A collaborative optimization control method and system for reducing pollution and carbon emissions in the entire boiler process
By using the LSTM-GSA model to predict the controlled variables in the boiler combustion system and formulating control strategies, the problems of adjustment lag and energy waste in the boiler pollution reduction and carbon reduction control are solved, and the energy conservation and emission reduction effect of the whole process is achieved.
Patent Information
- Application Number
- CN202410547762.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-06
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-05-06
AI Technical Summary
The existing boiler pollution reduction and carbon reduction control methods have problems such as lag, energy waste and frequent pollutant emissions exceeding the standard. This is mainly due to the difficulty of manual control to respond to changes in working conditions in a timely manner, and the control parameters are highly coupled, resulting in waste of resources and secondary pollution. .
The prediction model based on the LSTM-GSA model is used to predict the controlled variables of the boiler combustion system, formulate control strategies, perform equipment adjustments in advance, shorten the adjustment lag time, and achieve energy conservation and emission reduction.
Through the prediction model, the control strategy is formulated in advance, the adjustment lag time is shortened, the reaction speed is improved, energy waste and pollutant emissions are reduced, and the coordinated optimization of pollution reduction and carbon reduction in the entire boiler process is achieved.
Smart Images

Figure CN118483904B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of energy and environmental engineering control, and in particular to a method and system for collaborative optimization control of pollution reduction and carbon reduction in the entire process of a boiler. Background Art
[0002] The control method for reducing pollution and carbon emissions in the boiler process is generally controlled by the DCS system, supplemented by manual control. After the operating conditions change, it is difficult to make timely adjustments and responses, which easily leads to large energy and material consumption. Manual adjustments by operators have slow response speed and delayed adjustment, and high requirements for operators (such as system familiarity, operating experience, etc.); and the control parameters are strongly coupled. After the operating conditions change, other parameters are not optimized and adjusted accordingly, and the overall changes in boiler operation are not considered, which easily leads to energy waste and frequent excessive pollutant emissions. The operating parameters and outlet concentrations of pollutant removal equipment are affected by load fluctuations and changes in coal quality. The level of manual control is limited. Sometimes, in order to meet ultra-low emission requirements, excessive materials are added or the operating power is increased when the equipment is running, resulting in excessive investment costs, which easily leads to waste of resources and secondary pollution. Summary of the invention
[0003] The purpose of the present invention is to provide a collaborative optimization control method and system for pollution reduction and carbon reduction in the entire process of a boiler. Each controlled variable can be predicted through the prediction model of each controlled variable, a control strategy for the boiler combustion system can be formulated, equipment can be adjusted in advance, and the adjustment lag time can be shortened, thereby achieving the effect of energy conservation and emission reduction.
[0004] To achieve the above object, the present invention provides the following solutions:
[0005] A boiler whole-process pollution reduction and carbon reduction coordinated optimization control method, the method comprising:
[0006] Based on the prediction model of each controlled variable of the boiler combustion system, each controlled variable is predicted respectively to obtain the predicted value of each controlled variable;
[0007] Formulate control strategies for boiler combustion systems based on the predicted values of each controlled variable;
[0008] The boiler combustion system is controlled based on the control strategy.
[0009] Optionally, the controlled variables include: oxygen content, SO2 concentration, NO x concentration, total outlet particulate matter concentration, total outlet SO3 concentration, boiler outlet smoke concentration and denitrification outlet SO3 concentration;
[0010] Among them, oxygen content, SO2 concentration, NO xThe prediction model for concentration, total outlet particulate matter concentration and total outlet SO3 concentration is built based on the LSTM-GSA model.
[0011] Optionally, the process of building a prediction model based on the LSTM-GSA model includes:
[0012] Determine the influencing parameters of the target controlled variables; the target controlled variables are: oxygen content, SO2 concentration, NO x concentration, total outlet particulate matter concentration or total outlet SO3 concentration;
[0013] The influencing parameters of the target controlled variable are input into the prediction model to obtain the predicted value of the target controlled variable; the prediction model is obtained by training the LSTM model based on the GSA algorithm.
[0014] Optionally, in the process of training the LSTM model based on the GSA algorithm, the particle position of the GSA algorithm is a parameter of the LSTM model;
[0015] The fitness function of the GSA algorithm is the mean square error between the predicted value and the actual value of the training sample through the LSTM model:
[0016] is the predicted value of the boiler combustion system target controlled variable obtained by the LSTM model for the i-th particle, y i is the actual value of the target controlled variable of the boiler combustion system, and N represents the number of training samples;
[0017] The particle velocity update formula of the GSA algorithm is:
[0018]
[0019] is the updated velocity of the ith particle in the t+1th iteration, rand i is a random number between 0 and 1, is the velocity of the ith particle before update in the tth iteration, is the acceleration of the i-th particle in the t-th iteration;
[0020] The particle position update formula of the GSA algorithm is:
[0021]
[0022] is the updated position of the i-th particle in the t+1th iteration, is the position of the i-th particle before update in the t-th iteration.
