Full-technology integration method, device and equipment for intelligent desulfurization and medium
By collecting and processing key parameters in the wet limestone-gypsum desulfurization system, and combining mechanism modeling and a data-driven two-stage MPC structure, precise control of circulating slurry pH and outlet SO2 concentration was achieved. This solved the system's stability and economic issues under varying operating conditions, and improved the system's robustness and operating efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DATANG ENVIRONMENT IND GRP
- Filing Date
- 2025-12-18
- Publication Date
- 2026-05-12
AI Technical Summary
Under varying loads and fluctuating fuel sulfur content, the pH of the circulating slurry and the SO2 concentration at the outlet are prone to fluctuations in wet limestone-gypsum desulfurization. Traditional control methods are difficult to achieve high-precision, stable, and economical operation, and there are also problems such as particle size/density detection errors, unmeasurable circulating slurry flow rate, chloride ion accumulation leading to dehydration difficulties, and gypsum quality degradation.
By collecting key parameter data of the entire desulfurization process through DCS system and smart instruments, preprocessing and feature extraction are performed. Combined with mechanism modeling and data-driven approach, a two-stage MPC structure is adopted to design the control loop, which realizes precise control of circulating slurry pH and outlet SO2 concentration. Furthermore, the operating cost is reduced by optimizing the objective function of the entire process, and the entire system is linked for coordinated control.
It improves control robustness and system stability under complex and variable operating conditions, reduces equipment impact and frequent actions, lowers the overall operating cost, and ensures stable system operation and economy.
Smart Images
Figure CN122006436A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent desulfurization control technology, and to a fully integrated method, device, equipment and medium for intelligent desulfurization, particularly to an intelligent control and full process optimization method for wet desulfurization (FGD) devices in coal-fired power plants. Background Technology
[0002] In wet limestone-gypsum desulfurization processes, fluctuations in circulating slurry pH and outlet SO2 concentration are common under varying loads and fluctuating fuel sulfur content. Traditional control methods, relying primarily on manual experience and single-loop PID control, struggle to achieve high-precision, stable, and economical operation under complex disturbances. Furthermore, common engineering problems include particle size / density detection errors, unmeasurable circulating slurry flow rates, chloride ion accumulation leading to dewatering difficulties, and decreased gypsum quality. Therefore, there is an urgent need to integrate mechanistic modeling, data-driven approaches, advanced control, and online optimization to create a smart solution that can be safely implemented within a DCS (Distributed Control System) framework. Summary of the Invention
[0003] To overcome the problems existing in related technologies, this disclosure provides a fully integrated method, apparatus, equipment and medium for intelligent desulfurization, in order to solve the technical problems in related technologies.
[0004] This specification provides one or more embodiments of a fully integrated intelligent desulfurization technology method, including the following steps: Key parameter data of the entire desulfurization process are collected through DCS system and smart instruments; the collected data are preprocessed and feature extracted to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate; Based on the preprocessed data, thresholds are set according to the unit load and the rate of change of fuel sulfur content to determine the current operating condition, and the corresponding prediction model parameter set and control strategy are determined according to the operating condition. Based on the pre-processed data, the inlet SO2 mass flow rate and circulating slurry pH value, as well as the absorber outlet SO2 concentration, are predicted at a preset time according to the preset model. A two-stage MPC structure is adopted, with control loops designed for circulating slurry pH and outlet SO2 concentration respectively. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to set values, and the smoothness of system operation. Under the constraints of equipment and process, the optimal control scheme is obtained, including the combination of slurry flow rate and circulating pump start-stop. Under the premise of meeting SO2 emission standards and equipment constraints, an optimization objective function is constructed with the goal of minimizing the overall process operating cost. A slow-loop operation mode is adopted, and the optimal steady-state target is output as the MPC setpoint constraint. The obtained control scheme is decoupled to the pulping, dewatering, wastewater and urea hydrolysis subsystems to obtain executable instructions, thereby realizing the linkage and control of the entire system process.
[0005] This specification provides one or more embodiments of a fully integrated intelligent desulfurization device, including: The data acquisition and processing module is used to collect key parameter data of the entire desulfurization process through the DCS system and smart instruments; to preprocess and extract features from the collected data to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate; The operating condition identification module is used to determine the current operating condition based on the preprocessed data, the threshold set based on the unit load and the rate of change of fuel sulfur content, and to determine the corresponding prediction model parameter set and control strategy based on the operating condition. The predictive modeling module is used to predict the inlet SO2 mass flow rate and circulating slurry pH value at a preset time, as well as the SO2 concentration at the absorber outlet, based on the preprocessed data and a preset model. The MPC controller operation module is used to design control loops for circulating slurry pH and outlet SO2 using a two-level MPC structure. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to set values, as well as the smoothness of system operation. Under equipment and process constraints, the optimal control scheme is obtained, including slurry flow rate and circulating pump start-stop combination. The online input optimization control module is used to construct an optimization objective function with the goal of minimizing the overall process operating cost, under the premise of meeting SO2 emission standards and equipment constraints. It adopts a slow loop operation mode and outputs the optimal steady-state target as the MPC setpoint constraint. The collaborative execution module is used to decouple the obtained control scheme to subsystems such as pulping, dewatering, wastewater treatment, and urea hydrolysis to obtain executable instructions, thereby realizing the coordinated control of the entire system process.
