A method for optimizing and managing a desulfurization system

By establishing a data and machine fusion model and edge management device of the desulfurization system, the combination of slurry circulation pumps is optimized, and the problems of high power consumption and unstable operation of slurry circulation pumps in the existing technology are solved, and the long-term continuous and normal operation of slurry circulation pumps is achieved. and power consumption is reduced.

CN115355163BActive Publication Date: 2024-11-26HANGU YUNZHI (WUHAN) TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210937365.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-05
Publication Date
2024-11-26
Estimated Expiration
2042-08-05

AI Technical Summary

Technical Problem

The existing power plant limestone-gypsum wet desulfurization system has a large power consumption, and the operation of the slurry circulation pump depends on experience adjustment, and it is impossible to achieve accurate monitoring and management, resulting in frequent start-and-stop increasing the risk of motor burnout, and it is impossible to ensure long-term continuous and normal operation.

Method used

By establishing a desulfurization efficiency model that integrates data of the desulfurization system with the machine, we optimize the slurry circulation pump combination in real time, count its start-stop time interval and cumulative operating time, monitor health status, and use edge management devices to make real-time adjustments to reduce total power consumption and ensure Long-term continuous operation.

Benefits of technology

The total current of the slurry circulation pump is reduced, the total power consumption of the desulfurization system is reduced, the long-term continuous and normal operation of the slurry circulation pump is ensured, and the problem of high power consumption caused by empirical adjustment is solved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115355163B_ABST
    Figure CN115355163B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for optimizing and managing a desulfurization system. A desulfurization efficiency model integrating data and machinery is established through the operation data of the desulfurization system. The model is used for real-time optimization of the desulfurization system in an actual production process. Under the premise of ensuring the desulfurization efficiency, an optimal slurry circulation pump combination is output, so that on-site operators can adjust the slurry circulation pump combination to reduce the total current of the slurry circulation pump and the total power consumption of the desulfurization system. The start and stop time intervals and the accumulated operation time of the slurry circulation pump are counted through an edge management device, and the health status of the slurry circulation pump is monitored at the same time to ensure the long-term continuous and normal operation of the slurry circulation pump, thereby solving the problem that the empirical adjustment method is relatively extensive and causes large power consumption for desulfurization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of wet desulfurization technology, and more specifically, to a method for optimizing and managing a desulfurization system. Background Technology

[0002] The existing limestone-gypsum wet desulfurization system in power plants has a plant power consumption rate of around 1.5%, resulting in significant power consumption. The main power consumption of the desulfurization system is concentrated in the slurry circulation pump. The operation and combination of the slurry circulation pump rely entirely on the operating experience of the desulfurization operators. As the unit load is adjusted, the SO2 concentration at the outlet of the desulfurization system also fluctuates. This requires corresponding adjustments to the operating parameters of the desulfurization system and the combination of the slurry circulation pump. Precise management can ensure that the SO2 emissions of the desulfurization system meet the standards.

[0003] The defects of the existing technology are as follows: (1) At present, only the operating parameters of the slurry circulation pump, such as bearing temperature, motor coil temperature, and pump operating current, are detected. The comprehensive performance of the slurry circulation pump is not evaluated, nor is the cumulative operating time of the slurry circulation pump statistically analyzed. Therefore, accurate monitoring and management cannot be achieved. (2) The motor power of the slurry circulation pump is relatively large. The short start-stop interval or frequent start-stop of the same slurry circulation pump will have a large impact on the electrical circuit of the desulfurization system, affecting the electrical safety of the desulfurization system. Frequent start-stop of the slurry circulation pump will cause the motor to burn out. The health status and start-stop interval of the slurry circulation pump can only be judged by estimation. It is impossible to guarantee the long-term continuous normal operation of the slurry circulation pump. (3) The method of adjusting the desulfurization system using operating experience is relatively crude and consumes a lot of manpower, resulting in a large desulfurization power consumption. Summary of the Invention

[0004] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides an optimization and management method for desulfurization systems. It establishes a data-machine integrated desulfurization efficiency model based on the operating data of the desulfurization system, which is used in actual production processes to optimize the desulfurization system in real time. While ensuring desulfurization efficiency, it reduces the total current of the slurry circulation pumps, thereby lowering the total power consumption of the desulfurization system. Simultaneously, it outputs the optimal slurry circulation pump combination for on-site operators to adjust. An edge management device statistically analyzes the start-stop intervals and cumulative operating time of the slurry circulation pumps, and monitors their health status to ensure long-term continuous normal operation. This solves the problem of relatively crude experience-based adjustment methods leading to high desulfurization power consumption.

