KR101810968B1 - Power Plant Boiler Optimizer Having Boiler Model Based On Artificial Neuron Network, and Power Plant Boiler Having the Same - Google Patents

Abstract

Disclosed are a boiler combustion optimizer for a power plant boiler having an artificial neural network-based boiler model, and a power plant boiler including the same. The boiler combustion optimizer (100) for a power plant according to an embodiment of the present invention, which is a boiler combustion optimizer installed in a boiler comprising a combustion chamber (10) a heat exchange part (20), and an exhaust gas discharge part (30) in a mutually communicating structure, comprises: a plurality of first data detecting sensors (111) arranged to be a predetermined distance away from the combustion chamber (10); a plurality of second data detecting sensors (112) disposed to be a predetermined distance away from the heat exchange part (20); a plurality of third data detecting sensors (113) disposed to be a predetermined distance away from the exhaust gas discharge part (30); and an arithmetic processing unit (130) deriving an optimal operation control value based on data acquired through the first data detecting sensors (111), the second data detecting sensors (112) and the third data detecting sensors (113). According to the boiler combustion optimizer for the power plant boiler, it is possible to derive the optimal operation control value based on the obtained data, so remarkably improved combustion efficiency of the power plant boiler may be achieved. In addition, it is possible to derive an optimal operation control value for reducing generation of contaminants by providing an optimal combustion state.

Description

Technical Field [0001] The present invention relates to a power plant boiler combustion heater including an artificial neural network based boiler model, and a power plant boiler including the artificial neural network based boiler model,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a power plant boiler combustion electrometer and a power plant boiler, and more particularly, to a power plant boiler combustion plummeter and a power plant boiler including an artificial neural network based boiler model.

In the case of a boiler of a coal-fired power plant according to the prior art, an exothermic reaction occurring when coal is burned is used to heat water and produce steam required for power generation.

At this time, the boiler generates nitrogen oxides or carbon dioxide as exhaust gas. However, in the case of boilers of coal-fired power plants according to the prior art, the cost of managing such exhaust gases is further required.

Accordingly, a boiler combustion pressure optimizer capable of reducing the amount of generated gas while increasing the combustion efficiency of the boiler has been developed and operated.

However, in the case of the boiler combustion pressure meter according to the prior art, it is difficult to produce optimal combustion efficiency because it operates according to the proficiency of the boiler operation manager.

In the case of the boiler combustion pressure meter according to the related art, a virtual model having the same behavior as that of the actual boiler is created and utilized, but there is no technology to effectively utilize the model.

Accordingly, it is an object of the present invention to solve the problems of the prior art, to achieve a combustion boiler combustion efficiency which is remarkably improved, and at the same time to provide an optimum combustion state, And a technology for a power plant boiler including the steam generator is required.

Korean Registered Patent No. 10-1484496 (registered on Jan. 14, 2015)

It is an object of the present invention to provide a power plant boiler combustion pressure gauge capable of achieving a significantly improved power plant boiler combustion efficiency and at the same time providing an optimum combustion condition to thereby obtain an optimum operation control value for reducing the generation of pollutants and the like To provide a power plant boiler.

To achieve these and other advantages and in accordance with the purpose of the present invention, as embodied and broadly described herein, there is provided a boiler combustion microcomputer installed in a boiler including a combustion chamber, a heat exchange unit, and an exhaust gas discharge unit, A plurality of first data detection sensors arranged at a predetermined distance apart from each other; A plurality of second data detecting sensors disposed at a predetermined distance from the heat exchanging unit; A plurality of third data detection sensors disposed at predetermined intervals in the exhaust gas discharging portion; And an arithmetic processing unit that is operated by an artificial neural network algorithm based on data acquired through the first data detection sensor, the second data detection sensor, and the third data detection sensor to derive an optimal operation control value .

