CN113341716B - Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system - Google Patents
Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system Download PDFInfo
- Publication number
- CN113341716B CN113341716B CN202110646133.6A CN202110646133A CN113341716B CN 113341716 B CN113341716 B CN 113341716B CN 202110646133 A CN202110646133 A CN 202110646133A CN 113341716 B CN113341716 B CN 113341716B
- Authority
- CN
- China
- Prior art keywords
- coal
- power plant
- fired power
- trapping
- trapping system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention is suitable for the technical field of thermal process optimization scheduling, and provides a large-scale coal-fired power plant CO based on artificial intelligence 2 The trapping system optimizing and scheduling method comprises the following steps: construction of large-scale coal-fired power station CO by Deep Belief Network (DBN) and other artificial intelligent methods 2 Capturing system data to drive a steady-state model; consider large coal-fired power plant CO 2 Capturing generating capacity and CO of system unit 2 The emission and operation cost is used for constructing an objective function which can fully reflect the economic index and the operation constraint of the whole system; according to related information such as electricity price and carbon price, a Bayesian optimization method (BO) is utilized to realize efficient real-time solving of the objective function. The invention can give consideration to the CO of the coal-fired power plant 2 The operation requirements of the trapping system in the aspects of economy, safety and environmental protection find out the optimal scheduling mode meeting the minimum total operation cost, and realize the CO of the large-scale coal-fired power plant 2 The trapping system operates optimally.
Description
Technical Field
The invention belongs to the technical field of thermal process optimization scheduling, and particularly relates to a large-scale coal-fired power plant CO based on artificial intelligence 2 The trapping system optimizes the scheduling method.
Background
To achieve the ambitious goal of '2030 carbon peak and 2060 carbon neutralization', CO 2 Trapping and reducing CO of industrial system by using main greenhouse gases 2 Emissions are currently an important technical means. The coal-fired thermal power unit is CO in China 2 The most dominant source of emissions, national CO in 2018 2 The total emission amount is about 9.48Gt, and 77% of the total emission amount is from coal-fired thermal power units and other coal-fired processes. Because of the energy structure in China, the coal-fired thermal power generating unit is the main body for energy production in the current and future time. Thereby, CO is carried out on the coal-fired thermal power unit 2 Trapping is of great importance.
Currently, post-combustion CO is chemically absorbed based on solvents such as ethanolamine 2 The trapping is one of the most mature and commercial popularization value technologies in all carbon trapping technologies, and is successfully applied to more than ten tested devices in China. However, coal-fired thermal power generation unit and post-combustion CO 2 The trapping system has a strong coupling effect. On one hand, with the great popularization of new energy technology, the coal-fired thermal power unit needs to carry out deep peak regulation so as to maintain the supply and demand level of the power gridThe balance, the peak regulation process of the thermal power generating unit can lead to corresponding change of flue gas parameters, which can influence the operation of a downstream carbon capture system; on the other hand, post-combustion CO 2 The solvent regeneration of the trapping system consumes a large amount of heat, and the steam extraction of the steam turbine of the thermal power unit provides a heat source, so that the generating capacity of the thermal power unit is reduced, and the thermal efficiency of the unit is affected. To sum up, analysis is carried out on coal-fired thermal power unit and post-combustion CO 2 The operation modes of the trapping systems are mutually influenced and mutually collided. In order to maximize the benefit of the whole system, the operation requirement of the whole system needs to be comprehensively considered, and the CO of the coal-fired power plant is realized on the premise of ensuring the constraint condition of the system 2 And capturing the optimized schedule of the whole system.
According to the current research situation at home and abroad, CO of coal-fired power stations 2 The research on optimizing and scheduling the trapping system is relatively less, and the existing method does not consider CO of the coal-fired power plant 2 The complex nature of the trapping system makes it difficult to achieve optimal operation of the overall system. The invention utilizes artificial intelligence to build the CO of the coal-fired power station 2 The optimization scheduling system of the trapping system solves the defects of multiple system operation variables, complex process characteristics and large optimization scheduling calculation amount, can solve the optimal given value of the main controlled variable of the system in the scheduling period on line in real time, and is suitable for industrial application.