[0006] This specification provides one or more embodiments of a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the fully integrated intelligent desulfurization technology method as described above.
[0007] This specification provides one or more embodiments of a computer-readable storage medium storing a computer program that, when executed by a processor, implements the fully integrated intelligent desulfurization technology method as described above.
[0008] This disclosure provides a fully integrated method, apparatus, equipment, and medium. Its advantages include: a predictive model that combines process mechanisms with real-time data, balancing predictive accuracy with physical interpretability, thus avoiding the insufficient generalization ability of purely data-driven models; dynamic matching of the predictive model parameter set and control strategy through operating condition classification, improving control robustness under complex and varying operating conditions; a two-level MPC structure for targeted management of core parameters, balancing control accuracy and operational smoothness in the objective function, reducing equipment impact and frequent actions; a fast-slow loop collaborative mechanism that achieves dynamic and precise control through MPC while reducing overall process operating costs through online economic optimization, balancing compliance and economy; and a robust safety assurance mechanism and coordinated execution of all subsystems to ensure stable system operation. Attached Figure Description
[0009] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0010] Figure 1 A flowchart illustrating a fully integrated intelligent desulfurization technology method provided for one or more embodiments of this specification; Figure 2 A block diagram of a fully integrated intelligent desulfurization device provided for one or more embodiments of this specification; Figure 3 This is a schematic diagram of the structure of a computer device provided for one or more embodiments of this specification. Detailed Implementation
[0011] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this invention.
[0012] The present invention will now be described in detail with reference to specific embodiments and accompanying drawings.
[0013] Method Implementation Examples According to embodiments of the present invention, a fully integrated method for intelligent desulfurization technology is provided, such as... Figure 1The diagram shown is a flowchart of the intelligent desulfurization full technology integration method provided in this embodiment. The intelligent desulfurization full technology integration method according to this embodiment includes the following steps: Step S1, Data Acquisition and Processing: Collect key parameter data of the entire desulfurization process through the DCS system and smart instruments, including flue gas, slurry and equipment operation data collected through the DCS system, and slurry particle size, slurry density, chloride ion concentration and gypsum moisture content collected through smart instruments; preprocess and extract features from the collected data to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate.
[0014] In one embodiment, the data collected by the DCS system includes the inlet flue gas flow rate F. g (k), inlet SO2 concentration C in (k), unit load L(k), coal feed rate or current, absorber level, spray layer pressure drop, circulating pump operating status, including at least the circulating pump operating frequency. The operating parameters of the slurry pump should include at least the slurry flow rate Q. s (k) Oxidation system parameters, basic parameters of gypsum dewatering system, including hydrocyclone pressure, filter cake thickness, belt speed and vacuum degree, pressure, temperature, steam valve opening of urea hydrolysis tank, denitrification steam consumption, etc.
[0015] In one embodiment, the slurry particle size is collected by a laser scattering principle monitoring instrument, the slurry density is collected by a tuning fork vibration principle density meter, the chloride ion concentration is collected by a sensor in a key pipeline or filtrate tank, and the gypsum moisture content is obtained by an online detector.
[0016] In this embodiment, data processing includes time alignment, anomaly removal, data interpolation, feature extraction, and noise suppression of the raw data to ensure data quality and usability, and to meet the requirements of subsequent modeling and control. The specific details are as follows: Time alignment: All data are synchronized with a uniform time offset step length Δt to ensure the time sequence consistency of data from different sources.
[0017] Anomaly removal: Identify and remove abnormal data caused by sensor malfunctions or transmission interference, such as jumps and out-of-range data.
[0018] Data imputation: Imputation algorithms are used to supplement missing small amounts of data to ensure data continuity, such as linear imputation.
[0019] Feature extraction is performed by collecting the inlet flue gas flow rate F. g (k) and inlet SO2 concentration C in (k) Calculate the inlet SO2 mass flow rate; If the circulating slurry flow rate is unmeasurable, it can be determined by calculating the rated flow rate of the already operational circulating pump and the frequency converter, as shown in the following formula: , ; in, For the first j The rated flow rate of the circulating pump For the first j Taiwan circulating pump k The actual operating frequency at any given time, For the first j The rated frequency of the circulating pump.
