CN115015134A - Boiler complex high-temperature flue gas component and concentration inversion method and system - Google Patents
Boiler complex high-temperature flue gas component and concentration inversion method and system Download PDFInfo
- Publication number
- CN115015134A CN115015134A CN202210178859.6A CN202210178859A CN115015134A CN 115015134 A CN115015134 A CN 115015134A CN 202210178859 A CN202210178859 A CN 202210178859A CN 115015134 A CN115015134 A CN 115015134A
- Authority
- CN
- China
- Prior art keywords
- flue gas
- boiler
- absorption spectrum
- absorption
- concentration
- 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.)
- Pending
Links
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 title claims abstract description 107
- 239000003546 flue gas Substances 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000000862 absorption spectrum Methods 0.000 claims abstract description 89
- 238000013136 deep learning model Methods 0.000 claims abstract description 34
- 239000000779 smoke Substances 0.000 claims abstract description 8
- 239000007789 gas Substances 0.000 claims description 71
- 238000010521 absorption reaction Methods 0.000 claims description 47
- 238000001228 spectrum Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 24
- 230000003287 optical effect Effects 0.000 claims description 21
- 238000011176 pooling Methods 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 12
- 239000000203 mixture Substances 0.000 claims description 11
- 230000008033 biological extinction Effects 0.000 claims description 9
- 230000003595 spectral effect Effects 0.000 claims description 7
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 239000002245 particle Substances 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 230000031700 light absorption Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000011541 reaction mixture Substances 0.000 claims description 3
- 238000004847 absorption spectroscopy Methods 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 abstract description 5
- 238000005259 measurement Methods 0.000 description 8
- 238000005260 corrosion Methods 0.000 description 5
- 230000007797 corrosion Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- -1 comprises CO 2 Chemical compound 0.000 description 1
- 238000002848 electrochemical method Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000004868 gas analysis Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biochemistry (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Computing Systems (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a method and a system for inverting components and concentration of complex high-temperature flue gas of a boiler, and belongs to the technical field of thermal engineering. The method comprises the following steps: acquiring smoke absorption spectrum data of a boiler to be detected; inputting the boiler flue gas absorption spectrum data to be detected into a multi-task deep learning model to obtain boiler flue gas components and concentration parameters; the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations. The deep learning algorithm is constructed based on the absorption spectrum data of the boiler flue gas with different temperatures and different concentrations, the constructed deep learning algorithm is used for quickly and accurately inverting the actually measured boiler flue gas absorption spectrum data to obtain various main components and concentration parameters of the boiler flue gas, the range of the components and the concentration of the boiler flue gas with complicated high temperature can be effectively covered, and redundant data are reduced.
Description
Technical Field
The invention relates to the technical field of thermal engineering, in particular to a boiler complex high-temperature flue gas composition and concentration inversion method based on an absorption spectrum and a boiler complex high-temperature flue gas composition and concentration inversion system based on the absorption spectrum.
Background
The boiler flue gas mainly comprises CO 2 、H 2 O、CO、O 2 、NO、NO 2 、SO 2 、H 2 S, etc., wherein H 2 S gas corrosion is the main cause of high-temperature corrosion of water-cooled walls, and CO and O 2 Gas concentration is related to water wall high temperature corrosion, H 2 S、CO、O 2 The measurement of the gases has important reference value for researching the corrosion mechanism of the water wall of the boiler and predicting the high-temperature corrosion condition of the water wall of the boiler. This is achieved byExternal, NO 2 、SO 2 When the gas is the main pollutant discharged by the boiler, CO 2 、H 2 Gases such as O are important data for evaluating the operation of the boiler. Therefore, the measurement of the smoke components of the boiler is one of the important contents of the thermal measurement of the boiler.