In one embodiment of the present invention, the arithmetic processing unit 130 analyzes the input value data, stores the analysis result, transmits the control value to the optimum route control unit using the normal network, An advisor for receiving and storing another output value to be transmitted to the optimum route control unit and transmitting an optimal control value among the previously stored control values to the optimum route control unit; And an optimal route control part for updating a control value for controlling each of the burners based on the optimal control value received from the adviser in real time and for determining an optimal control value for each input value data Lt; / RTI >

In one embodiment of the present invention, at least one first layered structure having a lattice structure spaced apart from each other by a predetermined distance is formed in the combustion chamber, and in the first layered structure, As shown in FIG.

In one embodiment of the present invention, at least two first data detecting sensors for detecting the state of the inside of the combustion chamber using optical methods may be mounted on the side surface of the combustion chamber so as to be separated from each other by a predetermined distance.

In one embodiment of the present invention, the heat exchange unit includes a plurality of heat exchange pipes, at least one second layered structure having a lattice shape at a position corresponding to the heat exchange pipe, 2 Data detection sensor can be mounted according to a predetermined pattern.

In one embodiment of the present invention, at least one third layered structure is formed in the exhaust gas discharging portion, the third layered structure having a lattice-shaped structure spaced apart from the exhaust gas discharging portion by a predetermined distance, and the third data detection sensor And can be mounted according to a predetermined pattern.

In one embodiment of the present invention, the first data detection sensor, the second data detection sensor, and the third data detection sensor detect the temperature of the flame, the temperature of the exhaust gas, the oxygen concentration, the carbon monoxide concentration, the nitrogen oxide concentration, And the flow rate of the passing fluid can be detected.

In one embodiment of the present invention, the arithmetic processing unit may be an artificial neural network based system.

In this case, the arithmetic processing unit can generate a virtual boiler model showing the same behavior based on the data obtained through the first data detecting sensor, the second data detecting sensor, and the third data detecting sensor.

The arithmetic processing unit may derive a correlation between an input value and an output value after substituting a plurality of different input values into the virtual boiler model and arithmetically processing output values corresponding thereto.

Further, the arithmetic processing unit can derive an optimal operation control value using the derived correlation.

According to the present invention, it is also possible to provide a power plant boiler including the power plant boiler combustion heater.

As described above, according to the power plant boiler combustion pressure meter of the present invention, by providing the first data detection sensor, the second data detection sensor, the third detection sensor and the arithmetic processing section of a specific configuration, Which can significantly improve the power plant boiler combustion efficiency and at the same time provide the optimal combustion state and thereby can provide an optimum operation control value for reducing the generation of pollutants, And a power plant boiler including the same.

In addition, according to the power plant boiler combustion pressure meter of the present invention, the first to third data detection sensors are mounted on the first to third layered structures in a lattice-shaped structure spaced apart from each other by a predetermined distance, , It is possible to accurately acquire various data according to each area, and thereby it is possible to derive a more specific correlation between the respective data.

Further, according to the power plant boiler combustion pressure meter of the present invention, based on the acquired data, the arithmetic processing unit generates a virtual boiler model by using only the artificial nerve based system, so that the virtual model exhibiting the same behavior as the actual power plant boiler Can be generated.

Also, according to the present invention, it is possible to easily derive the correlation between the input value and the output value by using the artificial neural based system, and to obtain the optimal operation control value.

Further, according to the power plant boiler of the present invention, by providing the power plant boiler combustion pressure meter according to the present invention, it is possible to achieve a remarkably improved power plant boiler combustion efficiency and at the same time provide an optimum combustion state, have.

1 is a schematic diagram showing a boiler according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a state where a power plant boiler is operated according to an optimal operation control value derived by an arithmetic processing unit according to an embodiment of the present invention.
FIG. 3 is a block diagram illustrating an artificial neural network algorithm for operating an arithmetic processing unit according to an exemplary embodiment of the present invention. The algorithm includes an optimal route control method (MPC) and a destination control and burner selection method FIG.
4 is a schematic diagram illustrating a process of processing data acquired by the arithmetic processing unit according to an embodiment of the present invention.
5 is a table showing an input value and an output value input to the arithmetic processing unit according to an embodiment of the present invention.
6 is a schematic diagram illustrating a system structure of an artificial neural network based on an arithmetic processing unit according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. Prior to the description, terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary meanings and should be construed in accordance with the technical concept of the present invention.