[0020] Noise suppression: Kalman filtering or exponential smoothing algorithms are used on the processed data to reduce the impact of measurement noise on subsequent modeling and control.
[0021] Step S2, Operating Condition Identification: Based on the preprocessed data, thresholds are set according to the unit load and the rate of change of fuel sulfur content to determine the current operating condition, and the corresponding prediction model parameter set and control strategy are determined according to the operating condition.
[0022] In this embodiment, the set unit load and the rate of change of fuel sulfur content are calculated as follows: The load change rate ΔL(k) = L(k) - L(k-1), where L(k) is the load at the current moment and L(k-1) is the load at the previous moment. The rate of change in sulfur content ΔS(k) = C in (k)-C in (k-1), C in (k) represents the current inlet SO2 concentration, characterizing the sulfur content of the fuel, C in (k-1) represents the SO2 concentration at the inlet at the previous moment.
[0023] In this embodiment, the operating conditions are determined based on the threshold of the rate of change of unit load and fuel sulfur content, including the following: (1) Expansion condition: Triggered when the unit load change is greater than the load change threshold, i.e., ΔL(k)>εL or the sulfur content change is greater than the sulfur content change threshold, i.e., ΔS(k)>εS, corresponding to the pre-start of the pump group and the set lifting strategy.
[0024] (2) Drop condition: Triggered when the load change is less than the load drop threshold, i.e., ΔL(k) < -(εL+h), where h is the buffer coefficient, or when the sulfur content change is less than the sulfur content drop threshold, i.e., ΔS(k) < -(εS+h), corresponding to the pump stop and the set adjustment strategy.
[0025] (3) Stable operating condition: Except for rising operating condition and falling operating condition, conventional parameter set control is used for other conditions.
[0026] In this embodiment, the prediction model parameter set is the set of coefficients, weights, and regression parameters used to fit data and calculate predicted values in the core prediction models such as the pH prediction model and the outlet SO2 concentration prediction model under different working conditions divided in the working condition identification stage.
[0027] Step S3, Predictive Modeling: Based on the preprocessed data, predict the inlet SO2 mass flow rate and circulating slurry pH value at the preset time and the SO2 concentration at the absorber outlet according to the preset model.
[0028] In another embodiment, step S3 specifically includes performing the following steps: The inlet SO2 mass flow rate at the next moment is predicted using a preset inlet SO2 mass flow rate prediction model based on the collected inlet SO2 mass flow rate. The circulating slurry pH value at the next moment is predicted using a preset circulating slurry pH value prediction model based on the collected historical circulating slurry pH value, slurry flow rate, inlet flue gas flow rate, and inlet SO2 concentration. The current absorber tower outlet SO2 concentration C is predicted using a preset outlet SO2 concentration prediction model based on the collected current slurry flow rate, inlet flue gas flow rate, circulating pump operating frequency, inlet SO2 concentration, inlet SO2 mass flow rate, and circulating slurry flow rate. out .
[0029] In one embodiment, the circulating slurry pH prediction model adopts the NARX structure, and the model expression is as follows: ; Where θ0~θ4 are model parameters, pH ( k - i )for k - i pH value of circulating slurry at time Q s ( k - i )for k - i The flow rate of the slurry at any given time, F so2,in ( k - i )for k - i The inlet SO2 mass flow rate at time t, w k The model error term is obtained by fitting historical data to obtain the optimal parameters for each working condition.
[0030] The export SO2 concentration prediction model adopts a mechanism-enhanced ARX structure, and the model expression is as follows: ; ; The absorption efficiency η(k) is calculated using the penetration model, where k L a is the mass transfer coefficient, τ is the residence time, ξ is the correction factor for pH and temperature T, and τ(k) = K(L / G) (K is the proportionality coefficient, L / G is the liquid-to-gas ratio). To enhance robustness, an error regression term is introduced for correction, expressed as: ; in, For regression parameters, C out ( k - i )for k - i The SO2 concentration at the outlet at time Q c ( k - i )for k - i The circulating slurry flow rate at any given time, pH ( k - i )for k - i pH value of circulating slurry at time v k This is the error term.
[0031] The inlet SO2 mass flow rate is predicted using an ARX model based on historical time series data and operating condition characteristics, providing feedforward disturbance prediction for subsequent control.
[0032] Step S4, MPC controller operation: A two-level MPC structure is adopted, and control loops are designed for circulating slurry pH and outlet SO2 respectively. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to the set values, as well as the smoothness of system operation. Under the constraints of equipment and process, the optimal control scheme is obtained, including the combination of slurry flow rate and circulating pump start-stop.