[0023] Optionally, the prediction model for the smoke concentration at the boiler outlet is:
[0024]
[0025] Among them, C (PM) represents the smoke concentration at the boiler outlet; Q represents the smoke volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; A ar Indicates the basic ash content of coal; Q net,ar It indicates the basic low calorific value of coal received; q4 indicates the incomplete combustion loss of boiler machinery; α fh Indicates the proportion of fly ash carried out by boiler exhaust; ΔM (CFB,PM) It indicates the amount of ash and slag produced by the desulfurizer required for a unit mass of coal in the circulating fluidized bed boiler furnace.
[0026] Optionally, the prediction model for the SO3 concentration at the denitrification outlet is:
[0027]
[0028] in, represents the SO3 concentration at the denitrification outlet; η1 represents the conversion rate of the basic sulfur in the coal to SO2; η2 represents the conversion rate of SO2 to SO3; S ar represents the basic sulfur content of coal; Q represents the flue gas volume at the boiler outlet; B g It represents the actual fuel consumption of the boiler; q4 represents the incomplete combustion loss of the boiler machinery.
[0029] Optionally, based on the predicted values of each controlled variable, a control strategy for the boiler combustion system is formulated, including:
[0030] Develop a control strategy to adjust the boiler's secondary fan speed based on the predicted oxygen content;
[0031] Based on export NO x The predicted value of the concentration is used to formulate a control strategy for adjusting the ammonia injection amount and the number of nozzles of the boiler's ammonia injection system;
[0032] Based on the predicted value of outlet SO2 concentration, formulate a control strategy to adjust the boiler slurry supply valve opening and oxidation fan damper opening;
[0033] Based on the total outlet particulate matter concentration and SO3 concentration predicted values, a control strategy is formulated to adjust the voltage of the boiler's dry electrostatic precipitator and wet electrostatic precipitator.
[0034] Optionally, based on the predicted value of outlet SO2 concentration, a control strategy for the boiler slurry supply valve opening is formulated, including:
[0035] The pH operation target value is obtained by back-calculating based on the predicted value of outlet SO2 concentration, and the pH difference is determined by the difference between the pH operation target value and the measured actual pH value;
[0036] Based on the pH difference, the amount of limestone required to eliminate the pH difference is calculated using the following formula:
[0037] S'=aΔ+b;
[0038] The boiler slurry valve opening is determined based on the amount of limestone required to eliminate the pH difference using the following formula:
[0039] L = 20 + 1.73 × (S + S' - J);
[0040]
[0041] In the formula, L represents the opening of the slurry supply valve, S' represents the amount of limestone required to eliminate the pH difference; △ represents the pH difference; a and b represent correction coefficients, which are continuously corrected according to big data analysis and actual conditions; S represents the theoretical consumption of limestone; Q gas Indicates the inlet flue gas volume; C SO2 represents the inlet SO2 concentration; η represents the desulfurization efficiency; θ represents the ratio of unit mole of calcium carbonate to sulfur dioxide; φ represents the calcium-sulfur ratio; k represents the limestone concentration; J represents the theoretical supply of limestone slurry; ρ represents the real-time density of the slurry.
[0042] A boiler full-process pollution reduction and carbon reduction collaborative optimization control system, the system comprising:
[0043] The controlled variable prediction module predicts each controlled variable based on the prediction model of each controlled variable of the boiler combustion system to obtain the predicted value of each controlled variable;
[0044] The control strategy formulation module formulates the control strategy of the boiler combustion system based on the predicted values of each controlled variable;
[0045] The boiler combustion system control module controls the boiler combustion system based on the control strategy.
[0046] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: each controlled variable is predicted respectively through the prediction model of each controlled variable, a control strategy for the boiler combustion system is formulated, equipment is adjusted in advance, the adjustment lag time is shortened, and the effect of energy conservation and emission reduction is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0048] Figure 1 A schematic flow chart of a method for collaborative optimization control of pollution reduction and carbon reduction in the entire process of a boiler provided by an embodiment of the present invention;
[0049] Figure 2 A flow chart of a collaborative optimization control method for reducing pollution and carbon emissions in the entire boiler process provided by an embodiment of the present invention;
[0050] Figure 3 LSTM-GSA model diagram of the boiler full-process pollution reduction and carbon reduction collaborative optimization control method provided by an embodiment of the present invention;
[0051] Figure 4 A schematic diagram of the module structure of the collaborative optimization control system for pollution reduction and carbon reduction in the entire boiler process provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0053] The purpose of the present invention is to provide a collaborative optimization control method and system for pollution reduction and carbon reduction in the entire process of a boiler. By predicting each controlled variable through the prediction model of each controlled variable, a control strategy for the boiler combustion system is formulated, and equipment adjustments are made in advance, thereby shortening the adjustment lag time and achieving the effect of energy conservation and emission reduction.