Throughout this specification, when a member is "on " another member, this includes not only when the member is in contact with another member, but also when there is another member between the two members. Throughout this specification, when an element is referred to as "including" an element, it is understood that it may include other elements as well, without departing from the other elements unless specifically stated otherwise.

FIG. 1 is a schematic diagram showing a boiler according to an embodiment of the present invention. FIG. 2 is a schematic view of a boiler according to an embodiment of the present invention when the power plant boiler is operated according to an optimum operation control value derived by an arithmetic processing unit And a figure showing a change is shown.

FIG. 3 is a block diagram illustrating an artificial neural network algorithm for operating an arithmetic processing unit according to an exemplary embodiment of the present invention. The algorithm includes an optimal route control method (MPC) and a destination control and burner selection method FIG.

FIG. 4 is a schematic diagram illustrating a process of processing data acquired by the arithmetic processing unit according to an embodiment of the present invention. FIG. 5 shows an input value And a table showing output values are shown. FIG. 6 is a schematic diagram illustrating a system structure of an artificial neural network based on an arithmetic processing unit according to an embodiment of the present invention.

Referring to these drawings, the power plant boiler combustion meth- oder 100 according to the present embodiment uses an artificial neural network, which is an engineering model of a human brain cell structure.

Artificial Neural Networks Artificial Neural Networks use algorithms that mimic the ability of human beings to have learning. Through this algorithm, the artificial neural network can generate the mapping between the input pattern and the output pattern, which expresses that the artificial neural network has learning ability. In addition, the artificial neural network has a generalization ability to generate relatively correct output for input patterns that were not used for learning based on the learned results.

Because of the two representative performances of learning and generalization, artificial neural networks are applied to problems that are hardly solved by conventional sequential programming methods. Artificial neural networks are widely applied to pattern classification problems, continuous mapping, nonlinear system identification, nonlinear control, and robot control.

In the case of the artificial neural network applied to the combustion boiler of the power plant boiler according to the present embodiment, the correlation between the generated gas and the combustion efficiency due to the control of the burner and OFA can be derived, have.

An artificial neural network represents a computational model implemented in software or hardware that mimics the computational capability of a biological system using a large number of artificial neurons connected by a link.

Artificial neurons use artificial neurons that simplify the function of biological neurons. Then, they connect each other through a connection line having a connection strength to perform human cognitive function or learning process. The connection strength is a specific value of a connection line, also called a connection weight.

Artificial neural networks can be divided into instructional learning and non - instructional learning.

The map learning is a method in which input data and output data corresponding thereto are put together into a neural network and the connection strength of the connecting lines is updated so as to output output data corresponding to the input data. Representative learning algorithms include delta rules and backpropagation learning.

Bidirectional learning is a method in which the artificial neural network learns its own connection strength using only input data without a target value. Unsupervised learning is a method of updating connection weights by correlation between input patterns.

The artificial neural network applied to the power plant boiler combustion presser 100 according to the present embodiment may be constituted by map learning or non-map learning.

The above-described artificial neural network algorithm is stored in a programmable manner in the arithmetic processing unit 130 according to the present embodiment.

Hereinafter, each constitution of the power plant boiler combustion meth- oder 100 according to the present embodiment will be described in detail.

1, the power plant boiler combustion press machine 100 according to the present embodiment is mounted on a boiler including a combustion chamber 10, a heat exchange section 20 and an exhaust gas discharge section 30 of a structure communicated with each other. Which is the boiler combustion pressure meter.

The power plant boiler combustion meth- oder 100 according to the present embodiment can derive an optimum operation control value based on the acquired data after detecting specific data through a plurality of data detection sensors.