[0005] To address the above problems, this invention provides a method for optimizing and managing a desulfurization system, comprising the following steps:

[0006] S100: Real-time acquisition of the current of the slurry circulation pump; comparison of the current value with the rated current of the pump motor to determine the operating status of the slurry circulation pump; accumulation of the operating time when the pump is running; accumulation of the shutdown time when the pump is stopped; setting the minimum start-stop interval of the slurry circulation pump to a hours; if the start-stop interval is less than a hours, the slurry circulation pump will not be allowed to start, and the start-stop interval of the slurry circulation pump will be displayed.

[0007] S200 collects historical operating data of the desulfurization system, analyzes and preprocesses the collected operating data to obtain model data and test data;

[0008] The desulfurization efficiency model expression for S300 is: η(k, θ) = f(a(k), X(k)), where η(k, θ) is the desulfurization efficiency at time k for the θ-th slurry circulation pump combination, X(k) is the operating variable vector, and a(k) is the model coefficient at time k, i.e., the regression coefficient. A recursive formula for the model coefficients is proposed, and a first-order difference relationship for the desulfurization efficiency is established. First difference relationship with each independent variable The mathematical expression between them is given by and Establish a time-varying first-order difference linear regression model: Where a(k) is the time k. The model coefficient matrix is ​​given by , b is the constant term matrix, and ε is the fitting error term. The model coefficients a(k) are obtained by training a time-varying first-order difference linear regression model. A time-varying algorithm is introduced into the first-order difference linear regression model, and the model is always trained using m data points before the current time.

[0009] Then, with the optimization objectives of minimizing the total current and switching frequency of the slurry circulation pump, and with SO2 emission requirements and upper and lower bounds of the optimization variables as constraints, a combined optimization model for the desulfurization system's slurry circulation pumps was established. The optimization variables in this model include continuous variables: unit load (MW) and flue gas volume (Nm³). 3 / h; SO2 concentration in the inlet flue gas, in mg / Nm³. 3 Liquid-to-gas ratio, in L / Nm³ 3 pH value of slurry; density of slurry, in kg / m³ 3 And a binary integer variable representing the operating state of the circulating pump. When the binary integer variable is 1, the pump is in normal condition; when it is 0, the pump is in fault condition. This operation optimization model is a mixed integer nonlinear programming problem.

[0010] S400, using a queue competition algorithm that simulates the laws of nature to solve the above multi-objective optimization problem of mixed integer nonlinear programming, outputs the total current C(θ) of the slurry circulation pump and the SO2 concentration:

[0011] Substituting the desulfurization operating parameters at time k into the desulfurization efficiency model, the desulfurization efficiency of each circulating pump combination is calculated. The circulating pump combination with the θth slurry circulation pump combination where η(k, θ) > η (i.e., the desulfurization efficiency at time k is greater than η) is selected as the compliant circulating pump combination. Then, the total current of the compliant circulating pump combination is calculated. The combination with the minimum total current of the slurry circulation pumps and the slurry circulation pumps in operation state 1 is selected as the optimal slurry circulation pump combination. Based on the desulfurization efficiency η of the optimal slurry circulation pump combination under the current operating conditions... θ The calculated SO2 concentration is used as the predicted value of SO2 concentration at the outlet of the absorber tower. The following expression is proposed for the combined optimization model of the desulfurization system slurry circulation pump under constraints: J=min(C(θ),η(k,θ));

[0012] Calculate (SO2)out, which is the SO2 concentration at the outlet of the absorption tower;

[0013] S500 inputs test data into the slurry circulation pump combination optimization model to obtain optimized data. The model is then verified by comparing the test data with the optimized data.

