A desulfurization byproduct generation whole-process intelligent regulation and control system and method

Abstract

The present application relates to a kind of desulfurization by-product generation whole process intelligent regulation and control system and method, first establish operation database, based on database parameter identification model is established and based on real-time monitoring data correction parameter γ, Y, further establish desulfurization tower bottom slurry sulfite ion concentration accurate prediction model, cyclone underflow solid content accurate prediction model, vacuum belt dewaterer outlet gypsum water content accurate prediction model, and based on model to gypsum discharge pump, oxidizing fan, cyclone and vacuum belt dewaterer are carried out real-time regulation and control, then combine particle swarm optimization algorithm, by DSC controller realizes desulfurization by-product generation whole process regulation and control.The present application is under the condition of gypsum quality standard, to by-product generation process is carried out with low energy consumption as target global optimization regulation and control, guarantee the high efficiency of system operation, realize desulfurization by-product generation whole process low energy consumption high quality intelligent control.

Description

Technical Field

[0001] This invention belongs to the field of air pollutant control technology, specifically relating to an intelligent control system and method for the entire process of desulfurization byproduct generation under quality and energy consumption constraints. Background Technology

[0002] Currently, most coal-fired power plants use the limestone-gypsum method to remove sulfur dioxide (SO2) from flue gas. However, the quality of desulfurization byproducts in the limestone-gypsum desulfurization process varies with operating conditions, and relying solely on the experience and knowledge of operators can lead to significant fluctuations. Setting key parameters for critical equipment such as the gypsum discharge pump, the number / pressure of gypsum hydrocyclones, the speed of the belt dewatering machine, and the vacuum degree of the vacuum pump solely based on the experience and knowledge of operators can result in problems such as the gypsum quality not consistently meeting national standards and high energy consumption during equipment operation.

[0003] Chinese patent CN109459988 A provides a big data-based optimization control method for the dehydration system of a desulfurization unit. Based on real-time operating data and combined with big data analysis and calculation, it provides suggestions for the optimal operation and start-up / shutdown schemes of the gypsum discharge pump, achieving the goal of low-energy-consumption operation of the system. However, this method uses methods such as manual on-site inspection, remote video monitoring with manual inspection, and on-site sampling followed by laboratory testing to monitor the gypsum quality of the wet desulfurization system. Visual inspection relies on human experience to judge whether the gypsum quality meets the requirements, and the judgment process and results are difficult to quantify, have poor repeatability, and are limited by human energy and mental capacity, making continuous online monitoring impossible. On-site sampling followed by laboratory testing has high accuracy and good repeatability, but is time-consuming and can only be implemented intermittently. Chinese patent CN113694714 A provides an online monitoring method for slurry conditions and gypsum quality, which can systematically monitor the formation process of some by-products, but the equipment's energy consumption is relatively high under certain operating conditions, making it difficult to achieve industrial application. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, this invention provides an intelligent control system and method for the entire process of desulfurization byproduct generation (full-process monitoring and global optimization control). It establishes precise prediction models for sulfite ion concentration during byproduct generation, underflow solids content during cyclone dehydration, and gypsum moisture content during belt dehydration. Based on these models, it performs real-time control of the gypsum discharge pump, oxidation fan, cyclone separator, and vacuum belt dehydrator. Then, combined with particle swarm optimization algorithm, it achieves full-process control of desulfurization byproduct generation through a DSC controller. Under the constraint of achieving gypsum quality standards, this invention performs global optimization control of the byproduct generation process with low energy consumption as the goal, ensuring high system efficiency and achieving low-energy, high-quality intelligent control of the entire desulfurization byproduct generation process.

[0005] A fully intelligent control system for the generation of desulfurization byproducts is disclosed. The system comprises a desulfurization byproduct generation system and an intelligent control system. The desulfurization byproduct generation system includes a desulfurization tower, a gypsum discharge pump, a hydrocyclone, a vacuum belt dewatering machine, and an oxidation fan connected to the desulfurization tower. The desulfurization tower, gypsum discharge pump, hydrocyclone, and vacuum belt dewatering machine are connected sequentially. The vacuum belt dewatering machine is also connected to a gypsum flushing water pump and a vacuum pump. The intelligent control system includes an online monitoring module, a key parameter prediction module, and a key parameter control module.

[0006] The parameters detected by the online monitoring module include the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, hydrocyclone outlet slurry volume, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum quality at the vacuum belt dewatering machine outlet.

[0007] The flue gas flow rate, flue gas temperature, SO2 concentration, and O2 concentration at the inlet and outlet of the desulfurization tower are monitored by a flue gas analyzer; the slurry density at the bottom of the desulfurization tower is monitored by a densitometer; and the gypsum spreading thickness of the vacuum belt dewatering machine is monitored by a thickness gauge. The input terminals of the flue gas analyzer, densitometer, and thickness gauge are respectively connected to the flue gas duct at the inlet and outlet of the desulfurization tower, the slurry pool at the bottom of the desulfurization tower, and the vacuum belt dewatering machine, and the output terminals are respectively connected to the DCS controller.

[0008] The key parameter prediction module includes a precise prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, a precise prediction model for the solid content in the underflow of the hydrocyclone, and a precise prediction model for the gypsum quality at the outlet of the vacuum belt dewatering machine.

[0009] The parameters controlled by the key parameter control module include the control of the operating parameters of the gypsum discharge pump, the oxidation fan, the hydrocyclone, and the vacuum belt dewatering machine.

[0010] This invention also provides a method for intelligent control of the entire process of desulfurization byproduct generation, comprising the following steps:

[0011] (1) The state of desulfurization byproducts is corrected by constructing an accurate prediction model and combining historical operation and online monitoring data. The accurate prediction model includes an accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, an accurate prediction model of solid content in the underflow of the hydrocyclone, and an accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine.

[0012] (2) Further, based on the established accurate prediction model of sulfite ion concentration in the bottom slurry of the desulfurization tower, accurate prediction model of solid content in the underflow of the hydrocyclone, and accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine, the operating parameters of the gypsum discharge pump, oxidation fan, hydrocyclone, and vacuum belt dewatering machine are adjusted by combining the particle swarm optimization algorithm to achieve global optimization and control of desulfurization byproduct generation.

[0013] Preferably, the construction of the accurate prediction models for the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, the solid content in the underflow of the hydrocyclone, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine includes the following steps:

[0014] Step S1: Based on historical and online operation data, establish a multi-dimensional operation database covering the entire process of desulfurization byproduct generation and dewatering, including flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, slurry volume at the hydrocyclone outlet, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum moisture content at the outlet of the vacuum belt dewatering machine.

[0015] Step S2: Based on the database established in Step S1, for the SO2 absorption and removal process, based on the SO2 absorption and oxidation mechanism, analyze the response relationship between the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, slurry pH, slurry density, oxidation fan power and the sulfite ion concentration at the bottom of the desulfurization tower. Construct a prediction model for the sulfite ion concentration at the bottom of the desulfurization tower under different operating conditions. Further combine historical operating data and online monitoring data of the sulfite ion concentration at the bottom of the desulfurization tower to revise the prediction model, forming an accurate prediction model for the sulfite ion concentration at the bottom of the desulfurization tower that combines mechanism and data.

