrostatic precipitator host 43, that is, W=W1+W2+W3), the total power consumption of the control example is 100%, and the power consumption reference value η of the test example is calculated, η=(total power consumption of the test example / total power consumption of the control example)*100%; several sampling tests are carried out continuously, which are recorded as the first sampling time period (18-19h), the second sampling time period (24-25h), the third sampling time period (30-31h), and the fourth sampling time period (36-37h). The test results are shown in Tables 1 and 2. Figure 7 、 Figure 8 As shown:

[0133] Table 1

[0134]

[0135] It can be seen from the test results that by optimizing the process parameters in Example 2, energy consumption can be effectively reduced while ensuring the dust removal effect.

[0136] The above are only preferred embodiments of the present invention and do not limit the present invention in any form. Any ordinary technician in this industry can smoothly implement the present invention as shown in the drawings and the above. However, any equivalent changes, modifications and evolutions made by technicians familiar with this profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are still within the scope of protection of the technical solution of the present invention.

Claims

1. A dust removal device for a flue gas electric heater, characterized in that: include: The flue gas inlet pipe, atomizing box, dust pre-agglomeration box, ultrasonic electric field composite dust collector and flue gas exhaust pipe are sequentially connected along the airflow direction; A first ultrasonic transmitter is provided in the dust pre-agglomeration box, and the ultrasonic electric field composite dust collector includes an electrostatic precipitator and a second ultrasonic transmitter provided in a dust removal chamber of the electrostatic precipitator; The flue gas inlet pipe is equipped with a gas flow controller, a humidity sensor, a temperature sensor and an intake dust sensor, and the flue gas exhaust pipe is equipped with an exhaust dust sensor; The dust removal device also includes a controller connected to the gas flow controller, the humidity sensor, the temperature sensor, the intake dust sensor, the water flow controller, the first ultrasonic transmitter, the second ultrasonic transmitter, the electrostatic precipitator host and the exhaust dust sensor; The controller includes a parameter acquisition module, a parameter optimization model, a dust cleaning judgment module and a control module. The parameter optimization model uses a machine learning algorithm to optimize the process parameters in a cycle to obtain the optimized process parameter Gy'. The optimized process parameter is the characteristic parameter Fx of the current flue gas and the dust concentration target value C m Under the condition of max When , the total power consumption W reaches the minimum; The total power consumption W is the sum of the power consumption W1 of the first ultrasonic transmitter, the power consumption W2 of the second ultrasonic transmitter, and the power consumption W3 of the electrostatic precipitator host, that is, W=W1+W2+W3; The process parameters Gy of the dust removal device include the atomized droplet ratio coefficient K, the power P1 and frequency f1 of the first ultrasonic transmitter, the power P2 and frequency f2 of the second ultrasonic transmitter, and the operating voltage U of the electrostatic precipitator host, denoted as Gy = (K, P1, f1, P2, f2, U); The atomized droplet ratio coefficient K is the ratio of the water flow Qs flowing into the water pipe to the flue gas flow Qw flowing into the flue gas inlet pipe, that is, K = Qs / Qw; During the operation of the dust removal device, in each cycle t0 of the process parameter optimization model, the exhaust dust sensor collects N dust concentration values ​​in the exhaust gas, which are recorded as the actual dust concentration C S The dust cleaning judgment module determines whether dust cleaning is required in each cycle t0 according to the following method: 1) Calculate the actual dust concentration C within the period t0 S Greater than the dust concentration target value C m The number of dust concentration data values ​​N1, if N1 / N ≥ 0.01 ~ 0.05, go to step 2), otherwise it is determined that dust cleaning is not required; 2) Calculate the total time of each cycle t0 as △t, and divide one cycle into four time periods: 0~0.25△t, 0.25△t~0.5△t, 0.5△t~0.75△t, 0.75△t~△t, which are recorded as time period 1, time period 2, time period 3, and time period 4 respectively. Calculate the actual dust concentration C in each time period S The average value of If satisfied If so, it is determined that dust cleaning is required, otherwise it is determined that dust cleaning is not required.