To this end, the power plant boiler combustion microcomputer 100 according to the present embodiment includes a first data detection sensor 111, a second data detection sensor 112, a plurality of third data detection sensors 113, and an arithmetic processing unit 130 ). ≪ / RTI >

As shown in FIG. 1, the first data detection sensor 111 may be disposed in the combustion chamber 10 at a predetermined distance.

More specifically, at least one first layered structure 121 is formed in the combustion chamber 10 and has a lattice structure spaced apart from each other by a predetermined distance. The first layered structure 121 is provided with a first data detection sensor 111, Can be mounted according to a predetermined pattern.

In some cases, more than two first data detecting sensors 111 for detecting the state of the inside of the combustion chamber 10 using an optical method may be spaced from each other by a predetermined distance on the side surface of the combustion chamber 10.

The second data detection sensor 112 may be disposed in the heat exchange unit 20 at a predetermined distance.

Specifically, as shown in FIG. 1, a plurality of heat exchange pipes 21 are disposed in the heat exchange unit 20, and a second layered structure 122 having a lattice shape at a position corresponding to the heat exchange pipe 21 And the second data detection sensor 112 may be mounted to the second layered structure 122 according to a predetermined pattern.

On the other hand, the third data detection sensor 113 may be arranged in a plurality of positions spaced apart by a predetermined distance from the exhaust gas discharging portion 30. [

More specifically, as shown in FIG. 1, at least one third layered structure 123 is formed in the exhaust gas discharging portion 30 and has a lattice-like structure spaced apart from each other by a predetermined distance. The third layered structure The third data detection sensor 113 may be mounted in accordance with a predetermined pattern.

The first data detection sensor 111, the second data detection sensor 112 and the third data detection sensor 113 described above can detect specific data at each position. For example, The temperature of the exhaust gas, the oxygen concentration, the carbon monoxide concentration, the nitrogen oxide concentration, the velocity of the passing fluid, and the flow rate of the passing fluid. The data that can be detected by the first data detection sensor 111, the second data detection sensor 112 and the third data detection sensor 113 is not limited to the above-mentioned example.

FIG. 4 is a schematic diagram illustrating a process of processing data acquired by the arithmetic processing unit according to an embodiment of the present invention. FIG. 5 illustrates an input value and an output value input to the arithmetic processing unit according to an embodiment of the present invention. ≪ / RTI > FIG. 5 is a schematic diagram illustrating an artificial neural network-based system structure of an arithmetic processing unit according to an embodiment of the present invention.

Referring to these figures, the arithmetic processing unit 130 according to the present embodiment calculates data obtained through the first data detection sensor 111, the second data detection sensor 112, and the third data detection sensor 113 It is possible to derive the optimum operation control value.

3, the arithmetic processing unit 130 according to the present embodiment includes a first data detection sensor 111, a second data detection sensor 112, and a third data detection Based on the data obtained through the sensor 113, a virtual boiler model showing the same behavior can be generated.

4 and 5, the arithmetic processing unit 130 of the present embodiment substitutes a plurality of different input values into a virtual boiler model, performs arithmetic processing on the corresponding output values, The correlation of the output values can be derived.

Furthermore, the arithmetic processing unit 130 according to the present embodiment can derive an optimal operation control value using the derived correlation.

The input values mentioned above are the coal flow rate, the secondary air flow rate, the secondary air flow entering through the overfiring air (OFA), and the temperature value of the secondary air entering the boiler as respective burners. In addition, the above-mentioned output values are FEGT (FURNACE EXIT GAS TEMPERATURE), gas temperature on the exit of the boiler economizer, residual oxygen, nitrogen oxide, and carbon monoxide amount.

6, the arithmetic processing unit 130 according to the present embodiment includes an input node layer x having n input nodes and an input node layer x from the input node layer x, A hidden node layer h having k hidden nodes through learning algorithms having different weights and h output node layers y corresponding to the output terminals of the arithmetic processing unit 130. [

The arithmetic processing unit 130 according to this embodiment has the first data detection sensor 111 installed inside or on the side of the combustion chamber 10 of the actual power plant boiler 10 and the second data detection sensor 111 installed inside the heat exchange unit 20. [ Based on the data acquired through the data detection sensor 112 and the third data detection sensor 113 provided in the exhaust gas discharge unit 30, each specific data having a high degree of correlation between the input value and the output value is selected .