[0016] Step S3: Based on the database established in Step S1, for the hydrocyclone dewatering process, analyze the response relationship between the gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones and the solid content of the hydrocyclone underflow, establish a prediction model for the solid content of the hydrocyclone underflow under different operating conditions, and further combine historical operating data and online monitoring data of the solid content of the hydrocyclone underflow to revise the prediction model of the solid content of the hydrocyclone underflow, forming a knowledge- and data-driven accurate prediction model for the solid content of the hydrocyclone underflow.

[0017] Step S4: Based on the database established in Step S1, analyze the response relationship between the slurry volume at the hydrocyclone outlet, the solid content of the hydrocyclone underflow, the speed of the vacuum belt dewatering machine, the vacuum pump pressure, the power of the gypsum flushing water pump, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine for the belt dewatering process. Establish a prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine under different operating conditions. Further combine the historical operating data and online monitoring data of the quality of gypsum at the outlet of the vacuum belt dewatering machine to revise the prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine, forming a knowledge- and data-driven accurate prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine.

[0018] Preferably, step S2 specifically includes:

[0019] Step S2.1: Based on the SO2 absorption and oxidation mechanism, establish models for the natural oxidation process and forced oxidation process of the desulfurization oxidation system:

[0020]

[0021]

[0022] In the formula, It is the natural oxidizing factor of SO2 in the desulfurization tower; It is a forced oxidizing agent for SO2 in the desulfurization tower; Q is the molar concentration of SO2 in the flue gas at the inlet of the desulfurization tower; Q is the flow rate of the flue gas at the inlet of the desulfurization tower. q represents the molar concentration of O2 in the flue gas at the inlet of the desulfurization tower; q represents the amount of spray slurry in the desulfurization tower. Slurry sprayed on the desulfurization tower value; Where W is the density of the spray slurry in the desulfurization tower, and W is the operating power of the oxidation fan.

[0023] Step S2.2: Based on the models of natural oxidation and forced oxidation processes in the desulfurization oxidation system, establish models for the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time.

[0024]

[0025]

[0026] In the formula, This refers to the amount of SO2 absorbed per unit time. For a unit of time period; SO2 removal efficiency; The amount of SO2 oxidized per unit time; Introduce airflow to the oxidation blower; The oxidation fan introduces the O2 molar concentration from the air;

[0027] Step S2.3: Based on the models of the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time, establish a prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower:

[0028]

[0029] Substituting the models of SO2 moles absorbed per unit time and SO2 moles oxidized per unit time into the prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, we obtain:

[0030]

[0031] In the formula, The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower; for The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower is constantly monitored.

[0032] Step S2.4: Based on historical operating data and online sampling detection data of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, revise the prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower, forming an accurate prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower that combines mechanism and data.

[0033]

[0034] In the formula, To correct the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, This is a correction factor;

[0035] In the prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower Parameter identification is performed (by establishing a correction parameter γ through a parameter identification model). The parameter identification process can be represented by an optimization problem, namely:

[0036]

[0037] In the formula, The root mean square error between the model-predicted value and the actual measured value of sulfite ion concentration in the slurry pool at the bottom of the desulfurization tower; This is the actual measured value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower; This refers to the number of slurry samples taken from the slurry pool at the bottom of the desulfurization tower.

[0038] Preferably, step S3 specifically includes:

[0039] Step S3.1: Establish a hydrocyclone separation efficiency model based on the hydrocyclone inlet pressure, slurry flow rate at the gypsum discharge pump outlet, number of hydrocyclones, hydrocyclone structural dimensions, slurry density at the gypsum discharge pump outlet, and sulfite ion concentration in the slurry at the gypsum discharge pump outlet.

[0040]

[0041] In the formula, For hydrocyclone separation efficiency; P This refers to the inlet pressure of the hydrocyclone. This refers to the slurry flow rate at the outlet of the gypsum discharge pump. n is the number of cyclones; D is the structural dimension of the cyclone separator. The density of the slurry at the outlet of the gypsum discharge pump;

[0042] Step S3.2: Based on the hydrocyclone separation efficiency, combined with the hydrocyclone outlet pressure drop, the volume percentage of solids in the feed material, and the specific gravity of the feed material, the separation efficiency model is corrected. The corrected hydrocyclone separation efficiency model is as follows:

[0043]

[0044] In the formula, To correct the separation efficiency of the hydrocyclone; The pressure drop at the hydrocyclone outlet is P* = P, since the hydrocyclone dehydration process takes place under atmospheric pressure. is the volume percentage of solids in the hydrocyclone feed; s is the specific gravity of the solids in the feed.

[0045] Step S3.3: Based on the corrected hydrocyclone separation efficiency, and combined with the slurry flow rate at the gypsum discharge pump outlet, slurry density, and sulfite ion concentration in the slurry at the bottom of the desulfurization tower, establish a knowledge- and data-driven accurate prediction model for the solids content of the hydrocyclone underflow:

[0046]

[0047] In the formula, m This refers to the solids content of the underflow from the hydrocyclone.

[0048] Preferably, step S4 specifically includes:

[0049] Step S4.1: Based on the slurry volume at the hydrocyclone outlet and the rotational speed of the vacuum belt dewatering machine, establish a formula for calculating the slurry thickness of the vacuum belt dewatering machine:

[0050]

[0051] In the formula, h is the slurry thickness of the vacuum belt dewatering machine; The rotational speed of the vacuum belt dewatering machine; b is the slurry volume at the hydrocyclone outlet; b is the filter cloth width of the vacuum belt dewatering machine.

[0052] Step S4.2: Based on the slurry thickness of the vacuum belt dewatering machine, the slurry volume at the hydrocyclone outlet, the solids content of the slurry at the hydrocyclone outlet, the speed of the vacuum belt dewatering machine, the vacuum pump pressure, and the spray liquid volume during the belt dewatering process, establish a prediction model for the dewatering amount during the belt dewatering process:

[0053]

[0054] In the formula, This refers to the amount of water removed during the belt dehydration process. p For vacuum pump pressure, This refers to the amount of spray liquid used during the belt dehydration process.

[0055] Step S4.3: Based on historical operational data and online sampling detection data of the dewatering process in the belt dewatering process, revise the dewatering amount prediction model for the belt dewatering process to form an accurate prediction model for the dewatering amount of the belt dewatering process that combines mechanism and data.

[0056]

[0057] In the formula, Y is the correction factor for the amount of water removed during the belt dehydration process.

[0058] In the prediction model of water removal amount in the belt dewatering process Parameter identification is performed (by establishing the corrected parameters Y through a parameter identification model). The parameter identification process can be represented by an optimization problem, namely:

[0059]

[0060] In the formula, The root mean square error between the model-predicted value and the actual measured value of water removal during the belt dewatering process; y represents the actual measured value of water removal during the belt dehydration process; y represents the number of samples of water removal during the belt dehydration process.

[0061] Step S4.4: Based on the modified belt dewatering process dewatering volume model, and combined with the hydrocyclone outlet slurry volume, hydrocyclone outlet slurry solids content, and belt dewatering process spray volume, establish a precise prediction model for the gypsum moisture content at the vacuum belt dewatering machine outlet, combining mechanism and data.