Disclosure of Invention
The invention provides a large-scale coal-fired power plant CO based on artificial intelligence 2 The trapping system optimizing and scheduling method aims at solving the problems.
The invention is realized in such a way that the large-scale coal-fired power station CO based on artificial intelligence 2 The trapping system optimizing and scheduling method comprises the following steps:
step one, selecting main steam pressure, intermediate point enthalpy value, unit generating capacity, trapping rate, reboiler temperature and CO 2 The yield is CO of large-scale coal-fired power plant 2 Controlled variable y of trapping system (k) Selecting the coal feeding amount, the water feeding flow, the opening of a main steam valve, the lean solution flow and the steam extraction flow of a reboiler as corresponding control variables u (k) ；
Step two, under the condition of open loop, simultaneously changing the coal feeding amount, the water feeding flow, the opening of a main steam valve, the lean solution flow and the steam extraction flow of a reboiler to obtain steady-state input and output data of the system under different generating capacity and capturing rate operation working conditions;
step three, carrying out normalization processing on input and output data, and constructing a coal-fired power station CO containing three-layer limited Boltzmann machines (Restricted Boltzmann Machines, RBM) by utilizing pretrainDBN and trainDBN functions in MATLAB 2 The trapping system DBN steady state model,
wherein, the CO of the coal-fired power plant 2 The trapping system DBN steady-state model function is as follows: y (k) =f DBN (u(k))，y (k) For the acquisition value of the controlled variable at the current moment, u (k) For controlling the acquisition value of the variable at the current moment, f DBN (. Cndot.) is a functional form of the DBN model;
step four, considering the generating capacity of the unit and CO 2 Emission, running cost and system constraint factors, and construction can reflect CO of coal-fired power plants 2 Capturing an objective function of the running cost of the system and a weight coefficient corresponding to the objective function,
wherein, the CO of the coal-fired power plant 2 The objective function of the operation cost of the trapping system is that
CO of the coal-fired power plant 2 The objective function of the trapping system running cost satisfies a series of constraints;
step five, setting an optimized time domain N i And a scheduling period T s ；
Step six, setting relevant parameters of a BO solver, including acquisition functions, the number of initial evaluation points and cycle period parameters;
step seven, solving an objective function by using a BO solver, and solving an optimal control variable u which meets the minimum of the objective function under the constraint condition of the system (k) ；
Step eight, utilizing CO of coal-fired power plant 2 Trapping system DBN steady-state model, solving input as control variable u (k) Is set to the optimum output variable given value y ref(k) ；
Step nine, outputting the optimal given value y ref(k) Realizing CO of coal-fired power plant 2 The optimized schedule of the trapping system is followed by repeating steps seven through nine in each optimized time domain.
Preferably, in the first step: with fuel quantity command (u) 1 ) Flow rate of feed water (u) 2 ) Main steam valve opening (u) 3 ) Lean solution flow rate (u) 4 ) And the flow rate of the extracted steam (u) 5 ) As the main control variable u of the system (k) At the main vapor pressure (y 1 ) Enthalpy value of middle point (y) 2 ) Turbine power (y) 3 ) Collection ratio (y) 4 ) Reboiler temperature (y) 5 ) And CO 2 Yield (y) 6 ) As the main controlled variable y of the system (k) Will control the variable u (k) As input, the controlled variable y (k) As an output.
Preferably, in the second step: the output data can cover the main operation interval of 40% -95% trapping rate and 300MW-660MW generating capacity of the unit.
Preferably, in the third step: the number of units of the three-layer limited boltzmann machine is 100, 100 and 50 respectively.