At present, components of boiler flue gas are mainly analyzed by a flue gas analyzer consisting of different gas electrochemical sensors to obtain various components and concentrations of the flue gas, and the electrochemical method is subjected to cross interference of various components of the flue gas, so that the measurement accuracy is low. The absorption spectrum technology utilizes the certain quantitative relation between the concentration of the gas to be measured and the light intensity attenuation degree of a specific wave band, the gas to be measured can be conveniently determined through the mathematical relation, the concentration value of the gas to be measured is monitored, and the method has the advantages of high response speed, online measurement, convenience in maintenance, real-time monitoring, strong cross interference resistance and the like, and can effectively improve the measurement precision by combining a multi-reflection absorption cell. However, the existing absorption spectrum technology for flue gas analysis usually analyzes a single gas, and does not consider the influence of the simultaneous existence of a plurality of complicated gases on the measurement of the concentration of the single gas.
Disclosure of Invention
The invention aims to provide a method and a system for inverting components and concentrations of complex high-temperature flue gas of a boiler.
In order to achieve the above object, a first aspect of the present invention provides a method for inverting components and concentrations of complex high-temperature flue gas of a boiler based on absorption spectrum, where the complex high-temperature flue gas of the boiler mainly includes CO 2 、H 2 O、 CO、O 2 、NO、NO 2 、SO 2 、H 2 S and the like, the method comprises the following steps:
acquiring smoke absorption spectrum data of a boiler to be detected;
inputting the boiler flue gas absorption spectrum data to be detected into a multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
Optionally, the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations is obtained through the following steps:
calculating the optical thickness OD of the complex boiler flue gas with different temperatures and different concentrations according to the absorption spectrum principle by taking the temperature of the complex high-temperature flue gas of the boiler, main gas components and concentration parameters as independent variables;
drawing according to the optical thicknesses OD of different wave bands of the complex boiler flue gas with the same temperature and the same concentration to obtain the absorption spectrum data of the complex boiler flue gas with the same temperature and the same concentration;
and changing the temperature, main gas components and concentration parameters of the complex high-temperature flue gas of the boiler, and drawing to obtain the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
Optionally, the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations is stored in an absorption spectrum library.
Optionally, the absorption spectrum principle is as follows:
for a single gas, under the irradiation of light with different wavelengths, the incident light and the intensity of the transmitted light meet the following relationship:
I(λ)＝I 0 (λ)exp[-σ(λ)CL] (1)
in formula (1), I (λ) is the transmitted light intensity of light with wavelength λ emitted by the light source after passing through the gas to be measured, I 0 (λ) is the initial light intensity of light with wavelength λ emitted by the light source, L is the gas optical path, C is the concentration of the gas to be measured, σ (λ) is the absorption cross section of the gas to be measured, related to wavelength λ;
according to the principle that the complex high-temperature flue gas light attenuation of the boiler comprises extinction caused by Rayleigh scattering and Mie scattering, the formula (1) is changed into the formula:
in the formula (2), epsilon R Is the Rayleigh scattering extinction coefficient, epsilon M Is the mie scattering extinction coefficient;
according to the fact that various gas components of the complex high-temperature flue gas of the boiler absorb light emitted by a light source, the influence of other gas molecules and particles is considered, and the formula (2) is expressed as follows:
in formula (3), σ i And C i Respectively showing the absorption cross section and concentration of the ith gas for light absorption;
according to the principle of slow change and fast change of gas absorption, the absorption section sigma (lambda) of the gas to be measured is expressed as follows:
σ(λ)＝σ s (λ)+σ f (λ) (4)
in the formula (4), σ s (λ) is the gas absorption slow-change absorption cross-section, σ f (lambda) gas absorption fast-changing absorption cross section;
according to the 'slow change' and 'fast change' principle in gas absorption, obtaining broadband absorption and narrow-band absorption:
the optical thickness OD at different wavelengths λ is defined as:
in the formula (6), the reaction mixture is,
is broadband absorption, i.e. broadband absorption caused by broadband absorption and scattering in gas absorption;
is a narrow band absorption;
considering that the light attenuation of the complex high-temperature flue gas of the boiler is influenced by the temperature, the optical thickness OD is subjected to temperature correction according to the formula (7):
the accurate optical thickness of the complex high-temperature flue gas of the boiler can be calculated through the method.