[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0055] Embodiment 1
[0056] like Figure 1-Figure 2 As shown, the present invention provides a boiler full-process pollution reduction and carbon reduction coordinated optimization control method, the method comprising:
[0057] S101. Based on the prediction model of each controlled variable of the boiler combustion system, each controlled variable is predicted to obtain the predicted value of each controlled variable; the controlled variables include: oxygen content, SO2 concentration, NO x concentration, total outlet particulate matter concentration, total outlet SO3 concentration, boiler outlet smoke concentration and denitrification outlet SO3 concentration; among them, the prediction models of oxygen content, SO2 concentration, NOx concentration, total outlet particulate matter concentration and total outlet SO3 concentration are built based on the LSTM-GSA model.
[0058] S102. Based on the predicted values of each controlled variable, formulate a control strategy for the boiler combustion system.
[0059] S103. Control the boiler combustion system based on the control strategy.
[0060] like Figure 3 As shown, as an optional implementation, the process of building a prediction model based on the LSTM-GSA model includes:
[0061] Determine the influencing parameters of the target controlled variables; the target controlled variables are: oxygen content, SO2 concentration, NO x concentration, total outlet particulate matter concentration or total outlet SO3 concentration.
[0062] The influencing parameters of the target controlled variable are input into the prediction model to obtain the predicted value of the target controlled variable; the prediction model is obtained by training the LSTM model based on the GSA algorithm.
[0063] The specific process of the prediction model is as follows:
[0064] After data preprocessing, the influencing parameters of the target controlled variable are input into the LSTM network model to obtain the predicted value of the target controlled variable. When training the LSTM model, the particle position (including the number of hidden layer nodes, learning rate, and network training times) is optimized through the GSA gravitational search algorithm, and the optimal value is returned to the LSTM model structure, and finally the optimal predicted value of the target controlled variable is obtained.
[0065] The target controlled variable prediction model of the boiler combustion system based on the LSTM-GSA model is as follows:
[0066] The measured values of the influencing parameters of the target controlled variables of the target controlled variables of the boiler combustion system at the current time t are obtained, including the properties of the fuel entering the furnace (such as low calorific value, ash content, volatile matter, sulfur content, etc.), boiler load, coal feed rate, primary air volume, secondary air volume, fan frequency, etc., and input them into the LSTM prediction model to output the predicted values of the target controlled variables of the boiler combustion system in the next period.
[0067] The LSTM model algorithm includes the input gate i t 、Forget gate t , output gate o t , cell unit C t The calculation rules of the modules are as follows: the forget gate mainly determines whether to forget or save the specific information of the previous moment; the input gate mainly controls the input of information; and the output gate mainly controls the output of information. Taking the oxygen content prediction as an example, the prediction formula is as follows:
[0068] i t =σW i H t-1 ,x t ]+b i ).
[0069] f t =σ(W f [H t-1 ,x t ]+b f ).
[0070] o t =σ(W o [H t-1 ,x t ]+b o ).
[0071] C' t =tanh(W c [H t-1 ,x t ]+b c ).
[0072] C t =f t *C t-1 +i t *C′ t .
[0073] H t =o t *tanh(C t ).
[0074] Where: x t is the measured value of the influencing parameter of the target controlled variable of oxygen content in the memory unit at time t, x t =(Boiler load(x 1t ), coal feeding amount (x 2t ), primary air volume (x 3t ), secondary air volume (x 4t ), fan frequency (x 5t )); H tis the predicted value of oxygen content in the memory unit at time t; σ represents the Sigmoid activation function; tanh represents the hyperbolic tangent activation function; W i , W f , W o are the weight matrices of different gates; b i 、b f 、b o 、b c are the correction vectors corresponding to different gates and memory cells at time t, C' t The updated value or short-term memory for the current iteration number.
[0075] In order to improve the prediction accuracy of the LSTM model, the GSA algorithm is used to optimize the parameters such as the 2-particle position (including the number of hidden layer nodes, learning rate and network training times) of the LSTM model, and the mean square error of the LSTM model is used as the particle fitness function.
[0076] In the process of training the LSTM model based on the GSA algorithm, the particle positions of the GSA algorithm are parameters of the LSTM model.