Preferably, in the fourth step: CO of the coal-fired power plant 2 The objective function of the trapping system running cost satisfies a series of constraints:
y j ＝f DBN (u j ),j＝1,2,…,5 (2)
u min ≤u j ≤u max ,j＝1,2,…,5 (3)
y 1,i ≤y 1，max ,i＝1,2,…,N i (4)
y 4,min ≤y 4,i ≤y 4,max ,i＝1,2,…,N i (5)
y 5,min ≤y 5,i ≤y 5,max ,i＝1,2,...,N i (6)
wherein N is i Representing an optimized time domain; alpha 1,i To alpha 4,i Respectively corresponding weight coefficients of the performance indexes; u (u) min And u max Respectively a lower limit constraint and an upper limit constraint of the amplitude of the input variable; y is 1，max Is the main steam pressure (y 1 ) Upper limit constraints of (2); y is 4,min And y 4,max Respectively the collecting rate (y) 4 ) Lower and upper limit constraints of (2); y is 5,min And y 5,max Respectively reboiler temperatures (y 5 ) Lower and upper limit constraints of (2); y is 6,m i n And y 6,max CO respectively 2 Yield (y) 6 ) Lower and upper constraints on the total amount in the optimization time domain;
J 1,i to J 4,i CO of large-scale coal-fired power plant 2 The specific expression of the objective function related to the unit operation safety and the operation economy in the trapping system is as follows:
J 1,i ＝|y 3,i -E uld | (8)
J 2,i ＝u 1,i (9)
J 3,i ＝(1-y 4,i ) (10)
J 4,i ＝y 4,i (11)
wherein J is 1,i Represents the load tracking error of the thermal power unit, J 2,i Representing the fuel quantity consumed in the running process of the unit, J 3,i Indicating CO emissions 2 Punishment of J 4,i Representing costs of operation and maintenance of PCC systems, E uld For AGC load commands.
Preferably, in the fourth step: the weight coefficients corresponding to the objective function in the formula (1) are respectively:
α 1,i ＝0.3C grid,i (12)
α 2,i ＝3.6C fuel,i (13)
α 4,i ＝0.215q g,i C O&M,i (15)
wherein C is grid,i The current online electricity price (USD/MWh) is represented; c (C) fuel，i Indicating the current time fuel quantity price (USD/ton). Meanwhile, the station service electricity and the throttling heat loss of the thermal power generating unit can be converted into the fuel consumption; q g,i The mass flow of flue gas (kg/s) is indicated,for the current moment CO 2 Price of emissions (USD/ton); c (C) O&M,i For individual carbon capture system operation and maintenance costs (USD/ton).
Preferably, in step seven: the BO solver solves the objective function by using a BO algorithm, wherein the BO algorithm is realized by a bayesopt function in MATLAB
Preferably, large-scale coal-fired power plant CO 2 The trapping system comprises five main units of a boiler, a steam turbine, a generator, an absorption tower and a separation tower; the main variables include: fuel quantity command (u) 1 ) Flow rate of feed water (u) 2 ) Main steam valve opening (u) 3 ) Lean solution flow rate (u) 4 ) Flow rate of extraction (u) 5 ) Main vapor pressure (y) 1 ) Enthalpy value of middle point (y) 2 ) Turbine power (y) 3 ) Collection ratio (y) 4 ) Reboiler temperature (y) 5 ) And CO 2 Yield (y) 6 )。
Compared with the prior art, the invention has the beneficial effects that: the invention relates to a large-scale coal-fired power plant CO based on artificial intelligence 2 Method for optimizing and scheduling trapping system and establishing CO (carbon monoxide) of coal-fired power station by using DBN (direct base station) 2 Capturing a steady-state data model of a system, and considering generating capacity and CO of a unit 2 Emission and CO of coal-fired power plant 2 The system economy index is established according to the operation cost of the trapping system, and according to the information such as the real-time carbon price, the electricity price and the like, the optimal control variable given value meeting the system constraint is solved by utilizing the BO algorithm, so that the CO of the coal-fired power plant is realized 2 The invention can give consideration to the optimization scheduling of the trapping system and the CO of the coal-fired power station 2 The operation requirements of the trapping system in the aspects of economy, safety and environmental protection find out the optimal scheduling mode meeting the minimum total operation cost, and realize the CO of the large-scale coal-fired power plant 2 The trapping system operates optimally.