Optionally, the multitask deep learning model is established through the following steps:
converting the absorption spectrum data in the absorption spectrum library into a spectrogram by adopting a spectrum two-dimensional image conversion algorithm to obtain a two-dimensional spectrum information matrix;
and constructing a multi-task deep learning model according to the two-dimensional spectral information matrix.
Optionally, the converting the absorption spectrum data in the absorption spectrum library into a spectrogram by using a spectrum two-dimensional image conversion algorithm to obtain a two-dimensional spectrum information matrix includes:
processing the absorption spectrum data according to formula (8):
S＝XX T (8)
in the formula (8), S is a two-dimensional spectrum information matrix, and X is a spectrum data column vector, a typical two-dimensional spectrum information matrix can be obtained:
in the formula (9), a i Is one-dimensional spectral data.
Optionally, the constructing a multitask deep learning model according to the two-dimensional spectral information matrix includes:
constructing a common layer comprising a plurality of convolution kernels and a maximum pooling layer;
constructing convolution kernel branches of different tasks to obtain an initial multi-task deep learning model comprising a common layer and the convolution kernel branches of the different tasks;
inputting the two-dimensional spectrum information matrix into the initial multi-task deep learning model, performing repeated iterative training, and determining principal component factors and optimal weights of the initial multi-task deep learning model to obtain the multi-task deep learning model;
wherein each convolution kernel branch corresponds to different temperatures, different main gas components and different concentration parameters.
Optionally, constructing a common layer including a plurality of convolution kernels and a maximum pooling layer includes:
the convolutional neural network of the formula (10) is adopted for construction:
in the formula (10), f (m) is a typical two-dimensional spectrum information matrix constructed by the formula (9), n is the length of the signal f (n), and S (n) is a convolution result sequence with the length of len (f (n)) + len (g (n)) -1.
The invention provides a boiler complex high-temperature flue gas component and concentration inversion system based on absorption spectrum, which comprises:
the data acquisition module is used for acquiring the smoke absorption spectrum data of the boiler to be detected;
the inversion module is used for inputting the boiler flue gas absorption spectrum data to be detected into the multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
In another aspect, the present invention provides a machine-readable storage medium, having stored thereon instructions for causing a machine to execute the method for inverting the complex high-temperature flue gas composition and concentration of a boiler based on absorption spectroscopy.
By the technical scheme, the method for inverting the components and the concentration of the complex high-temperature flue gas of the boiler based on the absorption spectrum is provided,the complex high-temperature flue gas of the boiler in the method mainly comprises CO 2 、H 2 O、CO、 O 2 、NO、NO 2 、SO 2 、H 2 And S, the method comprises the steps of obtaining absorption spectrum data of the boiler flue gas with different temperatures and different concentrations, constructing a multi-task deep learning algorithm based on the absorption spectrum data, and quickly and accurately inverting the measured boiler flue gas absorption spectrum data through the multi-task deep learning algorithm to obtain various main components and concentration parameters of the boiler flue gas, so that the range of the components and the concentration of the boiler flue gas with different temperatures and different concentrations can be effectively covered, and redundant data can be reduced.
The measured boiler flue gas absorption spectrum data is quickly and accurately inverted based on a multitask deep learning algorithm, various main components and concentration parameters of the boiler flue gas can be synchronously obtained, mutual interference and particle influence of various gas components can be effectively eliminated, and inversion accuracy is improved.