[0033] In this embodiment, step S4 specifically includes the following steps: Step S41: A two-stage MPC structure is adopted, which includes an MPC loop for circulating slurry pH and an MPC control loop for outlet SO2 concentration. In this embodiment, to avoid the impact of sudden changes in control quantity on the system, an incremental control quantity is set for the MPC controller, including the slurry flow rate increment ΔQ. s (k)=Q s (k)-Q s (k-1), and the frequency increment of the circulating pump Δf(k)=f(k)-f(k-1).
[0034] The MPC control loop for circulating slurry pH uses the inlet flue gas flow rate Fg(k) and the inlet SO2 concentration C. in (k) is combined with the operation of the circulating pump reflecting the change in the circulating slurry flow rate as the disturbance variable, the adjustment of the slurry pump as the operating variable, the current and historical pH values, and the circulating slurry pH (k+1)~pH (k+Np) output by the prediction model as the feedback input, the prediction step size of Np, the circulating slurry pH as the controlled variable, and the control objective is to stabilize within the set value ±0.2.
[0035] The MPC control loop for outlet SO2 concentration uses inlet flue gas parameters and the start / stop combination of the circulating pump as disturbance variables, and uses the current and historical outlet SO2 concentration values and the outlet SO2 concentration C output by the prediction model as the basis. out (k+1)~C out (k+Np) is the feedback input, the operated variables include the start-stop combination of multiple circulating pumps and the frequency of the variable frequency, the controlled variable is the outlet SO2 concentration, and the control target is to control it at 28±5 mg / Nm³ (example indicator) and not exceed the emission limit C. max .
[0036] Step S42 aims to optimize the control accuracy of the circulating slurry pH value and outlet SO2 concentration relative to the set values, as well as the smoothness of system operation, as follows: ; Among them, pH*, C out *These are the setpoints for the circulating slurry pH and the outlet SO2 concentration, respectively. , These are the weighting coefficients. To predict the step size, This is the controller's control window, and the control output window for each round of optimization.
[0037] In this optimization objective, the first and second parts of the formula are to make the future prediction time ( k + i While ensuring that the actual pH value and outlet SO2 concentration are as close as possible to the preset optimal setpoint, ensure that desulfurization emissions meet the standards (i.e., C). out (Not exceeding the upper limit) and pH stable within the optimal absorption range; the third part of the formula limits the increment of slurry flow rate (ΔQ). s To mitigate fluctuations in the frequency increment (Δf) of the circulating pump and prevent system shocks caused by sudden changes in control inputs. Constraints: pH operating range: pH min ≤pH(k+i) ≤ pH max ; SO2 emission limit: C out (k+i)≤C max ; Upper and lower limits of slurry flow rate: Q smin ≤Q s (k+i)≤Q smax ; Limit on slurry flow rate increment: |ΔQ s (k+i)|≤r s ; Pump frequency upper and lower limits: f min ≤f(k+i)≤f max ; Pump frequency increment limit: ; Equipment and process constraints: including the upper and lower limits of the liquid level in the absorption tower and the upper limit of the pressure drop in the spray layer.
[0038] Step S43: For the continuous variables of slurry flow rate and circulating pump frequency, the objective function is transformed into a quadratic programming problem, and the optimal incremental control quantity is solved using a numerical optimization algorithm. For solving the circulating pump start-stop combination, a two-layer design is adopted. The upper layer enumerates the circulating pump start-stop combination set U or obtains the optimal circulating pump start-stop combination u* through mixed integer programming, and a penalty term is added to the switching of the circulating pump start-stop combination. To reduce frequent starts, the lower layer uses a quadratic programming approach based on the optimal cycle pump start / stop combination u* to solve for continuous variables.
[0039] In this embodiment, the setting of the pump start-stop combination switching penalty item is to reduce the frequent start-stop of the circulating pump by setting a penalty coefficient, thereby reducing equipment wear and unnecessary energy consumption and extending equipment life.
[0040] Step S5, online optimization control: Under the premise of meeting SO2 emission standards and equipment constraints, construct an optimization objective function with the goal of minimizing the overall operating cost, adopt a slow loop operation mode, and output the optimal steady-state target as the MPC setpoint constraint.
[0041] In this embodiment, the optimization objective function is constructed with the goal of minimizing the overall process operating cost as follows: ; Similarity Law ; In the formula, λ e As the weight for power consumption, E pump For the power consumption of the circulating pump, λ l For the weight of limestone consumption, c slueey λ is the slurry consumption coefficient. chem As the weight of drug consumption, Q waste For wastewater discharge, C chem For drug cost; λ em Weighting for penalties for exceeding emission standards.