Drawings
FIG. 1 is a schematic illustration of an artificial intelligence based coal-fired power plant CO of the present invention 2 And a block diagram of an optimized scheduling method of the trapping system.
FIG. 2 is a schematic diagram of a large coal-fired power plant CO 2 A trapping system flow diagram.
Fig. 3 is a graph showing the trend of load command and internet power price.
FIG. 4 is a schematic illustration of a coal-fired power plant CO 2 Time-by-time cost comparison of the trapping system before and after optimization.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention provides a technical scheme that: coal-fired power plant CO based on artificial intelligence 2 The trapping system optimization scheduling method is implemented by a system as shown in fig. 1, and the system comprises: BO solver and coal-fired power plant CO 2 Capturing system objective function and coal-fired power plant CO 2 Capturing a DBN steady-state model of the system;
the BO solver utilizes a Bayesian algorithm to quickly solve an objective function;
coal-fired power plant CO 2 The capture system objective function considers electricity price, carbon price, system operation cost and input and output variable constraint, and can describe the CO of the coal-fired power station 2 The total cost of operation of the trapping system;
coal-fired power plant CO 2 DBN steady-state model of trapping system and CO of coal-fired power plant 2 The target function module of the trapping system inputs the control variable u (k) And outputting the controlled variable y (k) And gives the optimal output variable givenValue y ref (k)。
As shown in FIG. 2, large-scale coal-fired power plant CO 2 A trapping system comprising: main units such as a boiler, a steam turbine, a generator, an absorption tower, a separation tower and the like; the main variables include: fuel quantity command (u) 1 ) Flow rate of feed water (u) 2 ) Main steam valve opening (u) 3 ) Lean solution flow rate (u) 4 ) Flow rate of extraction (u) 5 ) Main vapor pressure (y) 1 ) Enthalpy value of middle point (y) 2 ) Turbine power (y) 3 ) Collection ratio (y) 4 ) Reboiler temperature (y) 5 ) And CO 2 Yield (y) 6 )。
Coal-fired power plant CO based on artificial intelligence 2 The trapping system optimizing and scheduling method comprises the following steps:
step one, selecting main steam pressure, intermediate point enthalpy value, unit generating capacity, trapping rate, reboiler temperature and CO 2 The yield is CO of large-scale coal-fired power plant 2 Controlled variable y of trapping system (k) Selecting the coal feeding amount, the water feeding flow, the opening of a main steam valve, the lean solution flow and the steam extraction flow of a reboiler as corresponding control variables u (k) ；
Step two, setting a sampling period T s =30 seconds. Under the open loop condition, simultaneously changing the coal feeding amount, the water feeding flow, the opening of a main steam valve, the lean solution flow and the steam extraction flow of a reboiler, obtaining steady-state input and output data of the system under the operation working conditions of different generated energy, trapping rate and the like, wherein the output data can cover main operation intervals of 40% -95% trapping rate, 300MW-660MW unit generated energy and the like;