[0062]

[0063] In the formula, The moisture content of the gypsum at the outlet of the vacuum belt dewatering machine.

[0064] As a preferred method, the operating parameters of the gypsum discharge pump are controlled as follows: based on the real-time measurement data of the slurry density meter at the bottom of the desulfurization tower, the gypsum discharge pump start and stop commands are sent to the DCS controller through the data transmission module, and then the DCS controller sends the commands to the field equipment.

[0065] DCS controller commands include:

[0066] When the density of the gypsum reaches its upper limit, it is recommended to start the gypsum discharge pump.

[0067] When the density of the plaster reaches the lower limit, it is recommended to shut down the plaster discharge pump.

[0068] As a preferred option, the control of the oxidation blower operating parameters is as follows: by using the established accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is calculated. Furthermore, by combining the real-time operating data of flue gas volume, temperature, SO2 concentration, and oxidation blower power at the desulfurization tower inlet, the oxidation blower power command is adjusted through an optimized control strategy and sent to the DCS controller through the data transmission module. The DCS controller then sends the command to the field equipment.

[0069] The predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is used as the given instruction, and the flue gas volume, temperature, SO2 concentration, and oxidation fan power at the inlet of the desulfurization tower are used as the feedforward instructions.

[0070] The optimized control strategy involves setting the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, selecting the method with the lowest oxidation fan power, and setting the upper and lower limits of the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower to ensure that the gypsum quality meets the standards.

[0071] DCS controller commands include:

[0072] When the sulfite ion concentration reaches the lower limit, it is recommended to reduce the power of the oxidation fan.

[0073] When the sulfite ion concentration reaches the upper limit, it is recommended to increase the power of the oxidation fan.

[0074] As a preferred option, the control of the hydrocyclone operating parameters is as follows: by using the established accurate prediction models of sulfite ion concentration in the slurry at the bottom of the desulfurization tower and the accurate prediction models of solid content in the hydrocyclone underflow, the predicted values ​​of sulfite ion concentration in the slurry at the bottom of the desulfurization tower and solid content in the hydrocyclone underflow are calculated. Furthermore, combined with the real-time operating data of slurry flow rate at the outlet of the gypsum discharge pump and hydrocyclone pressure, the hydrocyclone is adjusted through an optimized control strategy. The instructions for adding or removing the hydrocyclone are sent to the DCS controller through the data transmission module, and then the DCS controller sends the instructions to the field equipment.

[0075] The predicted solids content in the hydrocyclone underflow is used as a given command, while the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the slurry flow rate at the outlet of the gypsum discharge pump, and the hydrocyclone pressure are used as feedforward commands.

[0076] DCS controller commands include:

[0077] When the solids content in the underflow of a hydrocyclone reaches the lower limit, it is recommended to increase the number of hydrocyclones.

[0078] When the solids content in the underflow of a hydrocyclone reaches its upper limit, a suggestion is given to reduce the number of hydrocyclones.

[0079] As a preferred option, the control of the operating parameters of the vacuum belt dewatering machine is as follows: using the established accurate prediction models for the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, the solid content in the underflow of the hydrocyclone, and the gypsum quality at the outlet of the vacuum belt dewatering machine, the predicted values ​​of the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, the solid content in the underflow of the hydrocyclone, and the gypsum quality at the outlet of the vacuum belt dewatering machine are calculated. Further, combined with the slurry volume at the outlet of the hydrocyclone and the pressure of the vacuum pump, the multi-model prediction optimization control strategy is used to adjust the speed command of the vacuum belt dewatering machine and the power command of the gypsum flushing water pump. The command is then sent to the DCS controller through the data communication module, and the DCS controller sends the command to the field equipment.

[0080] The predicted value of gypsum quality at the outlet of the vacuum belt dewatering machine is used as the given instruction, and the predicted value of sulfite ion concentration at the bottom of the desulfurization tower, the predicted solid content of the hydrocyclone underflow, the slurry volume at the hydrocyclone outlet, and the vacuum pump pressure are used as feedforward instructions.

[0081] The optimization and control strategy involves setting the minimum speed of the vacuum belt dewatering machine and the minimum power of the gypsum flushing pump under the condition of setting the gypsum quality at the outlet of the vacuum belt dewatering machine, and setting the constraint conditions for the gypsum quality at the outlet of the vacuum belt dewatering machine to ensure that the gypsum quality meets the standards.

[0082] The quality constraints for gypsum are as follows:

[0083]

[0084]

[0085]

[0086]

[0087] In the formula, The chloride ion content of the gypsum at the outlet of the vacuum belt dewatering machine; pH is the pH value of the gypsum at the outlet.

[0088] DCS controller commands include:

[0089] When the slurry spreading thickness reaches the upper limit, the gypsum moisture content reaches the upper limit, and the gypsum chloride ion content reaches the lower limit, it is recommended to increase the speed of the belt dewatering machine and increase the power of the gypsum flushing water pump.

[0090] When the slurry spreading thickness reaches the lower limit, the gypsum moisture content reaches the lower limit, and the gypsum chloride ion content reaches the upper limit, it is recommended to reduce the speed of the belt dewatering machine and reduce the power of the gypsum flushing water pump.

[0091] As a preferred option, the adjustment process combined with the particle swarm optimization algorithm is as follows:

[0092] Global optimization and control are carried out with the goal of low energy consumption, i.e., optimal particle fitness.

[0093]

[0094] in, , These represent the costs of the by-product generation process, the cyclone dehydration process, and the belt dehydration process, respectively.

[0095] The cost of byproduct generation includes the power consumed by the oxidation blower. O And the power consumption generated by the operation of the plaster discharge pump P The cost of the cyclone dehydration process includes the energy loss due to friction loss. b The cost of the belt dewatering process includes the power consumption of the belt dewatering machine. B Power consumption of vacuum pump V and the power consumption of the belt flushing water pump W ;Right now:

[0096]

[0097]

[0098]

[0099] The optimal particle fitness is:

[0100]

[0101] The beneficial effects of this invention are as follows:

[0102] (1) In response to the frequent fluctuations in flue gas composition, SO2 concentration, and temperature of coal-fired power plants under different operating conditions, this invention provides an intelligent control method for the entire wet desulfurization process. By establishing a prediction model and combining historical operation and online monitoring data to correct the state of desulfurization byproducts, and further based on the accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the accurate prediction model of solid content in the underflow of the hydrocyclone, and the accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine, the operating parameters of the oxidation blower, gypsum discharge pump, hydrocyclone, vacuum belt dewatering machine, and gypsum flushing water pump are adjusted by combining the particle swarm optimization algorithm, so as to achieve global optimization control of desulfurization byproduct generation.

[0103] (2) Based on the goal of low energy consumption operation of equipment, this invention combines particle swarm optimization algorithm to carry out high-quality low-energy intelligent control of the entire process of desulfurization by-product generation. While ensuring the quality of gypsum, it realizes low-energy operation of desulfurization equipment. This invention proves that 23.7% of energy consumption can be saved in the by-product generation stage of wet desulfurization process of 1000MW coal-fired power generation unit, and the economic benefits are quite significant.