[0014] Furthermore, the recursive formula for the model coefficients in S300 is as follows:

[0015] Take the m data points before time k, arranged in reverse chronological order for the θ-th slurry circulation pump combination, i.e., the desulfurization efficiency values ​​η1, η2, ..., η m The values ​​of each variable [x] 1,1 ,x 1,2 ,…,x 1,m ] T 、[x 2,1 ,x 2,2 ,…,x 2,m ] T ...[x n,1 ,x n,2 ,…,x n,m ] T Establish the first-order difference relationship of desulfurization efficiency First difference relationship with each independent variable Establish and The mathematical expression:

[0016]

[0017] Furthermore, the constraints in S400 include:

[0018] P-δ1<χ1<P+δ1, where P is the current unit load in MW and δ1 is the fluctuation range;

[0019] Q-δ2<χ2<Q+δ2, where Q is the current flue gas flow rate in Nm³. 3 / h, δ2 is the fluctuation range;

[0020] S-δ3<χ3<S+δ3, where S is the current SO2 concentration in the inlet flue gas, in mg / Nm³. 3 δ3 represents the fluctuation range;

[0021] pH-δ4<χ4<pH+δ4, where pH is the current pH value of the slurry and δ4 is the fluctuation range;

[0022] Λ-δ5<χ5<Λ+δ5, where Λ is the current slurry density in kg / m³ 3 δ5 represents the fluctuation range;

[0023] C(θ) = C1 + C2 + C3 + ... + C n C(θ) represents the total current of the slurry circulation pumps in the θ-th slurry circulation pump combination, in A; C1 represents the current of slurry circulation pump No. 1, in A; C2 represents the current of slurry circulation pump No. 2, in A. n The current of slurry circulation pump n is given in amperes (A); the desulfurization efficiency η is proposed. θ Relationship with SO2 concentration.

[0024] Furthermore, the desulfurization efficiency η θ The relationship with SO2 concentration is as follows: Among them (SO2) in (SO2)out represents the SO2 concentration at the inlet of the absorption tower, and (SO2)out represents the SO2 concentration at the outlet of the absorption tower.

[0025] Furthermore, S100 includes: when the flue gas conditions at the inlet of the desulfurization system fluctuate, it is necessary to adjust the combined state of the slurry circulation pump. An edge management device is used to manage the health status of the slurry circulation pump, evaluate the health status of the pump in real time, collect the bearing temperature, motor coil temperature and operating current data of the slurry circulation pump and store them in a database. The edge management device directly obtains data from the database, counts the cumulative operating time of the slurry circulation pump, and prompts the operator to inspect, maintain and service the slurry circulation pump when the bearing temperature or coil temperature exceeds the set value, the operating current of the slurry circulation pump is lower than the set value, or the cumulative operating time of the slurry circulation pump exceeds the set value.

[0026] Furthermore, after collecting the operating data in S200, the operating data is first subjected to steady-state detection through the desulfurization mechanism to remove outliers and merge parameters of the same type with consistent trends. The data is then subjected to dimensionality reduction processing. Finally, the data is grouped according to different slurry circulation pump combinations. 80% of the pre-processed data is used as model data to build the model, and the remaining 20% ​​of the data is used as test data to test the model.

[0027] Furthermore, the operating variable vector includes the current of all slurry circulation pumps, in amperes (A); the unit load P, in MW; and the flue gas flow rate Q, in Nm³. 3 / h; SO2 concentration S in inlet flue gas, in mg / Nm³ 3 Flue gas oxygen content, in %; liquid-to-gas ratio, in L / Nm³ 3 ; Slurry pH value and slurry density Λ, in kg / m³ 3 Oxidation air flow rate, unit: Nm³ 3 / h.

[0028] Furthermore, the standard for verifying the model in S400 is: the outlet SO2 concentration is less than the limit value, and the total current of the optimized slurry circulation pump is not higher than the current total current value, avoiding frequent start-up and shutdown of the slurry circulation pump. If all pump combinations fail to meet the standard under the current operating conditions, the system outputs that the current operating conditions exceed the desulfurization limit and provides an alarm prompt.

[0029] Furthermore, the operational data in S200 includes flue gas volume, in Nm³. 3 / h; SO2 concentration in the flue gas at the desulfurization tower inlet, in mg / Nm³. 3 Liquid-to-gas ratio, in L / Nm³ 3 ; Slurry pH; Oxidation air volume, in Nm³ 3 / h, slurry density, unit: kg / m³ 3 Oxygen content in flue gas, in %; Slurry circulation pump current, in A; Unit load, in MW.