step three, carrying out normalization pretreatment on input and output data, and constructing a coal-fired power station CO containing three layers of Bernoulli-Bernoulli limited Boltzmann machines by utilizing pretrainDBN and trainDBN functions in MATLAB 2 Trapping system DBN steady state model. The number of units of each layer of limited Boltzmann machine is 100, 100 and 50 respectively, the learning rate is 0.5, and the DBN model is as follows: y (k) =f DBN (u(k))；
Step four, considering the generating capacity of the unit and CO 2 Emissions, operating costs, and system constraintsElement for constructing and reflecting CO of coal-fired power plant 2 Objective function of the trapping system running cost:
and satisfies the following constraints:
y j ＝f DBN (u j ),j＝1,2,…,5 (2)
u min ≤u j ≤u max ,j＝1,2,…,5 (3)
y 1,i ≤y 1，max ,i＝1,2,...,N i (4)
y 4,min ≤y 4,i ≤y 4,max ,i＝1,2,...,N i (5)
y 5,min ≤y 5,i ≤y 5,max ,i＝1,2,...,N i (6)
wherein, the constraint conditions are respectively set as follows: u (u) min ＝[20；200；0.4；100；30]T；u max ＝[80；600；1；600；250] T ；y 1,max ＝26；y 4,min ＝50％；y 4,max ＝90％；y 5,min ＝383；y 5,max ＝393；y 6,min ＝1000；y 6,max ＝1200；
J in the above formula (1) 1,i To J 4,i CO of large-scale coal-fired power plant 2 The specific expression of the objective function related to the unit operation safety and the operation economy in the trapping system is as follows:
J 1,i ＝|y 3,i -E uld | (8)
J 2,i ＝u 1,i (9)
J 3,i ＝(1-y 4,i ) (10)
J 4,i ＝y 4,i (11)
wherein J is 1,i Represents the load tracking error of the thermal power unit, J 2,i Representing the fuel quantity consumed in the running process of the unit, J 3,i Indicating CO emissions 2 Punishment of J 4,i Representing costs of operation and maintenance of PCC systems, E uld For AGC load command, AGC load command trend is shown in fig. 3.
Constructing weight coefficients corresponding to the objective function in the formula (1), wherein the weight coefficients are respectively as follows:
α 1,i ＝0.3C grid,i (12)
α 2,i ＝3.6C fuel,i (13)
α 4,i ＝0.215q g,i C O&M,i (15)
in the internet electricity price C grid,i The (USD/MWh) trend is shown in FIG. 3; fuel quantity price C at present fuel，i 91.4USD/ton; mass flow q of flue gas g,i ＝3.8756u 1 +264.2507 (kg/s); CO at the present time 2 Discharge price50USD/ton; individual carbon capture system operation and maintenance costs C O&M,i 4.862USD/ton;
step five, setting an optimized time domain N i ＝12；
Step six, setting relevant parameters of a BO solver, wherein an acquisition function is 'probability-of-improvement', the number of initial evaluation points is 4, and a cycle period is 100;
step seven, solving an objective function by using a BO solver, and solving an optimal control variable u which meets the minimum of the objective function under the constraint condition of the system (k) ；
Step eight, utilizing CO of coal-fired power plant 2 Trapping system DBN steady-state modeType, solve for input u (k) Is set to the optimum output variable given value y ref(k) ；
Step nine, outputting the optimal given value y ref(k) Realizing CO of coal-fired power plant 2 The optimized schedule of the trapping system is followed by repeating steps seven through nine in each optimized time domain.