2. The dust removal device for a flue gas electric heater according to claim 1, characterized in that: An atomizing nozzle is provided in the atomizing box, and the atomizing nozzle is connected to a water tank provided on the atomizing box through a water pipe for spraying atomized liquid droplets into the atomizing box. A water flow controller is provided on the water pipe.

3. The dust removal device for a flue gas electric heater according to claim 2, characterized in that: The electrostatic precipitator includes an electrostatic precipitator main unit, a discharge electrode and a dust collecting electrode arranged in the dust removal chamber, and a dust collecting hopper located below the dust removal chamber. The second ultrasonic transmitter is arranged in the dust removal chamber.

4. The dust removal device for a flue gas electric heater according to claim 1, characterized in that: A plurality of guide channels are formed inside the dust pre-agglomeration box by a plurality of guide plates which are parallel to the axial direction of the dust pre-agglomeration box and spaced apart from each other, and both ends of the guide plates have tapered tips.

5. The dust removal device for a flue gas electric heater according to claim 3, characterized in that: The parameter optimization model is based on the characteristic parameter Fx of the flue gas introduced into the flue gas inlet pipe and the preset target value C of the dust concentration in the exhaust gas discharged from the flue gas exhaust pipe. m , using a machine learning algorithm to optimize the process parameter Gy of the dust removal device to obtain an optimized process parameter Gy', and the control module controls the water flow controller, the first ultrasonic transmitter, the second ultrasonic transmitter, and the electrostatic precipitator host according to Gy'; The characteristic parameters Fx of the flue gas include the flue gas flow rate Qw entering the dust removal device, the relative humidity RH of the flue gas, the temperature T, the dust concentration C in the flue gas, and the dust particle size distribution parameter Lc, which can be expressed as Fx = (Qw, RH, T, C, Lc); wherein the dust particle size distribution parameter Lc includes the concentration C of PM1.0 in the dust. 1.0 、PM2.5 concentration C 2.5 and PM10.0 concentration C 10.0 .

6. The dust removal device for a flue gas electric heater according to claim 1, characterized in that: The cleaning judgment module is used to judge whether the discharge electrode and the dust collecting electrode of the electrostatic precipitator need to be cleaned. When cleaning is required, the control module controls the second ultrasonic transmitter to emit ultrasonic waves to clean the discharge electrode and the dust collecting electrode.

7. A dust removal method for a dust removal device for a flue gas electric heater according to any one of claims 5 to 6, characterized in that: The following steps are involved: S1. During the operation of the dust removal device, the parameter optimization model performs process parameter optimization on a periodic basis: First, the parameter acquisition module obtains the following parameters of the flue gas currently entering the flue gas inlet pipe: The flue gas flow rate Q detected by the gas flow controller t , relative humidity RH detected by the humidity sensor t , the temperature T detected by the temperature sensor t And the dust concentration C detected by the intake dust sensor t and dust particle size distribution parameter Lc t , these parameters are used as the characteristic parameters Fx of the flue gas in the current cycle t t , denoted as Fx t =(Q t RH t 、T t 、C t 、Lc t ); The parameter acquisition module then obtains the preset dust concentration target value C m ; S2, the parameter acquisition module collects the characteristic parameters Fx of the flue gas t , dust concentration target value C m Enter the parameter optimization model to obtain the optimized process parameters Gy' in the current cycle t t , Gy' t =(K' t 、P1' t 、f1' t 、P2' t 、f2' t 、U' t ); K' t Represents the optimized atomized droplet ratio coefficient, P1' t 、f1' t Respectively represent the power and frequency of the first ultrasonic transmitter obtained by optimization, P2' t 、f2' t Respectively represent the power and frequency of the optimized second ultrasonic transmitter, U' t Indicates the optimized working voltage of the electrostatic precipitator host; S3, the control module obtains the optimized process parameter Gy' according to the parameter optimization model t , control so that in the current cycle t: The ratio of the water flow rate in the water pipe to the flue gas flow rate in the flue gas inlet pipe is K' t , the power of the first ultrasonic transmitter is P1' t , frequency is f1' t , the power of the second ultrasonic transmitter is P2' t , frequency is f2' t , the working voltage of the electrostatic precipitator host is U' t ; S4. In the next cycle t+1, repeat steps S1-S3.