According to the invention, the absorption spectrum data of the complex boiler flue gas is converted into the spectrogram through a spectrum two-dimensional image conversion algorithm, so that a two-dimensional spectrum information matrix is established, a multitask network is constructed through a multitask convolution network form construction method, and multiple iterative training is carried out to find the principal component factors and the optimal weights of the network, so that the measured boiler flue gas absorption spectrum data is inverted through the obtained principal component factors and the optimal weights of the principal component factors, the iteration efficiency is effectively improved, and the inversion calculation time is reduced.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention and do not limit the embodiments. In the drawings:
FIG. 1 is a flow chart of a method for inverting the components and concentration of complex high-temperature flue gas of a boiler based on absorption spectrum according to an embodiment of the invention;
FIG. 2 is a diagram of a multitask network constructed by the method for constructing a form of a multitask convolutional network according to an embodiment of the present invention;
FIG. 3 is a block diagram of a system for inverting the complex high-temperature flue gas components and concentrations of a boiler based on absorption spectra according to an embodiment of the invention.
Detailed Description
The following describes in detail embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
Fig. 1 is a flowchart of a method for inverting the complex high-temperature flue gas components and concentration of a boiler based on absorption spectrum according to an embodiment of the present invention. As shown in fig. 1, the method includes:
the method comprises the following steps: acquiring smoke absorption spectrum data of a boiler to be detected; in this embodiment, the boiler flue gas absorption spectrum data to be measured is obtained by measuring the boiler flue gas to be measured according to the absorption spectrum technology.
Step two: inputting the boiler flue gas absorption spectrum data to be detected into a multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
In this embodiment, the complex high-temperature flue gas of the boiler mainly comprises CO 2 、H 2 O、CO、O 2 、 NO、NO 2 、SO 2 、H 2 S and the like.
In this embodiment, the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations is obtained through the following steps:
1) and calculating the optical thickness OD of the complex boiler flue gas with different temperatures and different concentrations according to the absorption spectrum principle by taking the temperature, main gas components and concentration parameters of the complex high-temperature flue gas of the boiler as independent variables.
2) And drawing according to the optical thicknesses OD of different wave bands of the complex boiler flue gas with the same temperature and the same concentration to obtain the absorption spectrum data of the complex boiler flue gas with the same temperature and the same concentration. The absorption spectrum is plotted with the optical thickness OD as the vertical axis and the wavelength band as the horizontal axis. The drawing process is a splicing process of optical thicknesses OD of different wave bands of the complex boiler flue gas with the same temperature and the same concentration on a horizontal axis.
3) And changing the temperature, main gas components and concentration parameters of the complex high-temperature flue gas of the boiler, and drawing to obtain the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
In some embodiments, the absorption spectrum data of the different temperatures, different concentrations of complex boiler flue gas is stored in an absorption spectrum library.
In the application, the principle of the absorption spectrum technology lies in that the concentration of the gas to be detected and the light intensity attenuation degree of a specific wave band have a certain quantitative relation, the gas to be detected can be conveniently determined through the mathematical relation, and the concentration value of the gas to be detected is monitored. In this embodiment, the principle of absorption spectrum is:
for a single gas, under the irradiation of light with different wavelengths, the incident light and the intensity of the transmitted light conform to the following relationship:
I(λ)＝I 0 (λ)exp[-σ(λ)CL] (1)
in formula (1), I (λ) is the transmitted light intensity of light with wavelength λ emitted by the light source after passing through the gas to be measured, I 0 (λ) is the initial light intensity of light with wavelength λ emitted by the light source, L is the gas optical path, C is the concentration of the gas to be measured, σ (λ) is the absorption cross section of the gas to be measured, related to wavelength λ;
according to the principle that the complex high-temperature flue gas light attenuation of the boiler comprises extinction caused by Rayleigh scattering and Mie scattering, the formula (1) is changed into:
in the formula (2), epsilon R Is the Rayleigh scattering extinction coefficient, epsilon M Is the mie scattering extinction coefficient;
the complex high-temperature flue gas is a mixture of a plurality of gas components according to the boiler, the light emitted by the light source is absorbed by the gas components, and other gas molecules and particle influences exist at the same time, so that the formula (2) is expressed as follows:
in formula (3), σ i And C i Respectively representing the absorption cross section and the concentration of the ith gas for light absorption;
according to the fact that the gas absorption comprises two parts of slow change and fast change, the absorption section sigma (lambda) of the gas to be measured is expressed as follows:
σ(λ)＝σ s (λ)+σ f (λ) (4)
in the formula (4), σ s (λ) is the gas absorption slow-change absorption cross-section, σ f (λ) gas absorption fast-changing absorption cross section;
according to the 'slow change' and 'fast change' principle in gas absorption, broadband absorption and narrow-band absorption can be obtained:
the optical thickness OD at different wavelengths λ is defined as:
in the formula (6), the reaction mixture is,
is broadband absorption, i.e., broadband absorption due to broadband absorption and scattering in gas absorption, which can be removed from the optical thickness by analysis;
is due to narrow band absorption, i.e. gas absorption;
considering that the light attenuation of the complex high-temperature flue gas of the boiler is influenced by the temperature, the optical thickness OD is subjected to temperature correction according to the formula (7):
the accurate optical thickness of the complex high-temperature flue gas of the boiler can be calculated through the method.