[0104] (3) This invention uses intelligent algorithms to analyze and control the oxidation fan, gypsum discharge pump, hydrocyclone and vacuum dehydration belt conveyor to achieve optimal gypsum quality control, thereby greatly reducing the workload of operators, making the whole system more stable, the gypsum quality higher, the energy consumption lower, and improving economic benefits. Attached Figure Description

[0105] Figure 1 This is a diagram of the desulfurization system of the present invention;

[0106] Figure 2 Intelligent control flow chart for the entire process of desulfurization by-product generation;

[0107] Figure 3 This is a graph showing the relationship between unit load and natural oxidation rate.

[0108] Figure 4 This is a graph showing the relationship between unit load and oxidation air demand. Detailed Implementation

[0109] The technical solution of the present invention will be further described in detail below through embodiments. These embodiments are for illustrative purposes only and are not intended to limit the present invention. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0110] Example 1

[0111] Reference Figure 1A fully intelligent control system for the entire process of desulfurization byproduct generation is disclosed. The system comprises a desulfurization byproduct generation system and an intelligent control system. The desulfurization byproduct generation system includes a desulfurization tower 1, a gypsum discharge pump 2, a hydrocyclone 3, a vacuum belt dewatering machine 4, and an oxidation fan 5 connected to the desulfurization tower. The desulfurization tower 1, gypsum discharge pump 2, hydrocyclone 3, and vacuum belt dewatering machine 4 are connected sequentially. The vacuum belt dewatering machine 4 is also connected to a gypsum flushing water pump 6 and a vacuum pump 7. The intelligent control system includes an online monitoring module, a key parameter prediction module, and a key parameter control module.

[0112] The parameters detected by the online monitoring module include the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, hydrocyclone outlet slurry volume, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum quality at the vacuum belt dewatering machine outlet.

[0113] The flue gas flow rate, flue gas temperature, SO2 concentration, and O2 concentration at the inlet and outlet of the desulfurization tower are monitored by a flue gas analyzer 8; the slurry density at the bottom of the desulfurization tower is monitored by a densitometer 9; and the gypsum spreading thickness of the vacuum belt dewatering machine is monitored by a thickness gauge 10. The input terminals of the flue gas analyzer 8, densitometer 9, and thickness gauge 10 are respectively connected to the inlet and outlet flue of the desulfurization tower 1, the slurry pool at the bottom of the desulfurization tower, and the vacuum belt dewatering machine 4, and their output terminals are respectively connected to the DCS controller 11.

[0114] The key parameter prediction module includes a precise prediction model for the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, a precise prediction model for the solid content in the underflow of the hydrocyclone, and a precise prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine.

[0115] The parameters controlled by the key parameter control module include the control of the gypsum discharge pump operating parameters (gypsum discharge pump start-stop control), the control of the oxidation fan operating parameters (oxidation fan power control), the control of the hydrocyclone operating parameters (hydrocyclone input and cut-off control), and the control of the vacuum belt dewatering machine operating parameters (vacuum belt dewatering machine speed control, gypsum flushing water pump power control).

[0116] Reference Figure 2 A method for intelligent control of the entire process of desulfurization byproduct generation, comprising the following steps:

[0117] (1) The state of desulfurization byproducts is corrected by constructing an accurate prediction model and combining historical operation and online monitoring data. The accurate prediction model includes an accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, an accurate prediction model of solid content in the underflow of the hydrocyclone, and an accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine.

[0118] (2) Further, based on the established accurate prediction model of sulfite ion concentration in the bottom slurry of the desulfurization tower, accurate prediction model of solid content in the underflow of the hydrocyclone, and accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine, the operating parameters of the gypsum discharge pump, oxidation fan, hydrocyclone, and vacuum belt dewatering machine are adjusted by combining the particle swarm optimization algorithm to achieve global optimization and control of desulfurization byproduct generation.

[0119] The construction of the accurate prediction models for the sulfite ion concentration in the bottom slurry of the desulfurization tower, the solid content in the underflow of the hydrocyclone, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine includes the following steps:

[0120] Step S1: Based on historical and online operation data, establish a multi-dimensional operation database covering the entire process of desulfurization byproduct generation and dewatering, including flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, slurry volume at the hydrocyclone outlet, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum moisture content at the outlet of the vacuum belt dewatering machine.

[0121] Step S2: Based on the database established in Step S1, for the SO2 absorption and removal process, based on the SO2 absorption and oxidation mechanism, analyze the response relationship between the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, slurry pH, slurry density, oxidation fan power and the sulfite ion concentration at the bottom of the desulfurization tower. Construct a prediction model for the sulfite ion concentration at the bottom of the desulfurization tower under different operating conditions. Further combine historical operating data and online monitoring data of the sulfite ion concentration at the bottom of the desulfurization tower to revise the prediction model, forming an accurate prediction model for the sulfite ion concentration at the bottom of the desulfurization tower that combines mechanism and data.

[0122] Step S3: Based on the database established in Step S1, for the hydrocyclone dewatering process, analyze the response relationship between the gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones and the solid content of the hydrocyclone underflow, establish a prediction model for the solid content of the hydrocyclone underflow under different operating conditions, and further combine historical operating data and online monitoring data of the solid content of the hydrocyclone underflow to revise the prediction model of the solid content of the hydrocyclone underflow, forming a knowledge- and data-driven accurate prediction model for the solid content of the hydrocyclone underflow.

[0123] Step S4: Based on the database established in Step S1, analyze the response relationship between the slurry volume at the hydrocyclone outlet, the solid content of the hydrocyclone underflow, the speed of the vacuum belt dewatering machine, the vacuum pump pressure, the power of the gypsum flushing water pump, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine for the belt dewatering process. Establish a prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine under different operating conditions. Further combine the historical operating data and online monitoring data of the quality of gypsum at the outlet of the vacuum belt dewatering machine to revise the prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine, forming a knowledge- and data-driven accurate prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine.

[0124] Furthermore, the establishment of a precise prediction model for the sulfite ion concentration in the bottom slurry of the desulfurization tower that combines mechanism and data, as described in step S2, specifically includes:

[0125] Step S2.1: Based on the SO2 absorption and oxidation mechanism, establish models for the natural oxidation process and forced oxidation process of the desulfurization oxidation system:

[0126]

[0127]

[0128] In the formula, It is the natural oxidizing factor of SO2 in the desulfurization tower; It is a forced oxidizing agent for SO2 in the desulfurization tower; Q is the molar concentration of SO2 in the flue gas at the inlet of the desulfurization tower; Q is the flow rate of the flue gas at the inlet of the desulfurization tower. q represents the molar concentration of O2 in the flue gas at the inlet of the desulfurization tower; q represents the amount of spray slurry in the desulfurization tower. Slurry sprayed on the desulfurization tower value; Where W is the density of the spray slurry in the desulfurization tower, and W is the operating power of the oxidation fan.

[0129] Step S2.2: Based on the models of natural oxidation and forced oxidation processes in the desulfurization oxidation system, establish models for the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time.