[0077] The fitness function of the GSA algorithm is the mean square error between the predicted value and the actual value of the training sample through the LSTM model:
[0078] is the predicted value of the boiler combustion system target controlled variable obtained by the LSTM model for the i-th particle, y i is the actual value of the target controlled variable of the boiler combustion system, and N represents the number of training samples.
[0079] The particle velocity update formula of the GSA algorithm is:
[0080]
[0081] is the updated velocity of the ith particle in the t+1th iteration, rand i is a random number between 0 and 1, is the velocity of the ith particle before update in the tth iteration, is the acceleration of the i-th particle in the t-th iteration.
[0082] The particle position update formula of the GSA algorithm is:
[0083]
[0084] is the updated position of the i-th particle in the t+1th iteration, is the position of the i-th particle before update in the t-th iteration.
[0085] The influencing parameters of the target controlled variables of the boiler outlet smoke concentration prediction model include fuel consumption, as-received ash content of coal, as-received lower calorific value of coal and boiler outlet smoke volume.
[0086] As an optional implementation, the prediction model of the smoke concentration at the boiler outlet is:
[0087]
[0088] Among them, C (PM) represents the smoke concentration at the boiler outlet; Q represents the smoke volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; A ar Indicates the basic ash content of coal; Q net,ar It indicates the basic low calorific value of coal received; q4 indicates the incomplete combustion loss of boiler machinery (should be calculated according to the data provided by the boiler factory); α fh Indicates the proportion of fly ash carried out by boiler exhaust (should be calculated according to the data provided by the boiler factory); ΔM (CFB,PM) It indicates the amount of ash and slag produced by the desulfurizer required for each unit mass of coal in the circulating fluidized bed boiler furnace (this item is not available when the boiler uses a pulverized coal furnace).
[0089] The influencing parameters of the target controlled variables of the denitrification outlet SO3 concentration prediction model include fuel consumption, as-received ash content of coal, as-received low calorific value of coal, boiler outlet flue gas volume, etc.
[0090] As an optional implementation, the prediction model of the SO3 concentration at the denitration outlet is:
[0091]
[0092] Wherein, C(SO3) represents the SO3 concentration at the denitration outlet; η1 represents the conversion rate of the basic sulfur in the coal to SO2; η2 represents the conversion rate of SO2 to SO3 (including the oxidation in boiler combustion and the catalytic oxidation of SCR denitration); S ar represents the basic sulfur content of coal; Q represents the flue gas volume at the boiler outlet; B g It represents the actual fuel consumption of the boiler; q4 represents the incomplete combustion loss of the boiler machinery.
[0093] As an optional implementation, a control strategy for the boiler combustion system is formulated based on the predicted values of each controlled variable, specifically including:
[0094] S1021. Based on the predicted value of oxygen content, formulate a control strategy for adjusting the speed of the secondary fan of the boiler.
[0095] Based on the predicted value of oxygen content, the secondary fan frequency adjustment strategy is issued in advance. When the oxygen content decreases, the frequency increase strategy is issued to increase the secondary fan speed. When the oxygen content increases, the wind reduction strategy is issued to reduce the secondary fan speed. After adjusting the secondary fan frequency, the measured values of the influencing parameters of the target controlled variables such as the secondary air volume and oxygen content detected in real time are fed back to the database for data update, and the oxygen content prediction value is optimized again. The secondary fan adjustment strategy is corrected and adjusted to ensure that the oxygen content fluctuates around the target value and reaches a stable state.
[0096] Based on the secondary fan adjustment increment, the fluctuation trend of the furnace pressure is predicted. When the furnace pressure exceeds the set range, the induced draft fan frequency is corrected through PID. The induced draft fan frequency adjustment increment is determined by fitting and analyzing the relationship between the primary air, secondary air, and induced draft fan frequencies. The adjustment formula is as follows:
[0097] △L 引 =0.55518△L 一次 +0.60237△L 二次 -0.00135L 引 +0.01363.
[0098] Where: △L 引 △L is the frequency control increment of the induced draft fan; 一次 is the frequency change of the primary fan; △L 二次 Secondary fan frequency change; L 引 It is the measured value of induced draft fan frequency.
[0099] After adjusting the induced draft fan frequency, the real-time detected values of the influencing parameters of the target controlled variables, such as the furnace pressure and induced draft fan frequency, are fed back to the database for data update, the oxygen content prediction value is optimized again, and the induced draft fan frequency adjustment strategy is corrected and adjusted to ensure that the furnace pressure fluctuates around the target value and reaches a stable state.
[0100] S1022. Based on export NO x The predicted value of the concentration is used to formulate a control strategy for adjusting the ammonia injection amount and the number of nozzles in the boiler's ammonia injection system.