In this embodiment, the deep learning model is built by the following steps:
1) converting the absorption spectrum data in the absorption spectrum library into a spectrogram by adopting a spectrum two-dimensional image conversion algorithm to obtain a two-dimensional spectrum information matrix, wherein the two-dimensional spectrum information matrix comprises the following steps:
processing the absorption spectrum data according to the following formula:
S＝XX T (8)
in the formula (8), S is a two-dimensional spectrum information matrix, X is a spectrum data column vector, and a typical two-dimensional spectrum information matrix is obtained:
in the formula (9), a i Is one-dimensional spectral data.
2) Constructing a multi-task deep learning model according to the two-dimensional spectral information matrix, comprising the following steps:
2-1) constructing a common layer comprising a plurality of convolution kernels and a maximum pooling layer;
2-2) constructing convolution kernel branches of different tasks to obtain an initial multi-task deep learning model comprising a common layer and the convolution kernel branches of different tasks;
2-3) inputting the two-dimensional spectrum information matrix into the initial multi-task deep learning model, performing repeated iterative training, and confirming principal component factors and optimal weights of the initial multi-task deep learning model to obtain a target multi-task deep learning model;
wherein each convolution kernel branch corresponds to different temperatures, different main gas components and different concentration parameters.
In this embodiment, the constructing a common layer including a plurality of convolution kernels and a max-pooling layer includes:
the convolutional neural network of the formula (10) is adopted for construction:
in the formula (10), f (m) is a typical two-dimensional spectrum information matrix constructed by the formula (9), n is the length of the signal f (n), and S (n) is a convolution result sequence with the length of len (f (n)) + len (g (n)) -1.
Constructing a plurality of shared convolution layers through the steps, setting a common layer comprising a series of convolution kernels and a maximum pooling layer, extracting spectrogram data from the common layer, and directing the information to a plurality of different branches, wherein each branch corresponds to a temperature parameter, a main gas component parameter and a concentration parameter, so as to form a multitask network.