CO 2 Discharge price C CO2,i Respectively into 10USD/ton, 50USD/ton and 150USD/ton, and CO of coal-fired power plant 2 The time-by-time cost of the trapping system before and after the optimization scheduling is as shown in fig. 4 (the upper half line distributed from bottom to top corresponds to the optimized carbon prices of 10USD/ton, 50USD/ton, 150USD/ton, and the lower half line distributed from bottom to top corresponds to the carbon prices of 10USD/ton, 50USD/ton, and 150USD/ton, respectively), and it is obvious that the time-by-time cost after the system optimization is greatly reduced.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (6)
1. Large-scale coal-fired power plant CO based on artificial intelligence 2 The trapping system optimizing and scheduling method is characterized in that: the method comprises the following steps:
step one, selecting a main vapor pressure (y 1 ) Enthalpy value of middle point (y) 2 ) Turbine power (y) 3 ) Collection ratio (y) 4 ) Reboiler temperature (y) 5 ) And CO 2 Yield (y) 6 ) CO for large-scale coal-fired power plant 2 Controlled variable y of trapping system (k) Selecting a fuel quantity command (u 1 ) Flow rate of feed water (u) 2 ) Main steam valve opening (u) 3 ) Lean solution flow rate (u) 4 ) Flow rate of extraction (u) 5 ) For the corresponding control variable u (k) ；
Step two, setting a sampling period T s =30 seconds, in the open loop case, the fuel quantity command (u 1 ) The water supply flow, the opening of a main steam valve, the lean solution flow and the steam extraction flow of a reboiler,acquiring steady-state input and output data of the system under the operating conditions of different generating capacity and trapping rate;
step three, carrying out normalization processing on input and output data, and constructing a coal-fired power station CO containing three-layer limited Boltzmann machines (Restricted Boltzmann Machines, RBM) by utilizing pretrainDBN and trainDBN functions in MATLAB 2 The trapping system DBN steady state model,
wherein, the CO of the coal-fired power plant 2 The trapping system DBN steady-state model function is as follows: y (k) =f DBN (u(k))，f DBN (. Cndot.) is a functional form of the DBN model;
step four, considering the generating capacity of the unit and CO 2 Emission, running cost and system constraint factors, and construction can reflect CO of coal-fired power plants 2 Capturing an objective function of the running cost of the system and a weight coefficient corresponding to the objective function,
wherein, the CO of the coal-fired power plant 2 The objective function of the operation cost of the trapping system is that
CO of the coal-fired power plant 2 The objective function of the trapping system running cost satisfies a series of constraints;
step five, setting an optimized time domain N i ＝12；
Step six, setting relevant parameters of a BO solver, including acquisition functions, the number of initial evaluation points and cycle period parameters;
step seven, solving an objective function by using a BO solver, and solving an optimal control variable u which meets the minimum of the objective function under the constraint condition of the system (k) ；
Step eight, utilizing CO of coal-fired power plant 2 Trapping system DBN steady-state model, solving input as control variable u (k) Is set to the optimum output variable given value y ref (k)；
Step nine, outputting the optimal given value y ref (k) Realizing CO of coal-fired power plant 2 Optimized scheduling of trapping systems, after which the steps are repeated for each optimized time domainSeventh to ninth steps;
in the fourth step: CO of the coal-fired power plant 2 The objective function of the trapping system running cost satisfies a series of constraints:
y j ＝f DBN (u j ),j＝1,2,...,5 (2)
u min ≤u j ≤u max ,j＝1,2,…,5 (3)
y 1,i ≤y 1，max ,i＝1,2,…,N i (4)
y 4,min ≤y 4,i ≤y 4,max ,i＝1,2,…,N i (5)
y 5,min ≤y 5,i ≤y 5,max ,i＝1,2,…,N i (6)
wherein y is 1 Is the main steam pressure, y 2 For the enthalpy value of the middle point, y 3 For turbine power, y 4 Is the trapping rate (y) 4 )，y 5 Is the reboiler temperature; u (u) 1 To select a fuel quantity command; u (u) 2 For the water supply flow rate (u) 2 )，u 3 Is the opening degree (u) of the main steam valve 3 )，u 4 Is the lean liquid flow rate (u) 4 )，u 5 For the flow rate of extraction (u) 5 )；
N i Representing an optimized time domain; alpha 1,i To alpha 4,i Respectively corresponding weight coefficients of the performance indexes; u (u) min And u max Respectively a lower limit constraint and an upper limit constraint of the amplitude of the input variable; y is 1，max Is the main steam pressure (y 1 ) Upper limit constraints of (2); y is 4,m i n And y 4,max Respectively the collecting rate (y) 4 ) Lower and upper limit constraints of (2); y is 5,m i n And y 5,max Respectively reboiler temperatures (y 5 ) Lower and upper limit constraints of (2); y is 6,min And y 6,max CO respectively 2 Yield (y) 6 ) Lower and upper constraints on the total amount in the optimization time domain;
J 1,i to J 4,i CO of large-scale coal-fired power plant 2 The specific expression of the objective function related to the unit operation safety and the operation economy in the trapping system is as follows:
J 1,i ＝|y 3,i -E uld | (8)
J 2,i ＝u 1,i (9)
J 3,i ＝(1-y 4,i ) (10)
J 4,i ＝y 4,i (11)
wherein J is 1,i Represents the load tracking error of the thermal power unit, J 2,i Representing the fuel quantity consumed in the running process of the unit, J 3,i Indicating CO emissions 2 Punishment of J 4,i Representing costs of operation and maintenance of PCC systems, E uld For AGC load commands.