[0101] Export NO x The influencing parameters of the target controlled variable of the concentration prediction value include the inlet NO x The prediction process is the same as the oxygen content prediction. x The predicted value of the concentration, combined with the outlet NO x The concentration target setting value and the measured value are used to issue the ammonia spray system adjustment strategy in advance, and the denitrification efficiency is improved by adjusting the ammonia spray amount and the number of nozzles of the spray system. At the same time, the furnace pressure, ammonia spray amount, flue gas temperature distribution, and inlet NO xThe influencing parameters of the target controlled variables such as concentration and oxygen content prediction value are fed back to the database for data update to optimize the outlet NO x The concentration prediction value is used to modify the regulation strategy of the ammonia injection system so that the outlet NO x The concentration is stabilized within the set target value range, reducing the loss of ammonia water.
[0102] S1023. Based on the predicted value of outlet SO2 concentration, formulate a control strategy for adjusting the boiler's slurry supply valve opening and oxidation fan damper opening.
[0103] As an optional implementation, step 1023 specifically includes:
[0104] Based on the predicted value of outlet SO2 concentration, formulate the control strategy of boiler slurry supply valve opening, including:
[0105] The pH operating target value is calculated based on the predicted value of the outlet SO2 concentration, and the pH difference is determined by the difference between the pH operating target value and the measured actual pH value.
[0106] The amount of limestone required to eliminate the pH difference was calculated based on the pH difference using the following formula.
[0107] S'=aΔ+b.
[0108] The boiler slurry supply valve opening is determined using the following formula based on the amount of limestone required to eliminate the pH difference.
[0109] L = 20 + 1.73 × (S + S' - J).
[0110]
[0111] In the formula, L represents the opening of the slurry supply valve, S' represents the amount of limestone required to eliminate the pH difference; △ represents the pH difference; a and b represent correction coefficients, which are continuously corrected according to big data analysis and actual conditions; S represents the theoretical consumption of limestone; Q gas Indicates the inlet flue gas volume; C SO2 represents the inlet SO2 concentration; η represents the desulfurization efficiency; θ represents the ratio of unit mole of calcium carbonate to sulfur dioxide; φ represents the calcium-sulfur ratio; k represents the limestone concentration; J represents the theoretical supply of limestone slurry; ρ represents the real-time density of the slurry.
[0112] The influencing parameters of the target controlled variable of the outlet SO2 concentration prediction value include the inlet SO2 concentration, oxygen content, limestone slurry flow rate, slurry pH value, etc. The prediction process is the same as the oxygen content prediction. Based on the outlet SO2 concentration prediction value, combined with the outlet SO2 concentration target setting value and the measured value, the slurry pH operation target value is optimized (according to the outlet SO2 concentration target setting value and the measured value, the pH is back-calculated through the outlet SO2 concentration prediction model as the next pH operation target value.) Ensure that the outlet SO2 concentration in the next step is within the set value ± 0.1 mg / m 3 Fluctuates within a range. According to the difference between the slurry pH operating target value and the actual pH value, the corresponding required limestone content is calculated, the slurry supply valve opening is determined, and the pH of the limestone slurry is accurately controlled in real time. Based on the predicted value of the outlet SO2 concentration and the trend of slurry pH changes, the oxidation fan adjustment strategy is issued in advance. The oxidation air volume is controlled by adjusting the damper opening of the oxidation fan through the electric regulating valve. There is a linear relationship between the damper opening and the air volume of the oxidation fan. The damper opening adjustment range is determined to be 25%-60%. When the damper is adjusted to 25%, the current of a single oxidation fan is reduced from 300A to about 275A, the power consumption reduction is about 10%, and the air volume of a single unit is reduced from 5000Nm 3 / h reduced to 4700Nm 3 / h or so.
[0113] S1024. Based on the total outlet particulate matter concentration and SO3 concentration predicted values, formulate a control strategy for adjusting the voltage of the boiler's dry electrostatic precipitator and wet electrostatic precipitator.
[0114] The influencing parameters of the target controlled variable of the total outlet SO3 concentration prediction value include the multi-field multi-channel operating voltage of the dry electrostatic precipitator, the number of limestone slurry spraying layers, the multi-field multi-channel operating voltage of the wet electrostatic precipitator, etc. The prediction process is the same as the oxygen content prediction. The influencing parameters of the target controlled variable of the total outlet particulate matter concentration prediction value include the boiler outlet smoke concentration prediction value, the multi-field multi-channel operating voltage of the dry electrostatic precipitator, the multi-field multi-channel operating voltage of the wet electrostatic precipitator, etc. The prediction process is the same as the oxygen content prediction.