In this embodiment, a multitask network is constructed as shown in fig. 2, where the multitask network includes a convolution kernel a, a pooling layer a, a convolution kernel B, a pooling layer B, a convolution kernel C, a pooling layer C, a convolution kernel D and different branches as common layers, and the absorption spectrum data of the boiler flue gas to be measured is first input into the convolution kernel a, and then is processed by the convolution kernel a, the pooling layer a, the convolution kernel B, the pooling layer B, the convolution kernel C, the pooling layer C and the convolution kernel D in sequence, and then is transmitted to different branches. FIG. 2 is another embodiment of a multitasking network constructed in accordance with the present application, showing at least 8 typical branches, such as H 2 S different concentrations, temperature spectrogram branches, and H can be obtained by inversion according to the branches 2 The concentration of S; and as SO 2 The spectrogram branches with different concentrations and temperatures can be inverted to obtain SO according to the branches 2 Concentration; for example, the spectrogram branches of NO with different concentrations and temperatures, and the NO concentration can be obtained through inversion according to the branches; also as NO 2 The spectrogram branches with different concentrations and temperatures can be inverted to obtain NO 2 Concentration; e.g. H 2 O different concentration, temperature spectrogram branch, and H can be obtained by inversion according to the branch 2 The concentration of O; e.g. O 2 Different concentrationsThe spectrogram branch of temperature and temperature can be inverted to obtain O 2 Concentration; for example, the spectrogram branches of CO with different concentrations and temperatures, and the CO concentration can be obtained by inversion according to the branches; for example CO 2 The spectral diagram branches with different concentrations and temperatures can be inverted to obtain CO 2 And (4) concentration.
In the above embodiments, the measurement of the composition and concentration of the complex flue gas of the boiler is only performed, and the inversion method of the concentration of any other gas or gases can also be used.
FIG. 3 is a block diagram of a system for inverting the complex high-temperature flue gas components and concentrations of a boiler based on absorption spectra according to an embodiment of the invention. As shown in fig. 3, the system includes:
the data acquisition module is used for acquiring the smoke absorption spectrum data of the boiler to be detected;
the inversion module is used for inputting the boiler flue gas absorption spectrum data to be detected into the multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
The embodiment of the invention also provides a machine-readable storage medium, wherein the machine-readable storage medium is stored with instructions, and the instructions are used for enabling a machine to execute the absorption spectrum-based boiler complex high-temperature flue gas composition and concentration inversion method.
Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
While the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solution of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications are within the scope of the embodiments of the present invention. It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, the embodiments of the present invention will not be described separately for the various possible combinations.
In addition, any combination of various embodiments of the present invention may be made, and the same should be considered as what is disclosed in the embodiments of the present invention as long as it does not depart from the spirit of the embodiments of the present invention.
Claims (10)
1. An inversion method of components and concentration of complex high-temperature flue gas of a boiler based on absorption spectrum is characterized by comprising the following steps:
acquiring smoke absorption spectrum data of a boiler to be detected;
inputting the boiler flue gas absorption spectrum data to be detected into a multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
2. The method for inverting the components and the concentrations of the complex high-temperature flue gas of the boiler based on the absorption spectrum as claimed in claim 1, wherein the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations are obtained by the following steps:
calculating the optical thickness OD of the complex boiler flue gas with different temperatures and different concentrations according to the absorption spectrum principle by taking the temperature of the complex high-temperature flue gas of the boiler, main gas components and concentration parameters as independent variables;
drawing according to the optical thicknesses OD of different wave bands of the complex boiler flue gas with the same temperature and the same concentration to obtain the absorption spectrum data of the complex boiler flue gas with the same temperature and the same concentration;
and changing the temperature, main gas components and concentration parameters of the complex high-temperature flue gas of the boiler, and drawing to obtain the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
3. The method for inverting the composition and concentration of the boiler complex high-temperature flue gas based on the absorption spectrum as recited in claim 2, wherein the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations are stored in an absorption spectrum library.