[0042] In this embodiment, the optimization frequency is operated in a slow loop, such as with a period of 60 seconds, and alternates with the MPC fast loop, such as with a period of 10 seconds. The optimal steady-state target of the optimized output slurry supply setpoint and the combination / frequency of the circulating pump start / stop is used as the setpoint constraint of the MPC controller to achieve coordinated control of optimal steady-state cost and dynamic constraint tracking.
[0043] Step S6, Collaborative Execution: Decouple the control scheme obtained in step S4 to subsystems such as pulping, dewatering, wastewater treatment, and urea hydrolysis to obtain executable instructions, thereby achieving full-process linkage control of the system.
[0044] In this embodiment, the control strategies for each subsystem are as follows: Pulping subsystem: Based on online particle size / density detection, adjust the wet ball mill pulping parameters to maintain the particle size distribution and concentration of the slurry within the target range, and feed the data back to the slurry supply / dewatering subsystem; Dewatering subsystem: Based on online detection of gypsum moisture content, it adjusts hydrocyclone pressure, filter cake thickness, belt speed, and vacuum level in a coordinated manner; Wastewater system: The wastewater discharge path and timing are determined based on the online chloride ion concentration and prediction results; Urea hydrolysis system: The solution concentration is calculated online using a density-concentration conversion model, and the pressure, temperature and steam valve opening of the hydrolysis tank are optimized with the goal of minimizing the consumption of denitrification steam.
[0045] This embodiment also includes the following steps: Step S7 ensures the security of interaction between the APC system and the DCS system through heartbeat monitoring, anti-miswrite, and emergency rollback.
[0046] In one specific embodiment, the system includes a communication protection mechanism between the host APC and DCS, anti-write logic, and an emergency rollback mechanism for parameter out-of-bounds errors or device failures, as detailed below: ① Communication and Status Monitoring: The upper-level APC (Advanced Process Control) system periodically writes flag bits to the DCS system, and the DCS system resets in real time, forming a heartbeat mechanism; if no heartbeat is received within the time limit, the DCS system automatically cuts off the output of the APC system and alarms. ② Anti-miswrite logic: The set value written to the DCS system must pass the threshold consistency check, that is, compare the difference between the APC output and the current value of the DCS. Only if the threshold condition is met can it be written. ③ Emergency Retreat: When parameters exceed limits or equipment malfunctions, the system automatically reverts to DCS local loop control to ensure continuous operation of the desulfurization system.
[0047] The beneficial effects of this invention are as follows: 1. Deep integration of data and mechanism: The predictive model combines process mechanism with real-time data, taking into account both prediction accuracy and physical interpretability, and avoiding the problem of insufficient generalization ability of pure data-driven models; 2. Adaptive control under operating conditions: Dynamic matching of the predictive model parameter set and the control strategy is achieved through operating condition classification, thereby improving the control robustness under complex and variable operating conditions; 3. Precise and smooth coordinated control: The two-level MPC structure specifically manages core parameters, and the objective function balances control precision and operational smoothness, reducing equipment impact and frequent actions; 4. Dual optimization of control and economy: The fast and slow loop collaborative mechanism achieves dynamic and precise control through MPC, and reduces the overall operating cost through online economic optimization, thus balancing compliance and economy; 5. Safe and reliable throughout the entire process: A comprehensive security mechanism and coordinated execution across all subsystems ensure stable system operation, while visualization and traceability functions support operation and maintenance optimization; 6. Data lifecycle management: From full data collection and processing to refined processing, it provides high-quality data support for subsequent modeling, control, and optimization, forming a closed-loop management and control system.
[0048] Device Examples According to embodiments of the present invention, a fully integrated intelligent desulfurization technology device is provided, such as... Figure 2 The diagram shown is a block diagram of the intelligent desulfurization integrated technology device provided in this embodiment. According to this embodiment of the intelligent desulfurization integrated technology device, it includes: The data acquisition and processing module 10 is used to collect key parameter data of the entire desulfurization process through the DCS system and smart instruments, including flue gas, slurry and equipment operation data collected through the DCS system, and slurry particle size, slurry density, chloride ion concentration and gypsum moisture content collected through smart instruments; the collected data is preprocessed and feature extracted to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate.
[0049] The operating condition identification module 20 is used to determine the current operating condition based on the preprocessed data, the threshold set based on the unit load and the rate of change of fuel sulfur content, and to determine the corresponding prediction model parameter set and control strategy based on the operating condition.
[0050] In this embodiment, the operating conditions are determined based on the threshold of the rate of change of unit load and fuel sulfur content, including the following: (1) Expansion condition: Triggered when the unit load change is greater than the load change threshold, i.e., ΔL(k)>εL or the sulfur content change is greater than the sulfur content change threshold, i.e., ΔS(k)>εS, corresponding to the pre-start of the pump group and the set lifting strategy.