[0030] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0031] 1. This invention provides an optimization and management method for a desulfurization system. The method includes a desulfurization optimization method and a management method. It establishes a data-machine integrated desulfurization efficiency model based on the operating data of the desulfurization system. This model is used in actual production to optimize the desulfurization system in real time. While ensuring desulfurization efficiency, it outputs the optimal slurry circulation pump combination for on-site operators to adjust. This reduces the total current of the slurry circulation pumps, thereby reducing the total power consumption of the desulfurization system. An edge management device statistically analyzes the start-stop interval and cumulative operating time of the slurry circulation pumps, and monitors the health status of the pumps to ensure long-term continuous normal operation. This solves the problem of relatively crude experience-based adjustment methods that result in high power consumption during desulfurization.

[0032] 2. This invention provides an optimization and management method for a desulfurization system. This optimization method collects several historical operating data points of the desulfurization system, mainly including flue gas volume, SO2 concentration in the flue gas at the desulfurization tower inlet, liquid-to-gas ratio, slurry pH, oxidation air volume, slurry density, flue gas oxygen content, current of all slurry circulation pumps, and unit load. Steady-state detection is performed on the operating data, outliers are removed, parameters with consistent trends are merged, and the data undergoes dimensionality reduction. Finally, based on different slurry circulation pump combinations, the data is grouped to obtain data for establishing an optimization model, as well as test data. The test data is input into the model for verification. If the model fails to meet the standards, the current operating condition exceeds the desulfurization limit, and an alarm is triggered. Attached Figure Description

[0033] Figure 1 A flowchart of a desulfurization system optimization and management method is provided for an embodiment of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0035] like Figure 1 As shown, the present invention provides a method for optimizing and managing a desulfurization system, specifically including the following steps:

[0036] S100: Real-time acquisition of the slurry circulation pump current, comparison of the current value with the rated current of the pump motor to determine whether the slurry circulation pump is in operation. When the pump is in operation, its running time is accumulated; when the pump is stopped, its stopping time is accumulated. The minimum start-stop interval T of the slurry circulation pump is set to a hours. If the start-stop interval is less than a hours, the slurry circulation pump will not be allowed to start. The start-stop interval of the slurry circulation pump is also displayed for the convenience of operators.

[0037] S101: When the flue gas conditions at the inlet of the desulfurization system fluctuate, it is necessary to adjust the combination status of the slurry circulation pump. An edge management device is used to manage the health status of the slurry circulation pump, evaluate the health status of the pump in real time, and collect data such as bearing temperature, motor coil temperature and operating current of the slurry circulation pump and store them in the database. The edge management device directly obtains data from the database and counts the cumulative running time of the slurry circulation pump. When the bearing temperature or coil temperature exceeds the set value, or the operating current of the slurry circulation pump is lower than the set value, or the cumulative running time of the slurry circulation pump exceeds the set value, the operator is prompted to inspect, maintain and service the slurry circulation pump.

[0038] S200: Collects historical operating data from multiple desulfurization systems, including flue gas volume (Nm³). 3 / h), SO2 concentration in flue gas at the desulfurization tower inlet (mg / Nm 3 ), liquid-to-gas ratio (L / Nm 3 ), slurry pH, oxidation air volume (Nm³) 3 / h), slurry density (kg / m 3 ), flue gas oxygen content (%), slurry circulation pump current (A), unit load (MW), etc.

[0039] The collected operational data were analyzed and preprocessed. First, steady-state detection of the operational data was performed through the desulfurization mechanism to remove outliers and merge parameters of the same type with consistent trends. The data was then subjected to dimensionality reduction. Finally, the data was grouped according to different combinations of slurry circulation pumps. 80% of the preprocessed data was used as model data to build the model, and the remaining 20% ​​of the data was used as test data to test the model.

[0040] S300: Based on the application scenario of limestone-gypsum wet desulfurization of 600MW units, the desulfurization efficiency model expression is proposed as: η(k, θ)=f(a(k),X(k));

[0041] In the expression, η(k) is the desulfurization efficiency at time k for the θ-th combination of slurry circulation pumps, and X(k) is the operating variable vector, which includes: the current (A) of all slurry circulation pumps, the unit load P (MW), and the flue gas flow rate Q (Nm³). 3 / h), SO2 concentration in inlet flue gas S (mg / Nm 3 ), flue gas oxygen content (%), liquid-to-gas ratio (L / Nm³) 3 ), pH value of slurry, density of slurry Λ (kg / Nm³) 3 ), Oxidation air flow rate (Nm 3 / h), etc., where a(k) is the model coefficient (i.e., regression coefficient) at time k, and the recursive formula for the model coefficient is as follows:

[0042] Take the m data points before time k, arranged in reverse chronological order for the θ-th slurry circulation pump combination, i.e., the desulfurization efficiency values ​​η1, η2, ..., η m The values ​​of each variable [x] 1,1 ,x 1,2 ,…,x 1,m ] T 、[x 2,1 ,x 2,2 ,…,x 2,m ] T ...[x n,1 ,x n,2 ,…,x n,m ] T Establish the first-order difference relationship of desulfurization efficiency First difference relationship with each independent variable and The mathematical expression is as follows:

[0043]

[0044] Depend on and Establish a time-varying first-order difference linear regression model: Where a(k) is the model coefficient matrix of X at time k, b is the constant term matrix, and ε is the fitting error term. The model coefficients a(k) are obtained by training a time-varying first-order difference linear regression model. A time-varying algorithm is introduced into the first-order difference linear regression model, which always uses m data points before the current time to train the model, in order to reduce the information content of old data and improve the effectiveness of new data.

[0045] Then, with the optimization objectives of minimizing the total current of the slurry circulation pump and the minimum number of switching operations, and with SO2 emission requirements met and the upper and lower bounds of the optimization variables as constraints, a combined optimization model for the desulfurization system's slurry circulation pumps was established. This model includes continuous variables in its optimization variables, namely, unit load (MW) and flue gas volume (Nm³). 3 / h), SO2 concentration in inlet flue gas (mg / Nm³) 3 ), liquid-to-gas ratio (L / Nm 3 ), pH value of slurry, density of slurry (kg / m³) 3 The model includes variables such as 1 (when the value is 1, the pump is in normal condition; when the value is 0, the pump is in fault condition), and a binary integer variable representing the operating state of the circulating pump (when the value is 1, the pump is in normal condition; when the value is 0, the pump is in fault condition). Therefore, the operation optimization model is a mixed integer nonlinear programming problem.

[0046] S400: The queue competition algorithm simulating the laws of nature is used to solve the multi-objective optimization problem of the mixed integer nonlinear programming above, outputting the optimized combination of slurry circulation pumps and the optimized total current C(θ), while predicting the SO2 concentration at the desulfurization system outlet. The specific process is as follows:

[0047] Substituting the desulfurization operating parameters at time k into the desulfurization efficiency model, the desulfurization efficiency of each circulating pump combination is calculated. The circulating pump combination with η(k, θ) > η (the θth slurry circulation pump combination when the desulfurization efficiency at time k is greater than η) is selected as the compliant circulating pump combination. Then, the total current of the compliant circulating pump combination is calculated. The combination with the minimum total current of the slurry circulation pumps and the slurry circulation pumps in operation state 1 is selected as the optimal slurry circulation pump combination. Based on the desulfurization efficiency η of the optimal slurry circulation pump combination under the current operating conditions... θ The calculated SO2 concentration is used as the predicted value of SO2 concentration at the outlet of the absorption tower. The expression of the desulfurization system slurry circulation pump combination optimization model is: J=min(C(θ),η(k,θ));

[0048] The constraints are:

[0049] P-δ1<χ1<P+δ1, where P is the current unit load (MW) and δ1 is the fluctuation range;

[0050] Q-δ2<χ2<Q+δ2, where Q is the current flue gas flow rate (Nm³). 3 / h), where δ2 is the fluctuation range;

[0051] S-δ3<χ3<S+δ3, where S is the current SO2 concentration in the inlet flue gas (mg / Nm³). 3 ), where δ3 is the fluctuation range;

[0052] pH-δ4<χ4<pH+δ4, where pH is the current pH value of the slurry and δ4 is the fluctuation range;

[0053] Λ-δ5<χ5<Λ+δ5, where Λ is the current slurry density (kg / m³) 3 ), where δ5 is the fluctuation range;

[0054] C(θ) = C1 + C2 + C3 + ... + C nC(θ) is the total current (A) of the slurry circulation pumps in the θ-th slurry circulation pump combination; C1 is the current (A) of the first slurry circulation pump; C2 is the current (A) of the second slurry circulation pump. n Let n be the current (A) of the slurry circulation pump.

[0055] Desulfurization efficiency η θ The relationship with SO2 concentration is as follows: Among them (SO2) in (SO2)out represents the SO2 concentration at the inlet of the absorption tower, and (SO2)out represents the SO2 concentration at the outlet of the absorption tower.