4. The method for inverting the complex high-temperature flue gas components and the concentration of the boiler based on the absorption spectrum as claimed in claim 2, wherein the principle of the absorption spectrum is as follows:
for a single gas, under the irradiation of light with different wavelengths, the incident light and the intensity of the transmitted light meet the following relationship:
I(λ)＝I 0 (λ)exp[-σ(λ)CL] (1)
in formula (1), I (λ) is the transmitted light intensity of light with wavelength λ emitted by the light source after passing through the gas to be measured, I 0 (λ) is the initial light intensity of light with wavelength λ emitted by the light source, L is the gas optical path, C is the concentration of the gas to be measured, σ (λ) is the absorption cross section of the gas to be measured, related to wavelength λ;
according to the principle that the complex high-temperature flue gas light attenuation of the boiler comprises extinction caused by Rayleigh scattering and Mie scattering, the formula (1) is changed into:
in the formula (2), epsilon R Is the Rayleigh scattering extinction coefficient, epsilon M Is the mie scattering extinction coefficient;
according to the fact that various gas components of the complex high-temperature flue gas of the boiler absorb light emitted by a light source, the influence of other gas molecules and particles is considered, and the formula (2) is expressed as follows:
in formula (3), σ i And C i Respectively showing the absorption cross section and concentration of the ith gas for light absorption;
according to the "slow change" and "fast change" principles in gas absorption, the absorption cross section σ (λ) of the gas to be measured is expressed as:
σ(λ)＝σ s (λ)+σ f (λ) (4)
in the formula (4), σ s (λ) is the gas absorption slow-change absorption cross-section, σ f (λ) gas absorption fast-changing absorption cross section;
according to the 'slow change' and 'fast change' principle in gas absorption, broadband absorption and narrow-band absorption are obtained:
the optical thickness OD at different wavelengths λ is defined as:
in the formula (6), the reaction mixture is,
is broadband absorption;
is a narrow band absorption;
considering that the light attenuation of the complex high-temperature flue gas of the boiler is influenced by the temperature, the optical thickness OD is subjected to temperature correction according to the formula (7):
5. the absorption spectrum-based boiler complex high-temperature flue gas composition and concentration inversion method according to claim 3, wherein the multitask deep learning model is established through the following steps:
converting the absorption spectrum data in the absorption spectrum library into a spectrogram by adopting a spectrum two-dimensional image conversion algorithm to obtain a two-dimensional spectrum information matrix;
and constructing a multi-task deep learning model according to the two-dimensional spectrum information matrix.
6. The method for inverting the complex high-temperature flue gas components and the concentration of the boiler based on the absorption spectrum according to claim 5, wherein the two-dimensional spectrum information matrix is obtained by converting the absorption spectrum data in the absorption spectrum library into a spectrogram by using a spectrum two-dimensional image conversion algorithm, and comprises the following steps:
processing the absorption spectrum data according to formula (8):
S＝XX T (8)
in the formula (8), S is a two-dimensional spectrum information matrix, X is a spectrum data column vector, and a typical two-dimensional spectrum information matrix is obtained:
in the formula (9), a i Is one-dimensional spectral data.
7. The absorption spectrum-based boiler complex high-temperature flue gas composition and concentration inversion method according to claim 6, wherein the constructing of the multitask deep learning model according to the two-dimensional spectrum information matrix comprises the following steps:
constructing a common layer comprising a plurality of convolution kernels and a maximum pooling layer;
constructing convolution kernel branches of different tasks to obtain an initial multi-task deep learning model comprising a common layer and the convolution kernel branches of the different tasks;
inputting the two-dimensional spectrum information matrix into the initial multi-task deep learning model, performing repeated iterative training, and confirming principal component factors and optimal weights of the initial multi-task deep learning model to obtain a multi-task deep learning model;
each convolution kernel branch corresponds to different temperatures, different main gas components and different concentration parameters.
8. The method for inverting the complex high-temperature flue gas composition and concentration of the boiler based on the absorption spectrum as claimed in claim 7, wherein the constructing a common layer comprising a plurality of convolution kernels and a maximum pooling layer comprises:
the convolutional neural network of the formula (10) is adopted for construction:
in the formula (10), f (m) is a typical two-dimensional spectrum information matrix constructed by the formula (9), n is the length of the signal f (n), and S (n) is a convolution result sequence with the length of len (f (n)) + len (g (n)) -1.
9. A boiler complex high-temperature flue gas component and concentration inversion system based on absorption spectrum is characterized by comprising:
the data acquisition module is used for acquiring the smoke absorption spectrum data of the boiler to be detected;
the inversion module is used for inputting the boiler flue gas absorption spectrum data to be detected into the multi-task deep learning model to obtain boiler flue gas components and concentration parameters;
the multitask deep learning model is established according to the absorption spectrum data of the complex boiler flue gas with different temperatures and different concentrations.