2. An artificial intelligence based large coal burning power plant CO as claimed in claim 1 2 The trapping system optimizing and scheduling method is characterized in that: in the first step: with fuel quantity command (u) 1 ) Flow rate of feed water (u) 2 ) Main steam valve opening (u) 3 ) Lean solution flow rate (u) 4 ) And the flow rate of the extracted steam (u) 5 ) As the main control variable u of the system (k) At the main vapor pressure (y 1 ) Enthalpy value of middle point (y) 2 ) Turbine power (y) 3 ) Collection ratio (y) 4 ) Reboiler temperature (y) 5 ) And CO 2 Yield (y) 6 ) As the main controlled variable y of the system (k) Will control the variable u (k) As input, the controlled variable y (k) As an output.
3. An artificial intelligence based large coal burning power plant CO as claimed in claim 1 2 The trapping system optimizing and scheduling method is characterized in that: in the second step: the output data can cover the main operation interval of 40% -95% trapping rate and 300MW-660MW generating capacity of the unit.
4. An artificial intelligence based large coal burning power plant CO as claimed in claim 1 2 The trapping system optimizing and scheduling method is characterized in that: in the third step: the number of units of the three-layer limited boltzmann machine is 100, 100 and 50 respectively.
5. An artificial intelligence based large scale coal burning power plant CO as claimed in claim 4 2 The trapping system optimizing and scheduling method is characterized in that: in the fourth step: the weight coefficients corresponding to the objective function in the formula (1) are respectively:
α 1,i ＝0.3C grid,i (12)
α 2,i ＝3.6C fuel,i (13)
α 4,i ＝0.215q g,i C O&M,i (15)
wherein C is grid,i The current online electricity price (USD/MWh) is represented; c (C) fuel，i Indicating a current time fuel quantity price (USD/ton); q g,i The mass flow of flue gas (kg/s) is indicated,for the current moment CO 2 Price of emissions (USD/ton); c (C) O&M,i For individual carbon capture system operation and maintenance costs (USD/ton).