[0051] (2) Drop condition: Triggered when the load change is less than the load drop threshold, i.e., ΔL(k) < -(εL+h), where h is the buffer coefficient, or when the sulfur content change is less than the sulfur content drop threshold, i.e., ΔS(k) < -(εS+h), corresponding to the pump stop and the set adjustment strategy.
[0052] (3) Stable operating condition: Except for rising operating condition and falling operating condition, conventional parameter set control is used for other conditions.
[0053] The predictive modeling module 30 is used to predict the inlet SO2 mass flow rate and circulating slurry pH value at a preset time and the SO2 concentration at the absorber outlet based on the preprocessed data and a preset model.
[0054] The predictive modeling module 30 is specifically configured to: predict the inlet SO2 mass flow rate at the next moment based on the collected inlet SO2 mass flow rate using a preset inlet SO2 mass flow rate prediction model; predict the circulating slurry pH value at the next moment based on the collected historical circulating slurry pH value, slurry flow rate, inlet flue gas flow rate, and inlet SO2 concentration using a preset circulating slurry pH value prediction model; and predict the current absorber tower outlet SO2 concentration C based on the collected current slurry flow rate, inlet flue gas flow rate, circulating pump operating frequency, inlet SO2 concentration, inlet SO2 mass flow rate, and circulating slurry flow rate using a preset outlet SO2 concentration prediction model. out .
[0055] In this embodiment, the circulating slurry pH prediction model adopts the NARX structure, and the model expression is as follows: ; Where θ0~θ4 are model parameters, PH( k - i )for [[ID=ip 19]]k - i pH value of circulating slurry at time Q s ( k - i )for k - i The flow rate of the slurry at any given time, F so2,in ( k - i )for k - i The inlet SO2 mass flow rate at time t, w k The model error term is obtained by fitting historical data to obtain the optimal parameters for each working condition.
[0056] The export SO2 concentration prediction model adopts a mechanism-enhanced ARX structure, and the model expression is as follows: ; ; The absorption efficiency η(k) is calculated using the penetration model, where k L a is the mass transfer coefficient, τ is the residence time, ξ is the correction factor for pH and temperature T, and τ(k) = K(L / G) (K is the proportionality coefficient, L / G is the liquid-to-gas ratio). To enhance robustness, an error regression term is introduced for correction, expressed as: ; in, For regression parameters, C out ( k - i )for k - i The SO2 concentration at the outlet at time Q c ( k - i )for k - i The circulating slurry flow rate at any given time, pH ( k - i )for k - i pH value of circulating slurry at time vk This is the error term.
[0057] The inlet SO2 mass flow rate is predicted using an ARX model based on historical time series data and operating condition characteristics, providing feedforward disturbance prediction for subsequent control.
[0058] The MPC controller operation module 40 is used to design control loops for circulating slurry pH and outlet SO2 respectively using a two-level MPC structure. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to the set values, as well as the smoothness of system operation. Under the constraints of equipment and process, the optimal control scheme is obtained, including the combination of slurry flow rate and circulating pump start-stop.
[0059] In this embodiment, a two-stage MPC structure is adopted, which includes an MPC loop for circulating slurry pH and an MPC control loop for outlet SO2 concentration. The MPC control loop for circulating slurry pH uses the inlet flue gas flow rate Fg(k) and the inlet SO2 concentration C. in (k) is combined with the operation of the circulating pump reflecting the change in the circulating slurry flow rate as the disturbance variable, the adjustment of the slurry pump as the operating variable, the current and historical pH values, and the circulating slurry pH (k+1)~pH (k+Np) output by the prediction model as the feedback input, the prediction step size of Np, the circulating slurry pH as the controlled variable, and the control objective is to stabilize within the set value ±0.2.
[0060] The MPC control loop for outlet SO2 concentration uses inlet flue gas parameters and the start / stop combination of the circulating pump as disturbance variables, and uses the current and historical outlet SO2 concentration values and the outlet SO2 concentration C output by the prediction model as the basis. out (k+1)~C out (k+Np) is the feedback input, the operated variables include the start-stop combination of multiple circulating pumps and the frequency of the variable frequency, the controlled variable is the outlet SO2 concentration, and the control target is to control it at 28±5 mg / Nm³ (example indicator) and not exceed the emission limit C. max .
[0061] In another embodiment, the optimization objectives are the control accuracy of the circulating slurry pH, the outlet SO2 concentration relative to the set value, and the smoothness of system operation, as detailed below: ; Among them, pH*, C out *These are the setpoints for the circulating slurry pH and the outlet SO2 concentration, respectively. , These are the weighting coefficients. To predict the step size, This is the controller's control window, and the control output window for each round of optimization.