[0115] Based on the predicted values of the total outlet particulate matter concentration and SO3 concentration, combined with the target setting value and measured value of the total outlet particulate matter concentration, the operating target values such as the multi-field and multi-channel operating voltage of the dry electrostatic precipitator and the multi-field and multi-channel operating voltage of the wet electrostatic precipitator are optimized to ensure that the total outlet particulate matter emission concentration and SO3 emission concentration in the next step are within the set value of ±0.1mg / m 3According to the target value and actual value of the multi-field and multi-channel operating voltage of the electrostatic precipitator, the multi-field and multi-channel power supply output is determined (when the current flue gas passes through the electric field area, the control program determines the current operating state and calculates the optimal voltage control value of the current system (multi-field and multi-channel environment). The control program is an internal company secret), and the multi-field and multi-channel power supply operating voltage is accurately controlled in real time.
[0116] The optimal voltage control value is calculated as follows:
[0117]
[0118] Wherein, α1-α3 are the electric zone coefficients, U is the electric zone power supply, I is the electric zone current, P and Q are the flue gas parameters flowing through the electric field, P and f are the induced draft fan parameters, and in and out are the temperatures before and after the absorption tower.
[0119] Real-time detection of influencing parameters of target controlled variables such as load, inlet SO2 concentration, slurry density, slurry pH, oxidation air volume, etc., feedback the measured values of influencing parameters of target controlled variables to the database for data update, optimize the predicted value of outlet SO2 concentration, optimize the slurry pH operating target value, correct and adjust the slurry supply valve and oxidation fan adjustment strategy, so that the outlet SO2 concentration is stable within the set target value range, reducing fan power consumption and limestone slurry loss.
[0120] Based on the predicted values of the target controlled variables of the combustion system, the changing trends of furnace pressure and pollutant emission concentration are considered in a coordinated manner, and control strategies are issued in advance to achieve stable operation of the entire boiler process.
[0121] Embodiment 2
[0122] like Figure 4 As shown, the present invention also discloses a boiler full-process pollution reduction and carbon reduction coordinated optimization control system, the system comprising:
[0123] The controlled variable prediction module predicts each controlled variable based on the prediction model of each controlled variable of the boiler combustion system, and obtains the predicted value of each controlled variable.
[0124] The control strategy formulation module formulates the control strategy of the boiler combustion system based on the predicted values of each controlled variable.
[0125] The boiler combustion system control module controls the boiler combustion system based on the control strategy.
[0126] The present invention predicts each controlled variable respectively through the prediction model of each controlled variable, formulates the control strategy of the boiler combustion system, performs equipment adjustment in advance, shortens the adjustment lag time, and achieves the effect of energy saving and emission reduction.
[0127] The present invention establishes the boiler combustion system target controlled variables (such as oxygen content, furnace pressure, furnace temperature, outlet SO2 concentration, outlet NO) based on the DCS system data reading. x The prediction model of target controlled variables (concentration, outlet particulate matter concentration, outlet SO3 concentration, etc.) is firstly used to collect the influencing parameters of target controlled variables such as boiler load, coal feed, primary air volume, secondary air volume, fan frequency, and coal quality entering the furnace through big data, and establish a database related to the target controlled variables. Then, the influencing parameter data of the target controlled variables are input into the LSTM-GSA algorithm model, and a prediction model of target controlled variables of the boiler combustion system is established to predict and optimize the target controlled variables. Finally, according to the predicted value of the optimal target controlled variable, a control strategy is formulated and issued in advance to adjust the equipment, thereby shortening the adjustment lag time and realizing the coordinated optimization of pollution reduction and carbon reduction in the whole process. The present invention uses the change trend prediction of the target controlled variables of the boiler combustion system to propose a control strategy for the key equipment of the combustion system in advance, improve the adjustment response speed, and reduce the pressure of manual adjustment and the requirements for manual skills. The present invention considers the correlation between the target controlled variables of the boiler combustion system, establishes relevant prediction models, collaboratively considers the impact on variables such as furnace pressure and original concentration of pollutants, globally considers the changes in boiler operation, and reduces furnace pressure, pollutant emission concentration and fluctuations. The present invention predicts and optimizes the target controlled variables of the boiler combustion system, proposes the best adjustment strategy for key equipment, and reduces the energy and material consumption caused by long-term start-up and shutdown of equipment.