10. A machine-readable storage medium having instructions stored thereon for causing a machine to perform the absorption spectroscopy-based boiler complex high temperature flue gas composition and concentration inversion method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210178859.6A CN115015134A (en) | 2022-02-25 | 2022-02-25 | Boiler complex high-temperature flue gas component and concentration inversion method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210178859.6A CN115015134A (en) | 2022-02-25 | 2022-02-25 | Boiler complex high-temperature flue gas component and concentration inversion method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115015134A true CN115015134A (en) | 2022-09-06 |
Family
ID=83067801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210178859.6A Pending CN115015134A (en) | 2022-02-25 | 2022-02-25 | Boiler complex high-temperature flue gas component and concentration inversion method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115015134A (en) |
-
2022
- 2022-02-25 CN CN202210178859.6A patent/CN115015134A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111239071B (en) | Method for detecting concentration of nitrate in seawater by spectrometry | |
| CN110161013B (en) | Laser-induced breakdown spectroscopy data processing method and system based on machine learning | |
| CN102539377B (en) | Intermediate infrared absorption spectra based method for multi-component mixed gas qualitative and quantitative analysis | |
| KR102662656B1 (en) | Gas analysis methods and devices | |
| Dai et al. | Quantitative determination of Al–Cu–Mg–Fe–Ni aluminum alloy using laser-induced breakdown spectroscopy combined with LASSO–LSSVM regression | |
| Wang et al. | Rapid quantitative analysis of the acidity of iron ore by the laser-induced breakdown spectroscopy (LIBS) technique coupled with variable importance measures-random forests (VIM-RF) | |
| JPH1137935A (en) | Method and apparatus for determining a plurality of components in sewage | |
| CN109142251B (en) | LIBS quantitative analysis method of random forest auxiliary artificial neural network | |
| Cai et al. | Estimating the aging grade of heat-resistant steel by using portable laser-induced breakdown spectroscopy | |
| CN115015134A (en) | Boiler complex high-temperature flue gas component and concentration inversion method and system | |
| Burikov et al. | New approaches to determination of temperature and salinity of seawater by laser Raman spectroscopy | |
| CN107941758B (en) | Coal quality heat value measuring method based on spectrum pretreatment | |
| CN116399836A (en) | Cross-talk fluorescence spectrum decomposition method based on alternating gradient descent algorithm | |
| CN113723471B (en) | Nanoparticle concentration and particle size estimation method and device | |
| CN113435115B (en) | Fluorescence spectrum characteristic wavelength screening method, fluorescence spectrum characteristic wavelength screening device, computer equipment and readable storage medium | |
| Xie et al. | Quantitative analysis of steel samples by laser-induced-breakdown spectroscopy with wavelet-packet-based relevance vector machines | |
| Zhu et al. | Determination of ash content, volatile matter, and calorific value in coal by OLS combined with laser-induced breakdown spectroscopy based on PC recombination | |
| JP4211172B2 (en) | Method for calculating the concentration of SO3 gas in the flue | |
| JP4211173B2 (en) | Method for calculating the concentration of SO3 gas in the flue | |
| CN117388204B (en) | Nitric oxide gas analysis system, method and computer readable storage medium | |
| CN117233120B (en) | Polluted gas detection method and system based on machine learning and hyperspectral image | |
| CN117805024B (en) | Crisp pear saccharin degree detection method and device, cloud equipment and computer device | |
| JP7331058B2 (en) | Measuring device, measuring method and measuring device control program | |
| Yang et al. | Detection of Pb Content in Water Using Graphite Enrichment Combined with Laser-Induced Breakdown Spectroscopy | |
| KR20230049962A (en) | Method for estimating chlorine concentration in contrete structures using artificial intelligence |
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 |