[0062] Constraints: pH operating range: pH min ≤pH(k+i) ≤ pH max ; SO2 emission limit: C out (k+i)≤C max ; Upper and lower limits of slurry flow rate: Q smin ≤Q s (k+i)≤Q smax ; Limit on slurry flow rate increment: |ΔQ s (k+i)|≤r s ; Pump frequency upper and lower limits: f min ≤f(k+i)≤f max ; Pump frequency increment limit: ; Equipment and process constraints: including the upper and lower limits of the liquid level in the absorption tower and the upper limit of the pressure drop in the spray layer.
[0063] The online input optimization control module 50 is used to construct an optimization objective function with the goal of minimizing the overall operating cost, under the premise of meeting SO2 emission standards and equipment constraints. It adopts a slow-loop operation mode and outputs the optimal steady-state target as the MPC setpoint constraint.
[0064] In this embodiment, the optimization objective function is constructed with the goal of minimizing the overall process operating cost as follows: ; ; In the formula, λ e As the weight for power consumption, E pump For the power consumption of the circulating pump, λ l For the weight of limestone consumption, c slueey λ is the slurry consumption coefficient. chem As the weight of drug consumption, Q waste For wastewater discharge, C chem For drug cost; λ em Weighting for penalties for exceeding emission standards.
[0065] The collaborative execution module 60 is used to decouple the obtained control scheme to subsystems such as pulping, dewatering, wastewater treatment, and urea hydrolysis to obtain executable instructions, thereby realizing the coordinated control of the entire system process.
[0066] The embodiments of the present invention are device embodiments corresponding to the above method embodiments. The specific operations of each module processing step can be understood with reference to the description of the method embodiments, and will not be repeated here.
[0067] likeFigure 3 It should be noted that there seems to be a misspelling in "[[ID=ip 19]]", it should probably be "". If this is not a typo, please clarify for a more accurate translation. As shown, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the intelligent desulfurization technology integration method described in the above embodiments.
[0068] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the full technology integration method for intelligent desulfurization in the above embodiments.
[0069] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0070] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0071] Furthermore, the functional modules in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.
[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention 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. These 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 the present invention, and the contents not described in detail in the specification of the present invention are well known to those skilled in the art.
Claims
1. A fully integrated intelligent desulfurization technology method, characterized in that, Includes the following steps: Key parameter data of the entire desulfurization process are collected through DCS system and smart instruments; the collected data are preprocessed and feature extracted to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate; Based on the preprocessed data, thresholds are set according to the unit load and the rate of change of fuel sulfur content to determine the current operating condition, and the corresponding prediction model parameter set and control strategy are determined according to the operating condition. Based on the pre-processed data, the inlet SO2 mass flow rate and circulating slurry pH value, as well as the absorber outlet SO2 concentration, are predicted at a preset time according to the preset model. A two-stage MPC structure is adopted, with control loops designed for circulating slurry pH and outlet SO2 concentration respectively. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to set values, and the smoothness of system operation. Under the constraints of equipment and process, the optimal control scheme is obtained, including the combination of slurry supply flow rate and circulating pump start-stop. Under the premise of meeting SO2 emission standards and equipment constraints, an optimization objective function is constructed with the goal of minimizing the overall process operating cost. A slow-loop operation mode is adopted, and the optimal steady-state target is output as the MPC setpoint constraint. The obtained control scheme is decoupled to the pulping, dewatering, wastewater and urea hydrolysis subsystems to obtain executable instructions, thereby realizing the linkage and control of the entire system process.
2. The fully integrated intelligent desulfurization technology method as described in claim 1, characterized in that, The data collected by the DCS system includes inlet flue gas flow rate, inlet SO2 concentration, unit load, coal feed rate or current, absorber level, spray layer pressure drop, circulating pump operating frequency, slurry pump slurry flow rate, oxidation system parameters; and basic parameters of the gypsum dewatering system, including hydrocyclone pressure, filter cake thickness, belt speed and vacuum degree, urea hydrolysis tank pressure, temperature, steam valve opening, and denitrification steam consumption.
3. The fully integrated intelligent desulfurization technology method as described in claim 1, characterized in that, The specific steps for determining operating conditions based on the threshold of the rate of change of unit load and fuel sulfur content are as follows: Increased operating condition: Triggered when the unit load change rate is greater than the load change threshold or the sulfur content change rate is greater than the sulfur content change threshold, corresponding to the pre-start of the pump group and the setting of the boost strategy; Drop condition: Triggered when the load change rate is less than the load drop threshold, or the sulfur content change rate is less than the sulfur content drop threshold, corresponding to pump shutdown and setting down adjustment strategy; Stable operating conditions: Except for rising and falling operating conditions, conventional parameter set control is used.
4. The integrated intelligent desulfurization technology method as described in claim 1, characterized in that, The preset models include a circulating slurry pH prediction model and an outlet SO2 concentration prediction model. The pH prediction model for circulating slurry adopts the NARX structure, and the model expression is as follows: ； Where θ0~θ4 are model parameters, pH ( ki )for ki pH value of circulating slurry at time Q s ( ki )for ki The flow rate of the slurry at any given time, F so2,in ( ki )for ki The inlet SO2 mass flow rate at time t, w k The model error term is obtained by fitting historical data to obtain the optimal parameters for each working condition. The export SO2 concentration prediction model adopts a mechanism-enhanced ARX structure, and the model expression is as follows: ； ； The absorption efficiency η(k) is calculated using the penetration model, where k L a is the mass transfer coefficient, τ is the residence time, ξ is the correction factor for pH and temperature T, and τ(k) = K(L / G); Introducing an error regression term for correction, the expression is: ； in, For regression parameters, C out ( ki )for ki The SO2 concentration at the outlet at time Q c ( ki )for ki The circulating slurry flow rate at any given time, pH ( ki )for ki pH value of circulating slurry at time V k This is the error term.
5. The fully integrated intelligent desulfurization technology method as described in claim 1, characterized in that, The two-stage MPC structure includes an MPC loop for circulating slurry pH and an MPC control loop for outlet SO2 concentration, as detailed below: The MPC control loop for circulating slurry pH uses the combination of inlet flue gas flow, inlet SO2 concentration and circulating pump operation as disturbance variables, the slurry pump adjustment as the operating variable, the current and historical circulating slurry pH value and the circulating slurry at future time output by the prediction model as feedback input, and the circulating slurry pH as the controlled variable, and the control target is kept stable within the set value range. The MPC control loop for outlet SO2 concentration uses inlet flue gas parameters and the combination of recirculating pump start-stop as disturbance variables, and uses the current and historical outlet SO2 concentration values and the future outlet SO2 concentration output by the prediction model as feedback inputs. The operating variables include the start-stop combination of multiple recirculating pumps and the frequency of the variable frequency. The controlled variable is the outlet SO2 concentration, and the control objective is to control it within the set value range and not exceed the emission limit.
6. The fully integrated intelligent desulfurization technology method as described in claim 5, characterized in that, The optimization objectives are to achieve control accuracy of the circulating slurry pH value and outlet SO2 concentration relative to set values, as well as the smoothness of system operation, as detailed below: ； Among them, pH*, C out *These are the setpoints for the circulating slurry pH and the outlet SO2 concentration, respectively. , These are the weighting coefficients. To predict the step size, This is the controller's control window, and the control output window for each round of optimization.
7. The fully integrated intelligent desulfurization technology method as described in claim 5, characterized in that, The optimization objective function, which aims to minimize the overall process operating cost, is constructed as follows: ； ； In the formula, λ e As the weight for power consumption, E pump For the power consumption of the circulating pump, λ l For the weight of limestone consumption, c slueey λ is the slurry consumption coefficient. chem As the weight of drug consumption, Q waste For wastewater discharge, C chem For drug cost; λ em Weighting for penalties for exceeding emission standards.
8. A fully integrated intelligent desulfurization technology device, characterized in that, include: The data acquisition and processing module is used to collect key parameter data of the entire desulfurization process through the DCS system and smart instruments. The collected data were preprocessed and feature extracted to obtain key derived variables, including inlet SO2 mass flow rate and circulating slurry flow rate; The operating condition identification module is used to determine the current operating condition based on the preprocessed data, the threshold set based on the unit load and the rate of change of fuel sulfur content, and to determine the corresponding prediction model parameter set and control strategy based on the operating condition. The predictive modeling module is used to predict the inlet SO2 mass flow rate and circulating slurry pH value at a preset time, as well as the SO2 concentration at the absorber outlet, based on the preprocessed data and a preset model. The MPC controller operation module is used to design control loops for circulating slurry pH and outlet SO2 using a two-level MPC structure. The optimization objectives are the control accuracy of circulating slurry pH and outlet SO2 concentration relative to set values, as well as the smoothness of system operation. Under equipment and process constraints, the optimal control scheme is obtained, including slurry flow rate and circulating pump start-stop combination. The online input optimization control module is used to construct an optimization objective function with the goal of minimizing the overall process operating cost, under the premise of meeting SO2 emission standards and equipment constraints. It adopts a slow loop operation mode and outputs the optimal steady-state target as the MPC setpoint constraint. The collaborative execution module is used to decouple the obtained control scheme to the pulping, dewatering, wastewater and urea hydrolysis subsystems to obtain executable instructions, so as to realize the linkage and control of the entire system process.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the intelligent desulfurization full technology integration method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent desulfurization full technology integration method as described in any one of claims 1 to 7.