[0130]

[0131]

[0132] In the formula, This refers to the amount of SO2 absorbed per unit time. For a unit of time period; SO2 removal efficiency; The amount of SO2 oxidized per unit time; Introduce airflow to the oxidation blower; The oxidation fan introduces the O2 molar concentration from the air;

[0133] Step S2.3: Based on the models of the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time, establish a prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower:

[0134]

[0135] Substituting the models of SO2 moles absorbed per unit time and SO2 moles oxidized per unit time into the prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, we obtain:

[0136]

[0137] In the formula, The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower; for The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower is constantly monitored.

[0138] Step S2.4: Based on historical operating data and online sampling detection data of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, revise the prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower, forming an accurate prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower that combines mechanism and data.

[0139]

[0140] In the formula, To correct the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, This is a correction factor;

[0141] Furthermore, in the prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower... Parameter identification is performed, and the process of parameter identification can be represented by an optimization problem, namely:

[0142]

[0143] In the formula, The root mean square error between the model-predicted value and the actual measured value of sulfite ion concentration in the slurry pool at the bottom of the desulfurization tower; This is the actual measured value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower; This refers to the number of slurry samples taken from the slurry pool at the bottom of the desulfurization tower.

[0144] Furthermore, the establishment of a knowledge- and data-driven accurate prediction model for solids content in hydrocyclone underflow, as described in step S3, specifically includes:

[0145] Step S3.1: Establish a hydrocyclone separation efficiency model based on the hydrocyclone inlet pressure, slurry flow rate at the gypsum discharge pump outlet, number of hydrocyclones, hydrocyclone structural dimensions, slurry density at the gypsum discharge pump outlet, and sulfite ion concentration in the slurry at the gypsum discharge pump outlet.

[0146]

[0147] In the formula, For hydrocyclone separation efficiency; P This refers to the inlet pressure of the hydrocyclone. This refers to the slurry flow rate at the outlet of the gypsum discharge pump. n The number of cyclotrons; These are the structural dimensions of the hydrocyclone; The density of the slurry at the outlet of the gypsum discharge pump;

[0148] Step S3.2: Based on the hydrocyclone separation efficiency, combined with the hydrocyclone outlet pressure drop, the volume percentage of solids in the feed material, and the specific gravity of the feed material, the separation efficiency model is corrected. The corrected hydrocyclone separation efficiency model is as follows:

[0149]

[0150] In the formula, To correct the separation efficiency of the hydrocyclone; The pressure drop at the hydrocyclone outlet is P* = P, since the hydrocyclone dehydration process takes place under atmospheric pressure. is the volume percentage of solids in the hydrocyclone feed; s is the specific gravity of the solids in the feed.

[0151] Step S3.3: Based on the corrected hydrocyclone separation efficiency, and combined with the slurry flow rate at the gypsum discharge pump outlet, slurry density, and sulfite ion concentration in the slurry at the bottom of the desulfurization tower, establish a knowledge- and data-driven accurate prediction model for the solids content of the hydrocyclone underflow:

[0152]

[0153] In the formula, m This refers to the solids content of the underflow from the hydrocyclone.

[0154] Furthermore, the establishment of a knowledge- and data-driven prediction model for the moisture content of gypsum at the outlet of the vacuum belt dewatering machine, as described in step S4, specifically includes:

[0155] Step S4.1: Based on the slurry volume at the hydrocyclone outlet and the rotational speed of the vacuum belt dewatering machine, establish a formula for calculating the slurry thickness of the vacuum belt dewatering machine:

[0156]

[0157] In the formula, h is the slurry thickness of the vacuum belt dewatering machine; The rotational speed of the vacuum belt dewatering machine; b is the slurry volume at the hydrocyclone outlet; b is the filter cloth width of the vacuum belt dewatering machine.

[0158] Step S4.2: Based on the slurry thickness of the vacuum belt dewatering machine, the slurry volume at the hydrocyclone outlet, the solids content of the slurry at the hydrocyclone outlet, the speed of the vacuum belt dewatering machine, the vacuum pump pressure, and the spray liquid volume during the belt dewatering process, establish a prediction model for the dewatering amount during the belt dewatering process:

[0159]

[0160] In the formula, This refers to the amount of water removed during the belt dehydration process. p For vacuum pump pressure, This refers to the amount of spray liquid used during the belt dehydration process.

[0161] Step S4.3: Based on historical operational data and online sampling detection data of the dewatering process in the belt dewatering process, revise the dewatering amount prediction model for the belt dewatering process to form an accurate prediction model for the dewatering amount of the belt dewatering process that combines mechanism and data.

[0162]

[0163] In the formula, Y is the correction factor for the amount of water removed during the belt dehydration process.

[0164] Furthermore, in the prediction model for the amount of water removed during the belt dewatering process... Parameter identification is performed, and the process of parameter identification can be represented by an optimization problem, namely:

[0165]

[0166] In the formula, The root mean square error between the model-predicted value and the actual measured value of water removal during the belt dewatering process; y represents the actual measured value of water removal during the belt dehydration process; y represents the number of samples of water removal during the belt dehydration process.

[0167] Step S4.4: Based on the modified belt dewatering process dewatering volume model, and combined with the hydrocyclone outlet slurry volume, hydrocyclone outlet slurry solids content, and belt dewatering process spray volume, establish a precise prediction model for the gypsum moisture content at the vacuum belt dewatering machine outlet, combining mechanism and data.

[0168]

[0169] In the formula, The moisture content of the gypsum at the outlet of the vacuum belt dewatering machine.

[0170] Furthermore, the control of the operating parameters of the gypsum discharge pump (gypsum discharge pump start-stop control) is as follows: based on the real-time measurement data of the slurry density meter at the bottom of the desulfurization tower, after optimization and adjustment of the control strategy, the start-stop command of the gypsum discharge pump is sent to the DCS controller through the data transmission module, and then the DCS controller sends the command to the field equipment.

[0171] DCS controller commands include:

[0172] When the density of the gypsum reaches its upper limit, it is recommended to start the gypsum discharge pump.

[0173] When the density of the plaster reaches the lower limit, it is recommended to shut down the plaster discharge pump.

[0174] Furthermore, the control of the operating parameters of the oxidation blower (oxidation blower power control) is as follows: using the established accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is calculated. In combination with the real-time operating data of flue gas volume, temperature, SO2 concentration, and oxidation blower power at the desulfurization tower inlet, the oxidation blower power command is adjusted through an optimized control strategy and sent to the DCS controller through the data transmission module. The DCS controller then sends the command to the field equipment.

[0175] The predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is used as the given instruction, and the flue gas volume, temperature, SO2 concentration, and oxidation fan power at the inlet of the desulfurization tower are used as the feedforward instructions.

[0176] The optimized control strategy involves selecting the lowest possible oxidation fan power while setting the sulfite ion concentration at the bottom of the desulfurization tower. However, frequent adjustments to the oxidation fan power during actual operation can significantly impact the equipment's lifespan. Furthermore, to prevent substandard gypsum quality due to untimely adjustments to the oxidation fan power, upper and lower limits for the sulfite ion concentration at the bottom of the desulfurization tower are set to ensure gypsum quality meets standards.