8. The dust removal device for a flue gas electric heater according to claim 7, characterized in that: The parameter optimization model is constructed by the following method: S1-1. Construct training dataset Z: S1-1-1. The flue gas is passed into the dust removal device for treatment. First, the flue gas flow rate Qw and the target dust concentration value C in the exhaust gas are fixed. m ; Get different relative humidity RH i , temperature T i , dust concentration C i And the dust particle size distribution parameter Lc i Under this condition, the actual dust concentration in the exhaust gas should not exceed the dust concentration target value C m When the total power consumption W reaches the minimum, the process parameter Gy i , the total power consumption at this time is W i ; Among them, the process parameter Gy i Including the atomized droplet ratio coefficient K i , the power P1 of the first ultrasonic transmitter i and frequency f1 i , the power of the second ultrasonic transmitter P2 i and frequency f2 i , Working voltage U of electrostatic precipitator host i ; Qw、C m RH i 、T i 、C i 、Lc i and the corresponding Gy i 、W i Combine to get training data z i , thereby obtaining the corresponding flue gas flow Qw and dust concentration target value C m The n training data under, i = 1, 2, ..., n, n = 10-100; S1-1-2. Change the value of the incoming flue gas flow Qw and the target value of the dust concentration C m , use the same method as step S1-1-1 to obtain m training data z j ,i=1,2,...,m,m=50-5000; S1-1-3, combine all the training data to obtain the training data set Z; S1-2. Model training: Using the training data set Z, Qw, C m RH i 、T i 、C i 、Lc i For access, Gy i The CNN convolutional neural network is trained for the target output, and the parameter optimization model is obtained after the training is completed.

9. The dust removal device for a flue gas electric heater according to claim 7, characterized in that: During the operation of the dust removal device, in each cycle t0 of the process parameter optimization model, the exhaust dust sensor collects N dust concentration values ​​in the exhaust gas, which are recorded as the actual dust concentration C S , N = 5-200; the parameter acquisition module obtains the detection results of the exhaust dust sensor and transmits them to the cleaning judgment module. The cleaning judgment module determines whether cleaning is required in each cycle t0 according to the following method: 1) Calculate the actual dust concentration C within the period t0 S Greater than the dust concentration target value C m The number of dust concentration data values ​​N1, if N1 / N ≥ 0.01 ~ 0.05, go to step 2), otherwise it is determined that dust cleaning is not required; 2) Calculate the total time of each cycle t0 as △t, and divide one cycle into four time periods: 0~0.25△t, 0.25△t~0.5△t, 0.5△t~0.75△t, 0.75△t~△t, which are recorded as time period 1, time period 2, time period 3, and time period 4 respectively. Calculate the actual dust concentration C in each time period S The average value of If satisfied If it is not, it is judged that cleaning is required; otherwise, cleaning is not required; 3) When the cleaning judgment module determines that cleaning is required, after the current cycle ends, the control module controls the cleaning operation in the following manner: controlling the gas flow controller to be closed, stopping the introduction of flue gas, controlling the water flow controller to be closed, stopping the spraying of atomized droplets, controlling the first ultrasonic transmitter to stop working, controlling the electrostatic precipitator host to stop working, and controlling the second ultrasonic transmitter to work, emitting ultrasonic waves to clean the discharge electrode and the dust collecting electrode; After the dust cleaning is completed, the first ultrasonic transmitter, the second ultrasonic transmitter and the electrostatic precipitator host are controlled to work according to the optimized process parameter Gy' obtained in the previous cycle, and the water flow controller is controlled to open, and finally the gas flow controller is controlled to open to re-introduce the flue gas.

Images