After selecting each specific data with high correlation, each node constituting the input node layer (x) is connected to each node of the hidden layer through a connection line having different weights. At this time, the weights of the respective connection lines are changed through the horses algorithm. The number of nodes of the hidden node layer h may be changed to an arbitrary number according to the learning algorithm. The output node layer y has h output nodes and each appears as a plurality of outputs, as shown in Fig.

That is, it is possible to derive a correlation between each data by inputting a specific input value and an output value through an operation of an artificial neural network-based system, and then to derive the optimum control value of the boiler combustion operation of the power plant.

Specifically, the arithmetic processing unit 130 according to the present embodiment is operated by an artificial neural network algorithm as described above, and can be operated by map learning or non-map learning.

The power plant boiler combustion microcomputer 110 according to the present embodiment calculates the amount of secondary air flowing into the boiler according to the coal flow rate, the secondary air flow rate, the secondary air flow rate through OFA (Over Firing Air) 1 data detection sensor 111, the second data detection sensor 112, and the third data detection sensor 113, as shown in FIG.

The detected output values are FEGT (FURNACE EXIT GAS TEMPERATURE), gas temperature at the exit of the boiler economizer, residual oxygen, nitrogen oxide, and carbon monoxide.

The arithmetic processing unit 130, which is operated by the artificial neural network algorithm, learns output values changed by each input value, and then learns and stores the association with the output values.

FIG. 3 is a block diagram illustrating an artificial neural network algorithm for operating an arithmetic processing unit according to an embodiment of the present invention. The algorithm includes an optimal route control (MPC) and a destination control and burner selection method A configuration diagram is shown.

In some cases, as shown in FIG. 3, the neural network for operating the arithmetic processing unit according to the present embodiment may further include an adviser part for facilitating an optimal control value search.

As shown in FIG. 3, the advisor part analyzes the input value data (CFD), stores the analysis results (DB), and controls the burners by using a normal network (NN, Neural Network) Value is transmitted and controlled. At the same time, the control value sent to each burner is re-entered into the advisor unit and stored.

Thereafter, the other result values operated by each burner are analyzed by another analyzer, the result is again stored in the advisor part, and compared with the previously stored results The optimum control value is found.

The various correlations learned by the artificial neural network algorithm can be more easily derived using the above-mentioned adviser part.

Using these series of processes and methods, each burner can derive optimal coal flow rate, secondary air flow rate, secondary air flow through OFA (Over Firing Air), and temperature value of the secondary air entering the boiler.

The present invention can also provide a power plant boiler including the power plant boiler combustion meth- od 100 according to the present invention mentioned above.

As described above, according to the power plant boiler combustion pressure meter of the present invention, by providing the first data detection sensor, the second data detection sensor, the third detection sensor and the arithmetic processing section of a specific configuration, Which can significantly improve the power plant boiler combustion efficiency and at the same time provide the optimal combustion state and thereby can provide an optimum operation control value for reducing the generation of pollutants, And a power plant boiler including the same.

In addition, according to the power plant boiler combustion pressure meter of the present invention, the first to third data detection sensors are mounted on the first to third layered structures in a lattice-shaped structure spaced apart from each other by a predetermined distance, , It is possible to accurately acquire various data according to each area, and thereby it is possible to derive a more specific correlation between the respective data.

Further, according to the power plant boiler combustion pressure meter of the present invention, based on the acquired data, the arithmetic processing unit generates a virtual boiler model by using only the artificial nerve based system, so that the virtual model exhibiting the same behavior as the actual power plant boiler Can be generated.

Also, according to the present invention, it is possible to easily derive the correlation between the input value and the output value by using the artificial neural based system, and to obtain the optimal operation control value.