[0177] DCS controller commands include:

[0178] When the sulfite ion concentration reaches the lower limit, it is recommended to reduce the power of the oxidation fan.

[0179] When the sulfite ion concentration reaches the upper limit, it is recommended to increase the power of the oxidation fan.

[0180] Furthermore, the control of hydrocyclone operating parameters (hydrocyclone input and output control) is as follows: using the established accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower and the accurate prediction model of solid content in the hydrocyclone underflow, the predicted values ​​of sulfite ion concentration in the slurry at the bottom of the desulfurization tower and solid content in the hydrocyclone underflow are calculated. In addition, combined with the real-time operating data of slurry flow rate at the gypsum discharge pump outlet and hydrocyclone pressure, the hydrocyclone input and output commands are sent to the DCS controller through the data transmission module after optimization and adjustment strategies. The DCS controller then sends the commands to the field equipment.

[0181] The predicted solids content in the hydrocyclone underflow is used as a given command, while the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the slurry flow rate at the outlet of the gypsum discharge pump, and the hydrocyclone pressure are used as feedforward commands.

[0182] DCS controller commands include:

[0183] When the solids content in the underflow of a hydrocyclone reaches the lower limit, it is recommended to increase the number of hydrocyclones.

[0184] When the solids content in the underflow of a hydrocyclone reaches its upper limit, a suggestion is given to reduce the number of hydrocyclones.

[0185] Furthermore, the control of the operating parameters of the vacuum belt dewatering machine (vacuum belt dewatering machine speed control and gypsum flushing water pump power control) is as follows: Based on the established accurate prediction models of sulfite ion concentration at the bottom of the desulfurization tower, solid content at the underflow of the hydrocyclone, and gypsum quality at the outlet of the vacuum belt dewatering machine, the predicted values ​​of sulfite ion concentration at the bottom of the desulfurization tower, solid content at the underflow of the hydrocyclone, and gypsum quality at the outlet of the vacuum belt dewatering machine are calculated. In combination with the slurry volume at the hydrocyclone outlet and the vacuum pump pressure, the vacuum belt dewatering machine speed command and gypsum flushing water pump power command are adjusted through a multi-model prediction optimization control strategy. The data communication module sends the command to the DCS controller, and the DCS controller then sends the command to the field equipment.

[0186] The predicted value of gypsum quality at the outlet of the vacuum belt dewatering machine is used as the given instruction, and the predicted value of sulfite ion concentration at the bottom of the desulfurization tower, the predicted solid content of the hydrocyclone underflow, the slurry volume at the hydrocyclone outlet, and the vacuum pump pressure are used as feedforward instructions.

[0187] The optimized control strategy involves selecting the lowest possible speed of the vacuum belt dewatering machine and the lowest possible power for the gypsum flushing pump, while setting the desired gypsum quality at the outlet. However, frequent adjustments to these parameters during actual operation can significantly impact the equipment's lifespan. Furthermore, to prevent substandard gypsum quality due to untimely adjustments to these parameters, a constraint condition for the gypsum quality at the outlet of the vacuum belt dewatering machine is set to ensure that the gypsum quality meets the required standards.

[0188] The quality constraints for gypsum are as follows:

[0189]

[0190]

[0191]

[0192]

[0193] In the formula, Chloride ion content of gypsum at the outlet of the vacuum belt dewatering machine; PH pH value of the exported gypsum;

[0194] DCS controller commands include:

[0195] When the slurry spreading thickness reaches the upper limit, the gypsum moisture content reaches the upper limit, and the gypsum chloride ion content reaches the lower limit, it is recommended to increase the speed of the belt dewatering machine and increase the power of the gypsum flushing water pump.

[0196] When the slurry spreading thickness reaches the lower limit, the gypsum moisture content reaches the lower limit, and the gypsum chloride ion content reaches the upper limit, it is recommended to reduce the speed of the belt dewatering machine and reduce the power of the gypsum flushing water pump.

[0197] The process of adjusting the Particle Swarm Optimization algorithm is as follows:

[0198] Global optimization and control are carried out with the goal of low energy consumption, i.e., optimal particle fitness.

[0199]

[0200] in, , These represent the costs of the by-product generation process, the cyclone dehydration process, and the belt dehydration process, respectively.

[0201] The cost of byproduct generation includes the power consumed by the oxidation blower. O Power consumption generated by the operation of the gypsum discharge pump P The cost of the cyclone dehydration process includes the energy loss due to friction loss. b The cost of the belt dewatering process includes the power consumption of the belt dewatering machine. B Power consumption of vacuum pump V and the power consumption of the belt flushing water pump W ;Right now:

[0202]

[0203]

[0204]

[0205] The fitness of the optimal example is:

[0206]

[0207] In the intelligent control process of the entire desulfurization by-product generation process of this invention, an operational database is first established. Based on the database, a parameter identification model is built, and parameters γ and Y are corrected based on real-time monitoring data. Furthermore, knowledge- and data-driven desulfurization by-product generation models are established (accurate prediction models for sulfite ion concentration in the slurry at the bottom of the desulfurization tower, solids content in the underflow of the hydrocyclone, and gypsum moisture content at the outlet of the vacuum belt dewatering machine). Then, combined with a particle swarm optimization algorithm, the entire process of desulfurization by-product generation is controlled through a DSC controller. The data correction in this invention uses a particle swarm optimization algorithm (PSO algorithm). First, parameters γ and Y are initialized, and then the RMSE (evaluation criterion calculation) is calculated based on real-time operational data. The optimal position is then updated. If the obtained RMSE is minimized, the data correction ends; otherwise, the parameters are updated, and the RMSE calculation is repeated until the RMSE is minimized.

[0208] Example 2

[0209] An optimization operation test of the oxidation system was conducted on the desulfurization unit of a 1000MW unit. The relationship between unit load and natural oxidation rate is as follows: Figure 3 Under high load conditions (above 800MW), the natural oxidation rate of the desulfurization tower is low, less than 15%. Under medium and low load conditions (below 800MW), the natural oxidation rate of the desulfurization tower can reach up to 30%. The relationship between unit load and oxidation air demand is as follows: Figure 4 As shown, the oxidation air demand reaches 220 m³ / h under high load. 3 / min or above, the minimum oxidation air demand is only 107 m³. 3 / min.

[0210] The unit employs three control strategies (two parallel groups) for the oxidation blower, gypsum discharge pump, cyclone dewatering equipment, belt dewatering machine, vacuum pump, and belt flushing water pump. The first strategy is rated power operation, where the DCS does not control the equipment, allowing it to operate at its rated frequency continuously. The second strategy is an optimization strategy based on equipment operating experience. The DCS switches the equipment's operating frequency based on the operator's experience and the unit load. When the unit load is low to medium, the equipment's operating frequency is reduced; when the unit load is high, the equipment operates at its rated frequency. The third strategy is operation using the intelligent control method described in Example 1. The energy consumption comparison analysis results of different operating strategies are shown in Table 1. Based on the premise of ensuring the quality of the desulfurization byproduct gypsum, compared to operating at rated power, the strategy optimized based on operating experience can save 4.2% of energy consumption, while the strategy using the intelligent control method can save 23.7% of energy consumption.