[0056] S500: Input the test data into the slurry circulation pump combination optimization model to obtain optimized data. Compare the test data with the optimized data to verify the model. The verification standard is: the outlet SO2 concentration meets the standard, less than 35 mg / Nm³. 3 Furthermore, the optimized total current of the slurry circulation pump should not exceed the current total current value, and frequent start-stop of the slurry circulation pump should be avoided (the start-stop interval of the same slurry circulation pump should not be less than a hours). If all pump combinations fail to meet the standards under the current operating conditions, the system will output that the current operating conditions exceed the desulfurization limit and give an alarm prompt.

[0057] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing and managing a desulfurization system, It is characterized in that The following steps are involved: S100, real-time acquisition of the current of the slurry circulation pump, comparison of the current value with the rated current of the pump motor, determination of the operating state of the slurry circulation pump, accumulation of the operating time when the pump is in the operating state; accumulation of the pump stop time when the pump is in the stop state; setting the minimum start-stop time interval of the slurry circulation pump to a hour, if the start-stop time interval is less than a hour, the slurry circulation pump will not be allowed to start, and the start-stop time interval of the slurry circulation pump will be displayed; S200, collecting historical operation data of the desulfurization system, analyzing and preprocessing the collected operation data, and obtaining model data and test data; S300, the desulfurization efficiency model expression is: η(k, θ) = f(a(k), X(k)), where η(k, θ) is the desulfurization efficiency at time k and the θth slurry circulation pump combination, X(k) is the operating variable vector, and a(k) is the model coefficient at time k, i.e., the regression coefficient. The recursive formula of the model coefficient is proposed to establish the first-order difference relationship of the desulfurization efficiency. First-order difference relationship with each variable The mathematical expression between and Build a first-order difference linear regression model with time variation: Where a(k) is the kth moment The model coefficient matrix is ​​, b is the constant term matrix, and ε is the fitting error term; the model coefficient a(k) is obtained by training the first-order difference linear regression model with time variation; the time variation algorithm is introduced into the first-order difference linear regression model, and the m data before the current moment are always used to train the model; Then, the optimization objectives are to minimize the total current of the slurry circulation pump and the number of switching times, and to minimize the number of switching times. 2 The combined optimization model of the slurry circulation pump of the desulfurization system is established by taking the upper and lower bounds of the optimization variables as constraints to meet the emission requirements. The optimization variables of this model include continuous variables, namely, the unit load χ 1 , in MW; flue gas flow rate χ 2 , unit is Nm 3 / h; SO in inlet flue gas 2 The concentration of 3 , unit is mg / Nm 3 ; Liquid-gas ratio, unit is L / Nm 3 ; pH value of slurry x 4 ; Slurry density χ 5 , unit is kg / m 3 , and a binary integer variable representing the operating state of the circulating pump. When the binary integer variable takes 1, the pump is in normal state, and when it takes 0, the pump is in failure. The combined optimization model is a mixed integer nonlinear programming problem; S400, using the queue competition algorithm simulating the law of nature to solve the above multi-objective optimization problem of mixed integer nonlinear programming, output the total current C(θ) and SO of the slurry circulation pump 2 concentration: Substitute the desulfurization operation parameters at time k into the desulfurization efficiency model, calculate the desulfurization efficiency of each circulation pump combination, select the θth slurry circulation pump combination when η(k, θ)>η, that is, when the desulfurization efficiency at time k is greater than η, as the qualified circulation pump combination, and then calculate the total current of the qualified circulation pump combination, and take the combination with the smallest total current of the slurry circulation pump and the operating state of the opened slurry circulation pump as 1 as the optimal slurry circulation pump combination. According to the desulfurization efficiency η of the optimal slurry circulation pump combination under the current operating conditions θ Calculated SO 2 The concentration is the SO at the outlet of the absorption tower. 2 Concentration prediction value, proposed constraint conditions and expression of combined optimization model of slurry circulation pump of desulfurization system: J = min (C (θ), η (k, θ)); Calculation (SO 2 )out, that is, SO at the outlet of the absorption tower 2 concentration; S500, input the test data into the slurry circulation pump combination optimization model to obtain optimization data, compare the test data with the optimization data, and verify the model.