Further, according to the power plant boiler of the present invention, by providing the power plant boiler combustion pressure meter according to the present invention, it is possible to achieve a remarkably improved power plant boiler combustion efficiency and at the same time provide an optimum combustion state, have.

In the foregoing detailed description of the present invention, only specific embodiments thereof have been described. It is to be understood, however, that the invention is not to be limited to the specific forms thereof, which are to be considered as being limited to the specific embodiments, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims. .

That is, the present invention is not limited to the above-described specific embodiment and description, and various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims. And such variations are within the scope of protection of the present invention.

1: actual power plant boiler
10: Combustion chamber
11: Combustion nozzle
20: Heat exchanger
21: Heat exchange piping
30: exhaust gas discharging portion
100: Power plant boiler Combustion steamer
111: first data detection sensor
112: second data detection sensor
113: third data detection sensor
114: Laser sensor
121: first layer structure
122: second layer structure
123: Layer 3 structure
130: arithmetic processing unit

Claims (12)

1. A boiler combustion potentiometer (100) mounted on a boiler including a combustion chamber (10), a heat exchange section (20) and an exhaust gas discharge section (30)
A plurality of first data detection sensors (111) arranged at predetermined distances from the combustion chamber (10);
A plurality of second data detecting sensors 112 disposed at a predetermined distance from the heat exchanging unit 20;
A plurality of third data detection sensors 113 disposed at a predetermined distance from the exhaust gas discharging unit 30; And
A first data detection sensor 111, a second data detection sensor 112, and a third data detection sensor 113. The first data detection sensor 111, the second data detection sensor 112, and the third data detection sensor 113 are implemented by an artificial neural network- A virtual boiler model representing the same behavior is generated based on the data obtained through the calculation, and a plurality of different input values are substituted into a virtual boiler model, and the corresponding output values are arithmetically processed. Then, An arithmetic processing unit (130) for deriving a correlation of the input signal;
Lt; / RTI >
The arithmetic processing unit (130)
The control unit transmits the control value to the optimum route control unit using the normal network, inputs and stores another output value generated after the control value is transmitted, An advisor for transmitting the control value of the optimum route control unit to the optimum route control unit; And
An optimal route control part for updating a control value for controlling each burner on the basis of an optimum control value received from an advisor in real time and determining an optimum control value for each input value data;
Wherein the power generator boiler is a boiler.
delete The method according to claim 1,
At least one first layered structure 121 is formed in the combustion chamber 10 and has a lattice-like structure spaced apart from each other by a predetermined distance.
Wherein a first data detection sensor (111) is mounted on the first layered structure (121) in accordance with a predetermined pattern.
The method according to claim 1,
On the side surface of the combustion chamber (10)
Wherein at least two first data detection sensors (111) for detecting a state inside the combustion chamber (10) using optical methods are spaced apart from each other by a predetermined distance.
The method according to claim 1,
A plurality of heat exchange pipes 21 are disposed in the heat exchange unit 20,
At least one second layered structure 122 is formed at a position corresponding to the heat exchange pipe 21,
And a second data detection sensor (112) is mounted to the second layered structure (122) according to a predetermined pattern.
The method according to claim 1,
At least one third layered structure 123 is formed in the exhaust gas discharging part 30 and has a lattice-like structure spaced apart from each other by a predetermined distance.
And a third data detection sensor (113) is attached to the third layered structure (123) in accordance with a predetermined pattern.
The method according to claim 1,
The first data detection sensor 111, the second data detection sensor 112 and the third data detection sensor 113 are provided for detecting the temperature of the flame, the temperature of the exhaust gas, the oxygen concentration, the carbon monoxide concentration, the nitrogen oxide concentration, And the flow rate of the passing fluid is detected.
delete delete delete The method according to claim 1,
The arithmetic processing unit (130)
And derives an optimal operating control value using the derived correlation.
11. A power plant boiler, comprising a power plant boiler combustion pressurizer (100) according to any one of claims 1, 3, 7, and 11.

Images