[0128] The present invention is based on the idea of intelligent and precise control of multiple devices in the whole process of boiler combustion-flue gas treatment, and establishes a prediction-precise control model of the target controlled variables of the boiler combustion system and the multi-pollutant removal system to achieve coordinated optimization control of pollution reduction and carbon reduction in the whole process of the boiler. According to the changing trend of the fuel properties and the target controlled variables of the boiler combustion system, control instructions for key equipment such as secondary fans and induced draft fans are given in advance to keep the oxygen content and furnace pressure of the boiler stable in the optimal range. At the same time, the predicted value of the target controlled variable of the boiler combustion system is used to predict the NO at the boiler outlet in advance. xThe concentration, SO2 concentration, particulate matter concentration, and SO3 concentration changes were analyzed, and the cascade collaborative optimization and variable operating condition precision method of the boiler combustion-multi-pollutant removal system was further constructed. The pollutant concentration distribution of each section and the pollutant removal efficiency distribution of key devices were obtained, and then the ammonia / urea spray gun valve control instructions of the denitrification device, the multi-field, multi-channel, and multi-type power supply control instructions of the electrostatic precipitator, the limestone slurry circulation pump control instructions of the desulfurization device, and the multi-field, multi-channel, and multi-type power supply control instructions of the wet electrostatic precipitator were given in advance along the flue gas process. The key parameters of the boiler combustion system were coupled with the key parameters of the pollutant removal device to reduce the unit steam coal consumption and pollutant emissions throughout the process, achieving a unit steam / kWh coal consumption reduction of more than 1.5%, and the total outlet particulate matter concentration fluctuation ≤±0.11mg / m 3 , SO2 concentration fluctuation ≤ ±0.5mg / m 3 、NO x Concentration fluctuation ≤ ±0.5mg / m 3 , achieving high-precision control of pollutant emission concentration under variable load and variable fuel conditions, and collaboratively achieving SO3 emission concentration ≤ 1mg / m 3 The total emission concentration of five heavy metals, mercury, arsenic, lead, cadmium and chromium, is less than 20 μg / m 3 At the same time, the overall power consumption of the flue gas treatment system is reduced by more than 20%, and energy conservation and carbon reduction are achieved in a coordinated manner. The ammonia consumption is reduced by more than 20%, and the limestone consumption is reduced by more than 5%, achieving an effective reduction in CO2 emissions and realizing coordinated optimization of tiered pollution reduction and carbon reduction in the entire process.
[0129] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0130] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present invention. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A collaborative optimization control method for reducing pollution and carbon emissions in the entire boiler process. It is characterized in that The method comprises: Based on the prediction model of each controlled variable of the boiler combustion system, each controlled variable is predicted respectively to obtain the predicted value of each controlled variable; Based on the predicted values of each controlled variable, a control strategy for the boiler combustion system is formulated; the controlled variables include: oxygen content, SO 2 Concentration, NO x concentration, total outlet particulate matter concentration, total outlet SO 3 concentration, boiler outlet dust concentration and denitrification outlet SO 3 concentration; among them, oxygen content, SO 2 Concentration, NO x concentration, total outlet particulate matter concentration and total outlet SO 3 The prediction model of the concentration is built based on the LSTM-GSA model; among them, the prediction model of the smoke concentration at the boiler outlet is: Among them, C (PM) represents the smoke concentration at the boiler outlet; Q represents the smoke volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; A ar Indicates the basic ash content of coal; Q net,ar Indicates the basic low calorific value of coal; q 4 Indicates the incomplete combustion loss of boiler machinery; α fh Indicates the proportion of fly ash carried out by boiler exhaust; ΔM (CFB,PM) Indicates the amount of ash and slag produced by the desulfurizer required for a unit mass of coal in the circulating fluidized bed boiler furnace; Denitrification outlet SO 3 The prediction model for concentration is: Among them, C (SO3) Indicates the denitrification outlet SO 3 Concentration; η 1 Indicates that the sulfur in the coal is converted into SO 2 The conversion rate of 2 Indicates SO 2 To SO 3 Conversion rate; S ar represents the basic sulfur content of coal; Q represents the flue gas volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; q 4 Indicates incomplete combustion loss of boiler machinery; The boiler combustion system is controlled based on the control strategy.
2. According to claim 1, a boiler full-process pollution reduction and carbon reduction coordinated optimization control method, It is characterized in that The process of building a prediction model based on the LSTM-GSA model includes: Determine the influencing parameters of the target controlled variable; the target controlled variable is: oxygen content, SO 2 Concentration, NO x concentration, total outlet particulate matter concentration or total outlet SO 3 concentration; The influencing parameters of the target controlled variable are input into the prediction model to obtain the predicted value of the target controlled variable; the prediction model is obtained by training the LSTM model based on the GSA algorithm.
3. A boiler full-process pollution reduction and carbon reduction coordinated optimization control method according to claim 2, It is characterized in that In the process of training the LSTM model based on the GSA algorithm, the particle position of the GSA algorithm is a parameter of the LSTM model; The fitness function of the GSA algorithm is the mean square error between the predicted value and the actual value of the training sample through the LSTM model: is the predicted value of the boiler combustion system target controlled variable obtained by the LSTM model for the i-th particle, y i is the actual value of the target controlled variable of the boiler combustion system, and N represents the number of training samples; The particle velocity update formula of the GSA algorithm is: is the updated velocity of the ith particle in the t+1th iteration, rand i is a random number between 0 and 1, is the velocity of the ith particle before update in the tth iteration, is the acceleration of the i-th particle in the t-th iteration; The particle position update formula of the GSA algorithm is: is the updated position of the i-th particle in the t+1th iteration, is the position of the i-th particle before update in the t-th iteration.
4. A boiler full-process pollution reduction and carbon reduction coordinated optimization control method according to claim 1, It is characterized in that Based on the predicted values of each controlled variable, formulate the control strategy of the boiler combustion system, including: Develop a control strategy to adjust the boiler's secondary fan speed based on the predicted oxygen content; Based on export NO x The predicted value of the concentration is used to formulate a control strategy for adjusting the ammonia injection amount and the number of nozzles of the boiler's ammonia injection system; Based on export SO 2 Based on the predicted value of concentration, formulate control strategies for adjusting the boiler slurry supply valve opening and oxidation fan damper opening; Based on the total outlet particulate matter concentration, SO 3 Concentration prediction value, formulate control strategy for adjusting the voltage of boiler's dry electrostatic precipitator and wet electrostatic precipitator.
5. A boiler full-process pollution reduction and carbon reduction coordinated optimization control method according to claim 4, It is characterized in that Based on export SO 2 The predicted value of concentration is used to formulate the control strategy of the boiler slurry supply valve opening, including: Based on export SO 2 The predicted value of the concentration is back-calculated to obtain the pH operation target value, and the pH difference is determined by the difference between the pH operation target value and the measured actual pH value; Based on the pH difference, the amount of limestone required to eliminate the pH difference is calculated using the following formula: S'=aΔ+b; The boiler slurry valve opening is determined based on the amount of limestone required to eliminate the pH difference using the following formula: L = 20 + 1.73 × (S + S' - J); In the formula, L represents the opening of the slurry supply valve, S' represents the amount of limestone required to eliminate the pH difference; △ represents the pH difference; a and b represent correction coefficients, which are continuously corrected according to big data analysis and actual conditions; S represents the theoretical consumption of limestone; Q gas Indicates the inlet flue gas volume; C SO2 Indicates the entry SO 2 concentration; η represents desulfurization efficiency; θ represents the ratio of unit mole of calcium carbonate to sulfur dioxide; φ represents the calcium-sulfur ratio; k represents limestone concentration; J represents the theoretical supply of limestone slurry; ρ represents the real-time density of slurry.
6. A collaborative optimization control system for reducing pollution and carbon emissions throughout the boiler process. It is characterized in that The system comprises: The controlled variable prediction module predicts each controlled variable based on the prediction model of each controlled variable of the boiler combustion system to obtain the predicted value of each controlled variable; The control strategy formulation module formulates the control strategy of the boiler combustion system based on the predicted values of each controlled variable; the controlled variables include: oxygen content, SO 2 Concentration, NO x concentration, total outlet particulate matter concentration, total outlet SO 3 concentration, boiler outlet dust concentration and denitrification outlet SO 3 concentration; among them, oxygen content, SO 2 Concentration, NO x concentration, total outlet particulate matter concentration and total outlet SO 3 The prediction model of the concentration is built based on the LSTM-GSA model; among them, the prediction model of the smoke concentration at the boiler outlet is: Among them, C (PM) represents the smoke concentration at the boiler outlet; Q represents the smoke volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; A ar Indicates the basic ash content of coal; Q net,ar Indicates the basic low calorific value of coal; q 4 Indicates the incomplete combustion loss of boiler machinery; α fh Indicates the proportion of fly ash carried out by boiler exhaust; ΔM (CFB,PM) Indicates the amount of ash and slag produced by the desulfurizer required for a unit mass of coal in the circulating fluidized bed boiler furnace; Denitrification outlet SO 3 The prediction model for concentration is: in, Indicates the denitrification outlet SO 3 Concentration; η 1 Indicates that the sulfur in the coal is converted into SO 2 The conversion rate of 2 Indicates SO 2 To SO 3 Conversion rate; S ar represents the basic sulfur content of coal; Q represents the flue gas volume at the boiler outlet; B g Indicates the actual fuel consumption of the boiler; q 4 Indicates incomplete combustion loss of boiler machinery; The boiler combustion system control module controls the boiler combustion system based on the control strategy.