[0211] Table 1. Comparative Analysis of Energy Consumption under Different Operating Strategies

[0212]

[0213] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for intelligent control of the entire process of desulfurization byproduct generation, characterized in that: The control system includes a desulfurization byproduct generation system and an intelligent control system. The desulfurization byproduct generation system includes a desulfurization tower, a gypsum discharge pump, a hydrocyclone, a vacuum belt dewatering machine, and an oxidation fan connected to the desulfurization tower. The desulfurization tower, gypsum discharge pump, hydrocyclone, and vacuum belt dewatering machine are connected in sequence. The vacuum belt dewatering machine is also connected to a gypsum flushing water pump and a vacuum pump. The intelligent control system includes an online monitoring module, a key parameter prediction module, and a key parameter control module. The parameters detected by the online monitoring module include the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, hydrocyclone outlet slurry volume, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum quality at the vacuum belt dewatering machine outlet. The key parameter prediction module includes a precise prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, a precise prediction model for the solid content in the underflow of the hydrocyclone, and a precise prediction model for the gypsum quality at the outlet of the vacuum belt dewatering machine. The parameters controlled by the key parameter control module include the control of the operating parameters of the gypsum discharge pump, the oxidation fan, the hydrocyclone, and the vacuum belt dewatering machine. The method includes the following steps: (1) The state of desulfurization byproducts is corrected by constructing an accurate prediction model and combining historical operation and online monitoring data. The accurate prediction model includes an accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, an accurate prediction model of solid content in the underflow of the hydrocyclone, and an accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine. (2) Further based on the established accurate prediction model of sulfite ion concentration in the bottom slurry of the desulfurization tower, accurate prediction model of solid content in the underflow of the hydrocyclone, and accurate prediction model of gypsum moisture content at the outlet of the vacuum belt dewatering machine, the operating parameters of the gypsum discharge pump, oxidation fan, hydrocyclone, and vacuum belt dewatering machine are adjusted by combining the particle swarm optimization algorithm to achieve global optimization and control of desulfurization byproduct generation. The construction of the accurate prediction models for the sulfite ion concentration in the bottom slurry of the desulfurization tower, the solid content in the underflow of the hydrocyclone, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine includes the following steps: Step S1: Based on historical and online operation data, establish a multi-dimensional operation database covering the entire process of desulfurization byproduct generation and dewatering, including flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, oxidation fan power, sulfite ion concentration in the slurry at the bottom of the desulfurization tower, slurry pH, slurry density, gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones, solids content in the hydrocyclone underflow, slurry volume at the hydrocyclone outlet, vacuum belt dewatering machine speed, vacuum pump pressure, gypsum flushing water pump power, and gypsum moisture content at the outlet of the vacuum belt dewatering machine. Step S2: Based on the database established in Step S1, analyze the response relationship between the flue gas flow rate at the desulfurization tower inlet, flue gas temperature, SO2 concentration, O2 concentration, desulfurization tower spray slurry volume, slurry pH, slurry density, oxidation fan power, and sulfite ion concentration in the slurry at the bottom of the desulfurization tower. Construct a prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower under different operating conditions. Further combine historical operating data and online monitoring data on the sulfite ion concentration in the slurry at the bottom of the desulfurization tower to revise the prediction model, forming an accurate prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower that combines mechanism and data. Step S3: Based on the database established in Step S1, analyze the response relationship between the gypsum discharge pump flow rate, hydrocyclone pressure, number of hydrocyclones and solid content in the hydrocyclone underflow, establish a prediction model for solid content in the hydrocyclone underflow under different operating conditions, and further combine historical operating data and online monitoring data of solid content in the hydrocyclone underflow to revise the prediction model for solid content in the hydrocyclone underflow, forming a knowledge- and data-driven accurate prediction model for solid content in the hydrocyclone underflow. Step S4: Based on the database established in Step S1, analyze the response relationship between the slurry volume at the hydrocyclone outlet, the solid content in the hydrocyclone underflow, the speed of the vacuum belt dewatering machine, the pressure of the vacuum pump, the power of the gypsum flushing water pump, and the moisture content of the gypsum at the outlet of the vacuum belt dewatering machine. Establish a prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine under different operating conditions. Further combine the historical operating data and online monitoring data of the quality of gypsum at the outlet of the vacuum belt dewatering machine to revise the prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine, forming a knowledge- and data-driven accurate prediction model for the quality of gypsum at the outlet of the vacuum belt dewatering machine. The control of the gypsum discharge pump operating parameters is as follows: based on the real-time measurement data of the slurry density meter at the bottom of the desulfurization tower, the gypsum discharge pump start and stop commands are sent to the DCS controller through the data transmission module after optimization and adjustment strategy, and then the DCS controller sends the commands to the field equipment. The control of the oxidation blower operating parameters is as follows: by using the established accurate prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is calculated. Then, combined with the real-time operating data of flue gas volume, temperature, SO2 concentration, and oxidation blower power at the desulfurization tower inlet, the oxidation blower power command is adjusted through an optimized control strategy and sent to the DCS controller through the data transmission module. The DCS controller then sends the command to the field equipment. The predicted value of sulfite ion concentration in the slurry at the bottom of the desulfurization tower is used as the given instruction, and the flue gas volume, temperature, SO2 concentration, and oxidation fan power at the inlet of the desulfurization tower are used as the feedforward instructions. The optimized control strategy involves setting the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower, selecting the method with the lowest oxidation fan power, and setting the upper and lower limits of the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower to ensure that the gypsum quality meets the standards. The control of the hydrocyclone operating parameters is as follows: Based on the established accurate prediction models of sulfite ion concentration at the bottom of the desulfurization tower and solids content at the bottom of the hydrocyclone, the predicted values ​​of sulfite ion concentration at the bottom of the desulfurization tower and solids content at the bottom of the hydrocyclone are calculated. In addition, combined with the real-time operating data of slurry flow rate at the gypsum discharge pump outlet and hydrocyclone pressure, the hydrocyclone is adjusted through an optimized control strategy. The instructions for adding or removing the hydrocyclone are sent to the DCS controller through the data transmission module, and then the DCS controller sends the instructions to the field equipment. The predicted solids content in the hydrocyclone underflow is used as a given command, while the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the slurry flow rate at the outlet of the gypsum discharge pump, and the hydrocyclone pressure are used as feedforward commands. The control of the operating parameters of the vacuum belt dewatering machine is as follows: Based on the established accurate prediction models of sulfite ion concentration at the bottom of the desulfurization tower, solid content at the underflow of the hydrocyclone, and gypsum quality at the outlet of the vacuum belt dewatering machine, the predicted values ​​of sulfite ion concentration at the bottom of the desulfurization tower, solid content at the underflow of the hydrocyclone, and gypsum quality at the outlet of the vacuum belt dewatering machine are calculated. Further, combined with the slurry volume at the hydrocyclone outlet and the vacuum pump pressure, the multi-model prediction optimization control strategy is used to adjust the speed command of the vacuum belt dewatering machine and the power command of the gypsum flushing water pump. The command is then sent to the DCS controller through the data communication module, and the DCS controller sends the command to the field equipment. The predicted value of gypsum quality at the outlet of the vacuum belt dewatering machine is used as the given instruction, and the predicted value of sulfite ion concentration at the bottom of the desulfurization tower, the predicted solid content of the hydrocyclone underflow, the slurry volume at the hydrocyclone outlet, and the vacuum pump pressure are used as feedforward instructions. The optimization and control strategy involves setting the minimum speed of the vacuum belt dewatering machine and the minimum power of the gypsum flushing pump under the condition of setting the gypsum quality at the outlet of the vacuum belt dewatering machine, and setting the constraint conditions for the gypsum quality at the outlet of the vacuum belt dewatering machine to ensure that the gypsum quality meets the standards. The process of adjusting the Particle Swarm Optimization algorithm is as follows: Calculate the optimal particle fitness: fitness = cost min = cost a + cost b + cost c in, cost a , cost b , cost c These represent the costs of the by-product generation process, the cyclone dehydration process, and the belt dehydration process, respectively. The cost of byproduct generation includes the electricity consumption (Power0) generated by the oxidation blower and the electricity consumption (Power) generated by the gypsum discharge pump. P The cost of the cyclone dehydration process includes the energy loss due to friction loss. b The cost of the belt dewatering process includes the power consumption of the belt dewatering machine. B Power consumption of vacuum pump V and the power consumption of the belt flushing water pump W .