6. An artificial intelligence based large coal burning power plant CO as claimed in claim 1 2 The trapping system optimizing and scheduling method is characterized in that: step seven,: the BO solver solves the objective function using a BO algorithm, which is implemented by a bayesopt function in MATLAB.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110646133.6A CN113341716B (en) | 2021-06-10 | 2021-06-10 | Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110646133.6A CN113341716B (en) | 2021-06-10 | 2021-06-10 | Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113341716A CN113341716A (en) | 2021-09-03 |
| CN113341716B true CN113341716B (en) | 2023-08-04 |
Family
ID=77476167
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110646133.6A Active CN113341716B (en) | 2021-06-10 | 2021-06-10 | Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113341716B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011150462A1 (en) * | 2010-06-02 | 2011-12-08 | Co2Crc Technologies Pty Ltd | Optimising objectives of a plant and a carbon dioxide capture facility |
| CN110764419A (en) * | 2019-11-15 | 2020-02-07 | 江苏方天电力技术有限公司 | CO of large coal-fired power station2Capture global scheduling and predictive control system and method |
| CN112633560A (en) * | 2020-12-11 | 2021-04-09 | 西安交通大学 | Power station optimal scheduling method containing coal-fired cogeneration unit |
-
2021
- 2021-06-10 CN CN202110646133.6A patent/CN113341716B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011150462A1 (en) * | 2010-06-02 | 2011-12-08 | Co2Crc Technologies Pty Ltd | Optimising objectives of a plant and a carbon dioxide capture facility |
| CN110764419A (en) * | 2019-11-15 | 2020-02-07 | 江苏方天电力技术有限公司 | CO of large coal-fired power station2Capture global scheduling and predictive control system and method |
| CN112633560A (en) * | 2020-12-11 | 2021-04-09 | 西安交通大学 | Power station optimal scheduling method containing coal-fired cogeneration unit |
Non-Patent Citations (1)
| Title |
|---|
| 燃煤电站锅炉二氧化碳捕集封存技术经济性分析;田牧等;《锅炉技术》;20090531;第40卷(第03期);第36-40页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113341716A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107482688B (en) | Scheduling optimization method for carbon capture virtual power plant | |
| CN111552175B (en) | Overall optimization scheduling and rapid variable load control method for supercritical coal-fired power plant-carbon capture system after chemical adsorption combustion | |
| Chen et al. | The mutual benefits of renewables and carbon capture: Achieved by an artificial intelligent scheduling strategy | |
| CN110764419B (en) | CO of large coal-fired power plant2Capture global scheduling and predictive control system and method | |
| Wang et al. | Multi-energy coupling analysis and optimal scheduling of regional integrated energy system | |
| Li et al. | Energy, exergy and economic analyses of new coal-fired cogeneration hybrid plant with wind energy resource | |
| CN110932317A (en) | Design method of distributed energy system with complementary essential renewable energy sources | |
| CN102502631A (en) | Control method of post-combustion carbon capture generating station based on different flexible operation modes | |
| CN115238987A (en) | Energy efficiency improvement scheduling method considering multi-type low-carbon factors and demand response | |
| Fan et al. | Exergy analysis and dynamic control of chemical looping combustion for power generation system | |
| AU2021105749A4 (en) | Method for designing and modeling integrated energy system for realizing carbon cycle | |
| Huo et al. | Thermodynamic modeling and control of hybrid solar-fossil fuel power generation and storage system | |
| CN201962264U (en) | System for producing natural gas by wind, power and coal | |
| CN114282828A (en) | Carbon circulation system and application thereof | |
| CN113341716B (en) | Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system | |
| CN108508748A (en) | CO after a kind of effective photo-thermal auxiliary combustion2Trapping system progress control method | |
| Yueming et al. | Development status of advanced thermal power technology and low-carbon path of China's thermal power industry | |
| CN114519456B (en) | Green agriculture zero-carbon energy supply system and intelligent configuration hierarchical optimization algorithm thereof | |
| CN115660142A (en) | Source-load-storage coordination optimization scheduling method for park comprehensive energy system | |
| CN114091917A (en) | Dynamic environment-friendly economic dispatching method and system for cold-heat-power combined supply type micro-grid | |
| CN114000979A (en) | Electric water-hydrogen-methanol multi-generation energy source island and method | |
| Li et al. | Optimal Scheduling Method of Cogeneration System with Heat Storage Device Based on Memetic Algorithm | |
| FU et al. | A Integrated System of The Coal-Fired Unit With CO2 Capture to Produce Methanol | |
| Chen et al. | Semi-batch chemical-looping reactors integrated with combined cycle power plants operating at transient electricity demand | |
| Han et al. | Optimal Scheduling of Integrated Energy Systems with P2G Two-Stage and Carbon Capture Systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |