Apparatus for predicting combustion state of power plant boiler and method thereof

Abstract

The present invention discloses a combustion state prediction device for a power plant boiler and a method thereof. The combustion state prediction device for a power plant boiler according to the present invention comprises: an operating variable input module for receiving operating variables for operating the boiler; an operating information database for storing operating information of the boiler; an output module for displaying prediction data; and a processor operatively coupled to the operating variable input module, the operating information database, and the output module; wherein the processor drives an execution program to predict the combustion state of the boiler based on a prediction model based on the operating variables input from the operating variable input module and the boiler operating information stored in the operating information database, and outputs the prediction data through the output module.

Description

Technology Field

[0001] The present invention relates to a combustion state prediction device and method for a power plant boiler, and more specifically, to a combustion state prediction device and method for a power plant boiler that enables real-time prediction and monitoring of the combustion state inside a boiler in a coal-fired power plant through a prediction model based on a physical model. Background Technology

[0002] Generally, in the case of boilers in coal-fired power plants, the exothermic reaction generated during coal combustion is used to heat water and produce steam necessary for power generation, and electricity is produced by using this generated steam to rotate a turbine.

[0003] Combustion is a phenomenon in which a substance burns accompanied by a large amount of heat and light using oxygen in the air as a medium. The combustion process consists of a chain of chemical reactions and is classified as a relatively complex physical phenomenon, often involving the generation of turbulent flow.

[0004] In the case of large-scale external combustion engines such as coal-fired power plants, combustion takes place inside the boiler, and it is an environment where it is very difficult to directly monitor the internal combustion conditions, as the temperature of the combustion gases approaches 1,500 to 2,000 degrees Celsius and ash and other combustion byproducts flow at high speed along with the gases.

[0005] However, if problems such as incomplete combustion or flame imbalance occur over a wide combustion area, the ripple effect of the combustion conditions on the entire power plant is significant, as this leads to a decrease in the overall thermal efficiency and reliability of the plant.

[0006] Despite the importance of stable combustion, it is currently impossible to monitor the internal combustion status of the boiler in real time. Consequently, the best we can do is estimate the combustion status using partial data measured at the boiler's heat transfer section or furnace outlet, or perform only localized monitoring by installing special cameras on parts of the burner.

[0007] The background technology of the present invention is disclosed in Korean Registered Patent Publication No. 10-2363444 (published on February 16, 2022, apparatus and method for selecting an optimal boiler combustion model). The problem to be solved

[0008] To overcome such problems, facilities utilizing combustion, such as thermal power plants, employ methods like computational thermal fluid analysis to simulate physical phenomena occurring within the equipment, thereby indirectly predicting internal combustion phenomena and utilizing them for diagnosis.

[0009] However, computational thermal fluid analysis requires exponentially more computational resources and expertise as the structure and scale of the facility become more complex and the number of combustion chemical reactions increases, and the analysis time can also reach several days. Consequently, it is often utilized only during the design phase, and there is a problem in that it is difficult to use for monitoring the current status of the facility during the operation and management phase.

[0010] The present invention has been devised to improve upon the aforementioned problems. According to one aspect, the objective of the present invention is to provide a combustion state prediction device and method for a power plant boiler that can predict and monitor the combustion state inside the boiler in real time by receiving power plant operation data and operation variables through a prediction model based on a physical model. means of solving the problem

[0011] A combustion state prediction device for a power plant boiler according to one aspect of the present invention comprises: an operating variable input module for receiving operating variables for operating the boiler; an operating information database for storing operating information of the boiler; an output module for displaying prediction data; and a processor operatively coupled to the operating variable input module, the operating information database, and the output module, wherein the processor drives an execution program to predict the combustion state of the boiler based on a prediction model based on the operating variables input from the operating variable input module and the boiler operating information stored in the operating information database, and outputs the prediction data through the output module.

[0012] In the present invention, the operating variables are characterized by including one or more of the following: the amount of coal input, the opening rate of the primary air and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is on / off for each layer, and the calorific value of the coal being input.

[0013] In the present invention, the operating information is characterized by including operating data acquired from a measuring device installed in the boiler and control data used for controlling the boiler.

[0014] In the present invention, the prediction model is characterized as being a model generated through machine learning based on a physical model for predicting combustion phenomena in a boiler.

[0015] In the present invention, the prediction data is characterized by including one or more of the following: spatial coordinates inside the boiler, physical characteristic values ​​corresponding to the position of the spatial coordinates, air flow rate of the secondary air nozzle, a contour image of the physical quantity distribution over the entire furnace space of the boiler, and physical characteristic values ​​measured by a measuring device.

[0016] The present invention further includes a simulation condition input module for receiving simulated operating conditions of a boiler; and the processor predicts the combustion state of the boiler based on a prediction model based on the simulated operating conditions input through the simulation condition input module and the boiler operating information stored in the operating information database.

[0017] The present invention further comprises an optimization condition input module that receives optimization conditions for a boiler; and the processor derives optimal operating variables capable of satisfying the optimization conditions input through the optimization condition input module based on an optimization model, and predicts the combustion state of the boiler based on a prediction model based on the optimal operating variables and the boiler operating information stored in the operating information database.

[0018] A method for predicting the combustion state of a power plant boiler according to another aspect of the present invention comprises: a step in which a processor drives an execution program to receive operating variables for operating the boiler through an operating variable input module and receives operating information of the boiler from a measuring device and stores it in an operating information database; a step in which the processor predicts the combustion state of the boiler based on a prediction model based on the operating variables and operating information; and a step in which the processor outputs the predicted data of the combustion state through an output module.

[0019] In the present invention, the operating variables are characterized by including one or more of the following: the amount of coal input, the opening rate of the primary air and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is on / off for each layer, and the calorific value of the coal being input.

[0020] In the present invention, the operating information is characterized by including operating data acquired from a measuring device installed in the boiler and control data used for controlling the boiler.

[0021] In the present invention, the prediction model is characterized as being a model generated through machine learning based on a physical model for predicting combustion phenomena in a boiler.

[0022] In the present invention, the prediction data is characterized by including one or more of the following: spatial coordinates inside the boiler, physical characteristic values ​​corresponding to the position of the spatial coordinates, air flow rate of the secondary air nozzle, a contour image of the physical quantity distribution over the entire furnace space of the boiler, and physical characteristic values ​​measured by a measuring device.

[0023] The step of predicting the combustion state of a boiler in the present invention further comprises: a step in which a processor receives simulated operating conditions input through a simulated condition input module; and a step in which the processor predicts the combustion state of a boiler based on a prediction model based on the simulated operating conditions and the boiler operating information stored in an operating information database.

[0024] The step of predicting the combustion state of a boiler in the present invention further comprises: a step in which a processor receives an optimization condition of the boiler from an optimization condition input module; a step in which the processor derives an optimal operating variable capable of satisfying the optimization condition based on an optimization model; and a step in which the processor predicts the combustion state of the boiler based on a prediction model based on the optimal operating variable and the boiler operating information stored in an operating information database. Effects of the invention

[0025] A device and method for predicting the combustion state of a power plant boiler according to one aspect of the present invention can predict and monitor the combustion state inside a boiler in a coal-fired power plant in real time by receiving power plant operation data and operation variables through a prediction model based on a physical model, thereby allowing the causes to be analyzed by checking the internal temperature distribution and flow velocity of the boiler, and providing solutions to prevent high temperatures or increases in flow velocity in specific areas.

[0026] In addition, according to the present invention, through combustion tuning, various operating conditions for stabilizing boiler combustion in various situations, such as ammonia co-firing, woody fuels, and low-caloric coal, can be reviewed and simulated, thereby enabling rapid response to environmental changes, such as the introduction of new fuels and flexible operation, while maintaining a stable power supply.

[0027] In addition, according to the present invention, by verifying boiler efficiency under various combustion conditions and analyzing the causes that reduce efficiency, and by optimizing the secondary air distribution amount under various loads, the best efficiency can be achieved under any conditions, thereby contributing to economic efficiency and environmental friendliness.

[0028] In addition, according to the present invention, by tracking incomplete combustion within the combustion zone and identifying the zone where pollutants such as NOx are generated, it is possible to find operating conditions that minimize the finally emitted pollutants through detailed combustion tuning. Brief explanation of the drawing

[0029] FIG. 1 is a block diagram showing a combustion state prediction device for a power plant boiler according to one embodiment of the present invention. FIG. 2 is an example diagram showing the result of querying the combustion state in a combustion state prediction device for a power plant boiler according to one embodiment of the present invention. FIG. 3 is an example diagram showing the result of querying the state of the secondary air inlet in a combustion state prediction device for a power plant boiler according to one embodiment of the present invention. FIG. 4 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to one embodiment of the present invention. FIG. 5 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to another embodiment of the present invention. FIG. 6 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to another embodiment of the present invention. Specific details for implementing the invention

[0030] Hereinafter, an apparatus and method for predicting the combustion state of a power plant boiler according to the present invention will be described with reference to the attached drawings. In this process, the thickness of lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of explanation. Furthermore, the terms described below are defined considering their functions in the present invention, and these may vary depending on the intention or convention of the user or operator. Therefore, the definitions of these terms should be based on the content throughout this specification.

[0031] FIG. 1 is a block diagram showing a combustion state prediction device for a power plant boiler according to one embodiment of the present invention, FIG. 2 is an example diagram showing the result of querying the combustion state in the combustion state prediction device for a power plant boiler according to one embodiment of the present invention, and FIG. 3 is an example diagram showing the result of querying the secondary air inlet state in the combustion state prediction device for a power plant boiler according to one embodiment of the present invention.

[0032] As illustrated in FIG. 1, a combustion state prediction device for a power plant boiler according to one embodiment of the present invention may include an operating variable input module (10), an operating information database (60), an output module (70), a memory (40), and a processor (50), as well as a simulation condition input module (20) and an optimization condition input module (30).

[0033] The operating variable input module (10) can receive operating variables to operate by manipulating the combustion environment of the actual boiler.

[0034] In the case of a coal-fired boiler, there may be many operating variables, such as the amount of coal fed, the opening rate of the primary and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is turned on or off for each layer, and the calorific value of the fed coal.

[0035] The operating variables entered in this way are reflected in the operation of the boiler and can be stored in the operating information database (60).

[0036] The simulation condition input module (20) can receive simulation operating conditions of the boiler.

[0037] In other words, when there is a problem with the current actual operating conditions of the boiler or when seeking improved conditions, new virtual operating conditions can be entered as simulated operating conditions.

[0038] The optimization condition input module (30) can receive the optimization conditions of the boiler.

[0039] In other words, optimization conditions for optimizing operating variables can be set, for example, the temperature range of the boiler furnace outlet and the heat transfer rate range of each heat transfer unit, and operating variable conditions can be set to minimize the NOx concentration at the furnace outlet.

[0040] In addition, the allowable range of variable values ​​to be maintained among the major operating variables can be set together. For example, the output can be set to 500 MW ±5%.

[0041] The operation information database (60) can store operation information of the boiler.

[0042] Here, it may include operation data acquired in real time from a measuring device installed in the boiler and control data used for controlling the boiler. At this time, the operation information database (60) may include a data communication function, that is, a function to provide data to another program and record data provided by another program.

[0043] The output module (70) can display prediction data. That is, the output module (70) can refine the prediction data and convert it into an information form such as an image or graph that the user can intuitively accept, and display it, and in addition, it can also display key operating information such as the values ​​of each operating variable of the boiler and the output of the boiler.

[0044] The memory (40) can store an execution program and a learning model related to the operation of the combustion state prediction device of the power plant boiler, and the stored information can be selected by the processor (50) as needed.

[0045] That is, various types of data and commands generated during the execution of an operating system (O / S) or application (program or applet) for driving the combustion state prediction device of a power plant boiler are stored in the memory (40). At this time, the memory (40) can be implemented as non-volatile memory, volatile memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD). In addition, the memory (40) is accessed, and data reading / writing / modification / deletion / updating by the processor (50) can be performed.

[0046] The memory (40) can store a prediction model for predicting the combustion state.

[0047] Here, the prediction model can predict combustion phenomena inside the boiler by receiving operating variables in real time. In other words, the prediction model is a model that outputs physical behavior as similar as possible to the actual boiler when given operating variables, and can be considered a virtual boiler or a twin model of the boiler. Therefore, to predict combustion phenomena, the prediction model may be a model created by training the output of a model generated through necessary physical models such as thermodynamics, fluid dynamics, and chemistry, or by training multiple outputs using methods such as machine learning.

[0048] In addition, to predict boiler combustion phenomena more accurately, the prediction model may include not only the inside of the boiler but also the inlet of each nozzle where secondary air is injected and the part that controls the opening rate thereof, and the calculation result of this prediction model becomes the air flow rate for each secondary air nozzle.

[0049] Meanwhile, memory (40) can store an optimization model that can derive an optimal value that satisfies the conditions for the driving variable of interest as much as possible using optimization conditions as input.

[0050] Here, the optimization model is a model created based on a prediction model. For example, when an optimization condition is input to maintain the heat transfer rate while keeping the outlet NOx concentration at its lowest and maintaining the power plant output at 500 MW ±5%, the optimization model can derive values ​​such as an excess air ratio of 1.14, an upper inlet secondary air nozzle damper opening rate of 78%, and a lower burner secondary air nozzle damper opening rate of 65%, which are operating variables that satisfy the condition within a preset margin.

[0051] Alternatively, the optimization model may only suggest the direction of whether to increase or decrease specific variables when aiming to minimize the outlet NOx concentration.

[0052] The processor (50) is operatively coupled to the operating variable input module (10), the simulation condition input module (20), the optimization condition input module (30), the operating information database (60), the output module (70), and the memory (40) to control the overall operation of the combustion state prediction device of the power plant boiler by copying various programs stored in the memory (40) to RAM and executing them to perform various operations.

[0053] Here, although the processor (50) is described as including only one CPU, it may be implemented with multiple CPUs (or DSP, SoC, etc.) during implementation.

[0054] In various embodiments, the processor (50) may be implemented as a digital signal processor (DSP), a microprocessor, or a TCON (Time controller) that processes digital signals. However, it is not limited thereto, and may include or be defined by one or more of a central processing unit (CPU), a Micro Controller Unit (MCU), a micro processing unit (MPU), a controller, an application processor (AP), a communication processor (CP), or an ARM processor. Additionally, the processor (50) may be implemented as a System on Chip (SoC) or Large Scale Integration (LSI) with a built-in processing algorithm, or may be implemented in the form of a Field Programmable Gate Array (FPGA).

[0055] That is, the processor (50) can run an execution program to predict the combustion state of the boiler based on a prediction model based on the operating variables input from the operating variable input module (10) and the operating information of the boiler stored in the operating information database (60), and output the prediction data through the output module (70).

[0056] Here, the processor (50) can predict physical characteristic values ​​corresponding to the spatial coordinates inside the boiler and the location of the spatial coordinates as prediction data.

[0057] In this case, the physical characteristic values ​​corresponding to the position of the spatial coordinates may include the temperature inside the boiler, the concentration of chemical species of interest that the user wishes to monitor such as oxygen and carbon dioxide, and the gas flow rate.

[0058] Additionally, if the prediction model includes a secondary air inlet, the processor (50) may include the air flow rate for each secondary air nozzle in the prediction data.

[0059] In addition, the processor (50) may predict the physical quantity distribution in three dimensions for the entire boiler furnace space or in two dimensions for a part of it, as provided by the results of a conventional computational thermal fluid analysis, in addition to real values ​​as prediction data, and may calculate the physical characteristic values ​​measured by the measuring device as prediction data to verify the reliability of the prediction model itself. For example, the average gas temperature at a corresponding location can be calculated as prediction data through the prediction model to compare with a temperature sensor installed at the furnace outlet of a coal-fired boiler.

[0060] The processor (50) can convert the predicted data, which predicts the combustion state in this way, into an information form such as an image or graph that the user can intuitively accept through the output module (70) as shown in FIGS. 2 and 3, and display the boiler state, operating state, and inquiry results.

[0061] For example, as shown in FIG. 2, when the combustion state of the boiler is queried, there may be a part that displays operating information data indicating the current state of the boiler, a part that displays data of operating variables that set the operating conditions of the boiler, and a part that displays the combustion state generated by utilizing prediction data.

[0062] In addition, as shown in Fig. 3, the result of checking the status of the secondary air inlet can be displayed.

[0063] Meanwhile, the processor (50) can receive simulated operating conditions through the simulated condition input module (20), and generate predicted data by predicting the combustion state of the boiler based on a prediction model based on the simulated operating conditions and the boiler operating information stored in the operating information database (60).

[0064] Accordingly, not only the combustion state based on real-time boiler operating variables but also possible future operating conditions can be input as simulation conditions and output through the output module (70) so as to compare, review, and improve them in terms of combustion.

[0065] On the other hand, the processor (50) derives optimal operating variables that can satisfy the optimization conditions input through the optimization condition input module (30) based on the optimization model, and outputs prediction data that predicts the combustion state of the boiler based on the optimal operating variables and the boiler operating information stored in the operation information database (60) through the output module (70).

[0066] That is, when a user inputs the desired optimization conditions through the optimization condition input module (30), the processor (50) can automatically find the optimal driving variables to achieve this and calculate and retrieve prediction data based on this.

[0067] Therefore, it is possible to find the optimal operating conditions desired by the user and verify the combustion state at that time.

[0068] For example, when the processor (50) sets an optimization condition through the optimization condition input module (30) to maintain the heat transfer amount while keeping the outlet NOx concentration at a minimum and maintaining the power plant output at 500 MW ±5%, it can derive values ​​such as an excess air ratio of 1.14, an upper inlet secondary air nozzle damper opening rate of 78%, and a lower burner secondary air nozzle damper opening rate of 65%, which are operating variables that satisfy the condition within a preset margin through the optimization model.

[0069] Accordingly, the processor (50) can predict prediction data based on the boiler operation information stored in the operation information database (60) based on the derived optimal operation variables.

[0070] In addition, the processor (50) may only suggest the direction of whether to increase or decrease specific variables when trying to minimize the outlet NOx concentration through an optimization model.

[0071] As described above, according to the combustion state prediction device for a power plant boiler according to an embodiment of the present invention, by receiving power plant operation data and operation variables through a prediction model based on a physical model to predict and monitor the combustion state inside a boiler in a coal-fired power plant in real time, it is possible not only to track and resolve the cause of failure but also to stabilize boiler combustion by reviewing operating conditions in various situations through combustion tuning to achieve optimal efficiency and minimize pollutants by tracking incomplete combustion.

[0072] FIG. 4 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to one embodiment of the present invention.

[0073] As illustrated in FIG. 4, a method for predicting the combustion state of a power plant boiler according to one embodiment of the present invention first involves a processor (50) running an execution program to receive operating variables for operation through an operating variable input module (10) and receiving operating information of the boiler from a measuring device (S10).

[0074] Here, operating variables are conditions for operating by manipulating the combustion environment of the actual boiler, and in the case of a coal-fired boiler, may include, for example, the amount of coal input, the opening rate of the primary and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is on / off for each layer, and the calorific value of the coal being input.

[0075] In addition, the operation information may include operation data acquired in real time from a measuring device installed in the boiler and control data used to control the boiler.

[0076] At this time, the driving information database (60) may include a data communication function, that is, a function to provide data to another program and record data provided by another program.

[0077] When driving variables and driving information are received in step S10, the processor (50) stores the received driving variables and driving information in the driving information database (60) (S20).

[0078] After storing the operating variables and operating information in step S20, the processor predicts the combustion state of the boiler based on the prediction model based on the operating variables and operating information (S30).

[0079] Here, the prediction model can predict combustion phenomena inside the boiler by receiving operating variables in real time. In other words, the prediction model is a model that outputs physical behavior as similar as possible to the actual boiler when given operating variables, and can be considered a virtual boiler or a twin model of the boiler. Therefore, to predict combustion phenomena, the prediction model may be a model created by training the output of a model generated through necessary physical models such as thermodynamics, fluid dynamics, and chemistry, or by training multiple outputs using methods such as machine learning.

[0080] In addition, to predict boiler combustion phenomena more accurately, the prediction model may include not only the inside of the boiler but also the inlet of each nozzle where secondary air is injected and the part that controls the opening rate thereof, and the calculation result of this prediction model becomes the air flow rate for each secondary air nozzle.

[0081] After predicting the combustion state of the boiler based on a prediction model based on operating variables and operation information in step S30, the processor outputs the predicted data of the combustion state through an output module (S40).

[0082] Here, the prediction data may include spatial coordinates inside the boiler and physical characteristic values ​​corresponding to the location of the spatial coordinates.

[0083] In this case, the physical characteristic values ​​corresponding to the position of the spatial coordinates may include the temperature inside the boiler, the concentration of chemical species of interest that the user wishes to monitor such as oxygen and carbon dioxide, and the gas flow rate.

[0084] Additionally, if the prediction model includes a secondary air inlet, the processor (50) may include the air flow rate for each secondary air nozzle in the prediction data.

[0085] In addition, the processor (50) may predict the physical quantity distribution in three dimensions for the entire boiler furnace space or in two dimensions for a part of it, as provided by the results of a conventional computational thermal fluid analysis, in addition to real values ​​as prediction data, and may calculate the physical characteristic values ​​measured by the measuring device as prediction data to verify the reliability of the prediction model itself. For example, the average gas temperature at a corresponding location can be calculated as prediction data through the prediction model to compare with a temperature sensor installed at the furnace outlet of a coal-fired boiler.

[0086] FIG. 5 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to another embodiment of the present invention.

[0087] According to another embodiment of the present invention, not only the combustion state based on real-time boiler operating variables but also possible future operating conditions can be input as simulation conditions and output through the output module (70) so as to compare, review, and improve them in terms of combustion.

[0088] That is, as illustrated in FIG. 5, a combustion state prediction method, in which a processor (50) can receive simulated operating conditions through a simulated condition input module (20) (S310).

[0089] When there is a problem with the current actual operating conditions of the boiler or when seeking improved conditions, new virtual operating conditions can be entered as simulated operating conditions.

[0090] When simulated operating conditions are received at step S310, the processor (50) can generate predicted data by predicting the combustion state of the boiler based on a prediction model based on the simulated operating conditions and the boiler operating information stored in the operating information database (60) (S315).

[0091] In this way, when the combustion state of the boiler is predicted by receiving simulated operating conditions, it may be displayed through the output module (70) along with the predicted data based on operating variables so that they can be compared.

[0092] FIG. 6 is a flowchart illustrating a method for predicting the combustion state of a power plant boiler according to another embodiment of the present invention.

[0093] According to another embodiment of the present invention, when a user inputs desired optimization conditions, the optimal driving variables to achieve them are automatically found, and prediction data is calculated and retrieved based thereon.

[0094] As illustrated in FIG. 6, in a combustion state prediction method, the processor (50) can receive an optimization condition from the optimization condition input module (30) (S320).

[0095] When an optimization condition is input at step S320, the processor (50) derives an optimal driving variable that can satisfy the optimization condition based on the optimization model (S322).

[0096] Here, the optimization model is a model created based on a prediction model. For example, when an optimization condition is input to maintain the heat transfer rate while keeping the outlet NOx concentration at its lowest and maintaining the power plant output at 500 MW ±5%, the optimization model can derive values ​​such as an excess air ratio of 1.14, an upper inlet secondary air nozzle damper opening rate of 78%, and a lower burner secondary air nozzle damper opening rate of 65%, which are operating variables that satisfy the condition within a preset margin.

[0097] Alternatively, the optimization model may only suggest the direction of whether to increase or decrease specific variables when aiming to minimize the outlet NOx concentration.

[0098] After deriving optimal operating variables in step S322, the processor (50) can generate prediction data by predicting the combustion state of the boiler based on a prediction model based on the optimal operating variables and the boiler operating information stored in the operating information database (60) (S324).

[0099] Accordingly, when the processor (50) sets an optimization condition through the optimization condition input module (30) to maintain the heat transfer amount while keeping the outlet NOx concentration at a minimum and maintaining the power plant output at 500 MW ±5%, it can derive values ​​such as an excess air ratio of 1.14, an upper inlet secondary air nozzle damper opening rate of 78%, and a lower burner secondary air nozzle damper opening rate of 65%, which are operating variables that satisfy the condition within a preset margin through the optimization model.

[0100] In addition, the processor (50) may only suggest the direction of whether to increase or decrease specific variables when trying to minimize the outlet NOx concentration through an optimization model.

[0101] As described above, according to the method for predicting the combustion state of a power plant boiler according to an embodiment of the present invention, the combustion state inside a boiler in a coal-fired power plant can be predicted and monitored in real time by receiving power plant operation data and operation variables through a prediction model based on a physical model. This allows not only to track and resolve the cause of failure, but also to stabilize boiler combustion by reviewing operating conditions in various situations through combustion tuning to achieve optimal efficiency, and to track incomplete combustion to minimize pollutants.

[0102] The implementations described herein may be implemented, for example, as methods or processes, devices, software programs, data streams, or signals. Even if discussed only in the context of a single form of implementation (e.g., discussed only as a method), the implementation of the discussed features may also be implemented in other forms (e.g., devices or programs). Devices may be implemented in appropriate hardware, software, and firmware, etc. Methods may be implemented in devices such as processors, which generally refer to processing devices including, for example, computers, microprocessors, integrated circuits, or programmable logic devices. Processors also include communication devices such as computers, cell phones, portable / personal digital assistants ("PDAs"), and other devices that facilitate the communication of information between end-users.

[0103] Although the present invention has been described with reference to the embodiments illustrated in the drawings, this is merely illustrative, and those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom.

[0104] Therefore, the true technical scope of protection of the present invention should be determined by the claims below. Explanation of the symbols

[0105] 10: Driving variable input module 20: Simulation condition input module 30: Optimization condition input module 40: Memory 50: Processor 60: Driving Information Database 70 : Output module

Claims

Claim 1 An operating variable input module for receiving operating variables for operating a boiler; an operating information database for storing operating information of the boiler; and an output module for displaying predicted data; and a processor operatively coupled to the operating variable input module, the operating information database, and the output module; wherein the processor drives an execution program to predict the combustion state of the boiler based on a prediction model based on the operating variable input from the operating variable input module and the operating information of the boiler stored in the operating information database, outputs prediction data through the output module, and further includes an optimization condition input module that receives the optimization conditions of the boiler desired by the user; wherein the processor derives optimal operating variables capable of satisfying the optimization conditions input through the optimization condition input module based on the optimization model, and predicts the combustion state of the boiler based on the prediction model based on the optimal operating variables and the operating information of the boiler stored in the operating information database, wherein the prediction model is a model generated through machine learning based on a physical model for predicting combustion phenomena in the boiler, and the prediction data is regarding spatial coordinates inside the boiler, physical characteristic values ​​corresponding to the location of the spatial coordinates, the air flow rate of the secondary air nozzle, and the distribution of physical quantities for the entire furnace space of the boiler A combustion state prediction device for a power plant boiler characterized by including one or more of a contour image and physical characteristic values ​​measured by a measuring device. Claim 2 A combustion state prediction device for a power plant boiler according to claim 1, characterized in that the operating variables include one or more of the following: a coal input amount, the opening rate of primary and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is on / off for each layer, and the calorific value of the input coal. Claim 3 A combustion state prediction device for a power plant boiler according to claim 1, characterized in that the operation information includes operation data acquired from a measuring device installed in the boiler and control data used for controlling the boiler. Claim 4 delete Claim 5 delete Claim 6 A combustion state prediction device for a power plant boiler according to claim 1, further comprising a simulation condition input module for receiving simulation operating conditions of the boiler; wherein the processor predicts the combustion state of the boiler based on the prediction model based on the simulation operating conditions input through the simulation condition input module and the operation information of the boiler stored in the operation information database. Claim 7 delete Claim 8 The method comprises the steps of: a processor running an execution program to receive operating variables for the operation of a boiler through an operating variable input module and receiving operating information of the boiler from a measuring device and storing it in an operating information database; the processor predicting the combustion state of the boiler based on a prediction model based on the operating variables and the operating information; and the processor outputting the predicted data of the predicted combustion state through an output module; wherein the step of predicting the combustion state of the boiler includes the steps of: the processor receiving an optimization condition for the boiler desired by a user from an optimization condition input module; and the processor deriving an optimal operating variable capable of satisfying the optimization condition based on an optimization model. The method for predicting the combustion state of a power plant boiler further comprises the step of the processor predicting the combustion state of the boiler based on the prediction model, based on the optimal operating variables and the operating information of the boiler stored in the operating information database; wherein the prediction model is a model generated through machine learning based on a physical model for predicting combustion phenomena in the boiler, and the prediction data includes one or more of the following: spatial coordinates inside the boiler, physical characteristic values ​​corresponding to the position of the spatial coordinates, the air flow rate of the secondary air nozzle, a contour image of the physical quantity distribution over the entire furnace space of the boiler, and physical characteristic values ​​measured by a measuring device. Claim 9 A method for predicting the combustion state of a power plant boiler according to claim 8, wherein the above operating variables include one or more of the following: the amount of coal input, the opening rate of the primary air and secondary air nozzles, the tilt and yaw angles of the burner, whether the burner is on / off for each layer, and the calorific value of the input coal. Claim 10 A method for predicting the combustion state of a power plant boiler according to claim 8, wherein the above-mentioned operation information includes operation data acquired from a measuring device installed in the boiler and control data used for controlling the boiler. Claim 11 delete Claim 12 delete Claim 13 A method for predicting the combustion state of a power plant boiler according to claim 8, wherein the step of predicting the combustion state of the boiler further comprises: a step in which the processor receives simulated operating conditions input through a simulated condition input module; and a step in which the processor predicts the combustion state of the boiler based on the prediction model based on the simulated operating conditions and the operating information of the boiler stored in the operating information database. Claim 14 delete

Images