2. The intelligent control method for the entire process of desulfurization byproduct generation according to claim 1, characterized in that, Step S2 specifically includes: Step S2.1: Establish models for the natural oxidation process and forced oxidation process of the desulfurization oxidation system: ； In the formula, S 1 It is the natural oxidizing factor of SO2 in the desulfurization tower; S 2 It is a forced oxidizing agent for SO2 in the desulfurization tower; Q is the molar concentration of SO2 in the flue gas at the inlet of the desulfurization tower; Q is the flow rate of the flue gas at the inlet of the desulfurization tower. The molar concentration of O2 in the flue gas at the inlet of the desulfurization tower; q This refers to the volume of spray slurry used in the desulfurization tower. pH Slurry sprayed on the desulfurization tower pH value; ρ Where W is the density of the spray slurry in the desulfurization tower, and W is the operating power of the oxidation fan. Step S2.2: Based on the models of natural oxidation and forced oxidation processes in the desulfurization oxidation system, establish models for the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time. ； In the formula, X S This refers to the amount of SO2 absorbed per unit time. For a unit of time period; SO2 removal efficiency; X ab The amount of SO2 oxidized per unit time; Q a Introduce airflow to the oxidation blower; The oxidation fan introduces the O2 molar concentration from the air; Step S2.3: Based on the models of the amount of SO2 absorbed per unit time and the amount of SO2 oxidized per unit time, establish a prediction model for the sulfite ion concentration in the slurry at the bottom of the desulfurization tower: ； By combining the models of SO2 moles absorbed per unit time and SO2 moles oxidized per unit time with the prediction model of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, the concentration of sulfite ions in the slurry at the bottom of the desulfurization tower is obtained: ； In the formula, The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower; for The concentration of sulfite ions in the slurry at the bottom of the desulfurization tower is constantly monitored. Step S2.4: Based on historical operating data and online sampling detection data of sulfite ion concentration in the slurry at the bottom of the desulfurization tower, revise the prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower, forming an accurate prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower that combines mechanism and data. ； In the formula, To correct the sulfite ion concentration in the slurry at the bottom of the desulfurization tower, γ This is a correction factor; In the prediction model for sulfite ion concentration in the slurry at the bottom of the desulfurization tower γ Perform parameter identification.

3. The intelligent control method for the entire process of desulfurization byproduct generation according to claim 1, characterized in that, Step S3 specifically includes: Step S3.1: Establish a hydrocyclone separation efficiency model based on the hydrocyclone inlet pressure, slurry flow rate at the gypsum discharge pump outlet, number of hydrocyclones, hydrocyclone structural dimensions, slurry density at the gypsum discharge pump outlet, and sulfite ion concentration in the slurry at the gypsum discharge pump outlet. ； In the formula, η For hydrocyclone separation efficiency; P This refers to the inlet pressure of the hydrocyclone. Q 2 represents the slurry flow rate at the outlet of the gypsum discharge pump; n The number of cyclotrons; D These are the structural dimensions of the hydrocyclone; The density of the slurry at the outlet of the gypsum discharge pump; Step S3.2: Based on the hydrocyclone separation efficiency, combined with the hydrocyclone outlet pressure drop, the volume percentage of solids in the feed material, and the specific gravity of the feed material, the separation efficiency model is corrected. The corrected hydrocyclone separation efficiency model is as follows: ； In the formula, To correct the separation efficiency of the hydrocyclone; For hydrocyclone pressure drop, A V is the volume percentage of solids in the hydrocyclone feed; s is the specific gravity of the solids in the feed. Step S3.3: Based on the corrected hydrocyclone separation efficiency, and combined with the slurry flow rate at the gypsum discharge pump outlet, slurry density, and sulfite ion concentration in the slurry at the bottom of the desulfurization tower, establish a knowledge- and data-driven accurate prediction model for the solids content of the hydrocyclone underflow: ； In the formula, m This refers to the solids content of the underflow from the hydrocyclone.

4. The intelligent control method for the entire process of desulfurization byproduct generation according to claim 1, characterized in that, Step S4 specifically includes: Step S4.1: Based on the slurry volume at the hydrocyclone outlet and the rotational speed of the vacuum belt dewatering machine, establish a formula for calculating the slurry thickness of the vacuum belt dewatering machine: ； In the formula, h is the slurry thickness of the vacuum belt dewatering machine; v The rotational speed of the vacuum belt dewatering machine; Q 3 represents the slurry flow rate at the hydrocyclone outlet; b The width of the filter cloth for the vacuum belt dewatering machine; Step S4.2: Based on the slurry thickness of the vacuum belt dewatering machine, the slurry volume at the hydrocyclone outlet, the solids content of the slurry at the hydrocyclone outlet, the speed of the vacuum belt dewatering machine, the vacuum pump pressure, and the spray liquid volume during the belt dewatering process, establish a prediction model for the dewatering amount during the belt dewatering process: ； In the formula, This refers to the amount of water removed during the belt dehydration process. p For vacuum pump pressure, This refers to the amount of spray liquid used during the belt dehydration process. Step S4.3: Based on historical operational data and online sampling detection data of the dewatering process in the belt dewatering process, revise the dewatering amount prediction model for the belt dewatering process to form an accurate prediction model for the dewatering amount of the belt dewatering process that combines mechanism and data. ； In the formula, Y is the correction factor for the amount of water removed during the belt dehydration process. In the prediction model of water removal amount in the belt dewatering process Perform parameter identification; Step S4.4: Based on the modified belt dewatering process dewatering volume model, and combined with the hydrocyclone outlet slurry volume, hydrocyclone outlet slurry solids content, and belt dewatering process spray volume, establish a precise prediction model for the gypsum moisture content at the vacuum belt dewatering machine outlet, combining mechanism and data. ； In the formula, The moisture content of the gypsum at the outlet of the vacuum belt dewatering machine.

Images