2. A method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that The recursive formula of the model coefficient in S300 is as follows: Take the m data before time k of the θth slurry circulation pump combination after chronological reverse order, that is, the desulfurization efficiency value η 1 ,η 2 ,…,η m , the value of each variable [x 1,1 ,x 1,2 ,…,x 1,m ] T ,[x 2,1 ,x 2,2 ,…,x 2,m ] T ……[x n,1 ,x n,2 ,…,x n,m ] T , establish the first-order difference relationship of desulfurization efficiency First-order difference relationship with each variable Establish and The mathematical expression is:

3. The optimization and management method of a desulfurization system according to claim 1, It is characterized in that The constraints in S400 include: P-δ 1 < 1 <P+δ 1 , P is the current unit load, in MW, δ 1 is the fluctuation range; Q-δ 2 < 2 <Q+δ 2 , Q is the current flue gas flow rate, in Nm 3 / h,δ 2 is the fluctuation range; S-δ 3 < 3 <S+δ 3 , S is the current SO in the inlet flue gas 2 Concentration, in mg / Nm 3 , δ 3 is the fluctuation range; pH-δ 4 < 4 <pH+δ 4 , χ 4 is the pH value of the slurry, pH is the current pH value of the slurry, δ 4 is the fluctuation range; Λ-δ 5 < 5 <Λ+δ 5 , χ 5 is the slurry density, Λ is the current slurry density, the unit is kg / m 3 , δ 5 is the fluctuation range; C(θ)=C 1 +C 2 +C 3 +…+C n , C(θ) is the total current of the slurry circulation pump of the θth slurry circulation pump combination, in A, C 1 is the current of No. 1 slurry circulation pump, in A; C 2 is the current of No. 2 slurry circulation pump, in A, C n is the current of the slurry circulation pump No. n, in A; the desulfurization efficiency η θ and SO 2 Concentration relationship.

4. The method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that Desulfurization efficiency η θ and SO 2 The concentration relationship is: Among them (SO 2 ) in The absorption tower inlet SO 2 concentration, (SO 2 )out is the SO at the outlet of the absorption tower 2 concentration.

5. The method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that The S100 includes: when the flue gas conditions at the inlet of the desulfurization system fluctuate, it is necessary to adjust the combination state of the slurry circulation pump, use the edge management device to manage the health state of the slurry circulation pump, evaluate the health state of the circulation pump in real time, collect the bearing temperature, motor coil temperature and operating current data of the slurry circulation pump and store them in the database, the edge management device directly obtains data from the database, and counts the cumulative operating time of the slurry circulation pump. When the bearing temperature or coil temperature exceeds the set value, or the operating current of the slurry circulation pump is lower than the set value, or the cumulative operating time of the slurry circulation pump exceeds the set value, the operator is prompted to inspect, maintain and service the slurry circulation pump.

6. A method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that After the operation data is collected in S200, the operation data is firstly subjected to steady-state detection through the desulfurization mechanism, outliers in the data are eliminated, parameters of the same type with consistent change trends are merged, and the data is subjected to dimensionality reduction processing. Finally, the data is grouped and processed according to different slurry circulation pump combinations. 80% of the pre-processed data is used as model data for building the model, and the remaining 20% ​​of the data is used as test data for testing the model.

7. A method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that The operating variable vector includes all slurry circulation pump currents, in A; unit load P, in MW; Flue gas flow rate Q, in Nm 3 / h; inlet flue gas SO 2 Concentration S, unit is mg / Nm 3 ; Flue gas oxygen content, in %; Liquid-gas ratio, in L / Nm 3 ; Slurry pH value, slurry density Λ, unit is kg / m 3 ; Oxidation air flow rate, in Nm 3 / h.

8. The method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that The standard for verifying the model in S400 is: export SO 2 The concentration is less than the limit value, and the total current of the optimized slurry circulation pump is not higher than the current total current value, and frequent starting and stopping of the slurry circulation pump is avoided. If all pump combinations do not meet the standards under the current operating conditions, the system outputs that the current operating conditions exceed the desulfurization limit and gives an alarm prompt.

9. The method for optimizing and managing a desulfurization system according to claim 1, It is characterized in that The operating data in S200 includes the flue gas volume, in Nm 3 / h; desulfurization tower inlet flue gas SO 2 Concentration, in mg / Nm 3 ; Liquid-gas ratio, unit is L / Nm 3 ; pH of slurry; oxidation air volume, in Nm 3 / h, slurry density, unit is kg / m 3 ; Flue gas oxygen content, unit is %; Slurry circulation pump current, unit is A; Unit load, in MW.