Method and device for predicting full-wave-band thermal radiation energy in furnace

Abstract

The invention is suitable for the technical field of boiler detection, and provides a method and a device for predicting full-wave-band thermal radiation energy in a boiler. The method comprises the following steps: inputting obtained target visible light thermal radiation energy of each preset measuring point in the boiler into a trained prediction model to obtain target full-waveband thermal radiation energy corresponding to each preset measuring point; wherein the trained prediction model is obtained by training based on visible light thermal radiation energy and full-band thermal radiation energy of each preset measurement point under different working conditions; the full wave band comprises a visible light wave band to an infrared wave band; the visible light thermal radiation energy and the full-wave-band thermal radiation energy are determined according to the hearth temperature of the boiler and the particulate matter concentration and the gas concentration in the boiler. The corresponding target full-band thermal radiation energy can be directly and accurately obtained, namely, online accurate prediction of the thermal radiation energy in the furnace is achieved, meanwhile, an infrared radiation energy sensor does not need to be additionally arranged, cost is low, and implementation is easy.

Description

Technical Field

[0001] The present application relates to the technical field of boiler detection, and in particular to a method and device for predicting full-band thermal radiation energy in a furnace. Background Art

[0002] The traditional method of variable load regulation of existing coal-fired power generating units is to change the main steam flow rate and then adjust the output power of the unit. However, the delay and thermal inertia of processes such as boiler heat and mass transfer limit the regulation of unit output power. Therefore, it is necessary to quickly characterize the furnace signal of boiler combustion and heat release, and perform variable load regulation of the unit based on the furnace signal characterizing boiler combustion and heat release.

[0003] In actual applications, the combustion media contained in coal-fired boilers include solid particles such as soot, coke, and fly ash, as well as triatomic gas media such as carbon dioxide (CO2) and water vapor. Solid particles emit a continuous blackbody radiation spectrum at high temperatures, mainly in the visible light-near infrared band, while the thermal radiation emitted by triatomic gases such as CO2 is located in the infrared band and consists of multiple separate spectral bands or spectral lines. In other words, the thermal radiation energy generated by combustion in the furnace in the full range of visible light-infrared can be used as a furnace signal to characterize the heat release of boiler combustion.

[0004] However, the related art usually uses visible light radiation energy sensors and / or near-infrared radiation energy sensors to obtain the thermal radiation energy inside the furnace, which cannot fully represent the thermal radiation energy emitted by the combustion in the furnace in the visible light-infrared full band (i.e., full band), resulting in inaccurate thermal radiation energy inside the furnace. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a method and device for predicting the full-band thermal radiation energy in a furnace to solve the technical problem in the related art that the full-band thermal radiation energy in the furnace cannot be accurately obtained.

[0006] In a first aspect, an embodiment of the present application provides a method for predicting full-band thermal radiation energy in a furnace, comprising:

[0007] Obtain the target visible light thermal radiation energy of each preset measuring point in the boiler furnace;

[0008] The target visible light thermal radiation energy of each preset measuring point is input into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point;

[0009] Among them, the trained prediction model is obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions; the full band includes the visible light band to the infrared band;

[0010] The visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

[0011] In a possible implementation, before inputting the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy, the method further includes:

[0012] Obtaining the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler;

[0013] For each preset measuring point under each working condition, a radiation characteristic parameter is obtained according to the particle concentration and the gas concentration corresponding to the preset measuring point under the working condition; wherein the radiation characteristic parameter includes an absorption coefficient, a scattering coefficient and an attenuation coefficient; based on the furnace temperature and the radiation characteristic parameter corresponding to the preset measuring point under the working condition, the visible light thermal radiation energy and the full-band thermal radiation energy are obtained;

[0014] The visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions constitute a training set, and the preset prediction model is trained based on the training set to obtain a trained prediction model.

[0015] In a possible implementation, obtaining the furnace temperature, particulate matter concentration, and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler includes:

[0016] According to the design parameters and operating parameters of the boiler, different working conditions are divided;

[0017] For each operating condition, a combustion numerical simulation is performed according to the design parameters, the operating parameters, and the combustion mode of the boiler under the operating condition to obtain the overall furnace temperature, the overall particulate matter concentration, and the overall gas concentration under the operating condition;

[0018] Based on the position of each preset measuring point, as well as the overall furnace temperature, overall particle concentration and overall gas concentration under the operating condition, the furnace temperature, particle concentration and gas concentration corresponding to each preset measuring point under the operating condition are determined.

[0019] In a possible implementation, for each preset measuring point under each working condition, obtaining the visible light thermal radiation energy and the full-band thermal radiation energy based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition includes:

[0020] The furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition are input into a preset spectral imaging model to obtain spectral radiation energy; wherein the spectral imaging model inputs the furnace temperature and radiation characteristic parameters to output spectral radiation energy;

[0021] Integrating the spectral radiation energy along the wavelength in the visible light band to obtain visible light thermal radiation energy;

[0022] The spectral radiation energy is fully integrated along the wavelength to obtain the full-band thermal radiation energy.

[0023] In a possible implementation, the preset measuring points are arranged on the side wall of the boiler furnace, and the space formed by the positions of all the preset measuring points covers the combustion and heat release high temperature zone of the boiler; the combustion and heat release high temperature zone is an area where the heat is higher than the preset heat threshold.

[0024] In a possible implementation, for each preset measuring point under each working condition, obtaining the radiation characteristic parameter according to the particle concentration and the gas concentration corresponding to the preset measuring point under the working condition includes:

[0025] According to the particle concentration corresponding to the preset measuring point under the working condition, the absorption coefficient, scattering coefficient and attenuation coefficient of the particles are obtained;

[0026] The gas absorption coefficient is obtained according to the gas concentration corresponding to the preset measuring point under the working condition.

[0027] In a possible implementation, the preset prediction model includes a hidden layer;

[0028] The step of training a preset prediction model based on the training set to obtain a trained prediction model includes:

[0029] Input the visible light thermal radiation energy of a preset measuring point under a working condition in the training set into the preset prediction model to obtain the corresponding full-band thermal radiation energy prediction value;

[0030] Determine the loss value of the loss function according to the difference between the predicted value of the full-band thermal radiation energy and the corresponding full-band thermal radiation energy, and judge whether the loss value is less than a preset loss value;

[0031] If the loss value is less than the preset loss value, the training is completed and a trained prediction model is obtained;

[0032] If the loss value is greater than or equal to the preset loss value, the weights of all neurons in the hidden layer are updated, and the visible light thermal radiation energy of a preset measuring point under the next working condition in the training set is re-input into the preset prediction model until the training is completed.

[0033] In a possible implementation, a first loop line and a second loop line are arranged on the side wall of the furnace of the boiler, and the space formed by the first loop line and the second loop line covers the combustion heat release high temperature zone of the boiler;

[0034] Four first preset measuring points are arranged on the first loop line, and the four first preset measuring points are arranged on four side walls of the furnace respectively;

[0035] Four second preset measuring points are arranged on the second loop line, and the four second preset measuring points are arranged on four side walls of the furnace respectively.

[0036] In a second aspect, the embodiment of the present application provides a device for predicting full-band thermal radiation energy in a furnace, comprising:

[0037] The first acquisition module is used to acquire the target visible light thermal radiation energy of each preset measuring point in the boiler furnace.

[0038] The first obtaining module is used to input the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point.

[0039] Among them, the trained prediction model is obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions; the full-band includes visible light band to infrared band.

[0040] The visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

[0041] In a possible implementation manner, the device further includes:

[0042] The second acquisition module is used to obtain the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler.

[0043] The second obtaining module is used to obtain radiation characteristic parameters for each preset measuring point under each operating condition according to the particle concentration and the gas concentration corresponding to the preset measuring point under the operating condition; wherein the radiation characteristic parameters include absorption coefficient, scattering coefficient and attenuation coefficient; based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the operating condition, obtain visible light thermal radiation energy and full-band thermal radiation energy.

[0044] The third obtaining module is used to form a training set with the visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions, and train the preset prediction model based on the training set to obtain a trained prediction model.

[0045] It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0046] The method and device for predicting the full-band thermal radiation energy in a furnace provided in the embodiment of the present application input the obtained target visible light thermal radiation energy of each preset measuring point in the boiler furnace into the trained prediction model, thereby obtaining the target full-band thermal radiation energy corresponding to each preset measuring point, wherein the trained prediction model is obtained based on the visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions, and the visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, and the concentration of particulate matter and the concentration of gas in the boiler, so that when training the prediction model, it is not necessary to set up a variety of sensors to detect the thermal radiation energy to obtain the training set. The embodiment of the present application is based on the obtained target visible light thermal radiation energy, and the prediction model that has learned the mapping relationship between the visible light thermal radiation energy generated by the combustion in the furnace and the full-band thermal radiation energy, and can directly and accurately obtain the corresponding target full-band thermal radiation energy, that is, it realizes the online and accurate prediction of the thermal radiation energy in the furnace, and at the same time, because no additional infrared radiation energy sensor is required, it is low-cost and easy to implement.

[0047] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0049] Figure 1 It is a schematic diagram of an application scenario provided by an embodiment of the present application;

[0050] Figure 2 It is a flow chart of a method for predicting full-band thermal radiation energy in a furnace provided in an embodiment of the present application;

[0051] Figure 3 It is a schematic diagram of the arrangement of various preset measuring points in a boiler furnace provided in an embodiment of the present application;

[0052] Figure 4 It is a schematic diagram of the structure of a prediction model provided by an embodiment of the present application;

[0053] Figure 5 It is a structural schematic diagram of a device for predicting full-band thermal radiation energy in a furnace provided in one embodiment of the present application. DETAILED DESCRIPTION

[0054] The present application is described more clearly below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the effects of the present application, but are not intended to limit the present application in any form. It should be noted that, for those of ordinary skill in the art, several variations and improvements may be made without departing from the concept of the present application. These all fall within the scope of protection of the present application.

[0055] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0056] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0057] In the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0058] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0059] In addition, the “plurality” mentioned in the embodiments of the present application should be interpreted as two or more.

[0060] With the idea of ​​accurately predicting the thermal radiation energy in the furnace (i.e., full-band thermal radiation energy), the inventors have found that the acquired target visible light thermal radiation energy can be input into a prediction model that has learned the mapping relationship between the visible light thermal radiation energy generated by the combustion in the furnace and the full-band thermal radiation energy to accurately obtain the corresponding target full-band thermal radiation energy, that is, to achieve online and accurate prediction of the thermal radiation energy in the furnace, and no additional infrared radiation energy sensor is required, which is low-cost and easy to implement. In addition, the visible light thermal radiation energy and full-band thermal radiation energy used to train the prediction model are determined based on the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler. In this way, when training the prediction model, there is no need to set up multiple sensors to detect thermal radiation energy to obtain a training set.

[0061] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0062] First reference Figure 1 , Figure 1 The application scenario of the present application is schematically illustrated. In the application scenario, an electronic device and a boiler are included. Various preset measuring points are arranged in the boiler, and a trained prediction model is arranged on the electronic device.

[0063] The electronic device obtains the target visible light thermal radiation energy of each preset measuring point in the boiler furnace, and inputs the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point.

[0064] The trained prediction model is based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions. The full band includes visible light band to infrared band. The visible light thermal radiation energy and full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

[0065] The application scenario may also include a visible light radiation energy sensor, which is disposed in the boiler furnace and is in communication connection with the electronic device. The visible light radiation energy sensor detects the target visible light thermal radiation energy at each preset measuring point in the boiler, so that the electronic device obtains the above target visible light thermal radiation energy.

[0066] Combine the following Figure 1 ,refer to Figure 2-Figure 4 To describe the full-band thermal radiation energy prediction method in a furnace provided according to an exemplary embodiment of the present application.

[0067] Figure 2 FIG. 1 is a flow chart of a method for predicting full-band thermal radiation energy in a furnace provided in an embodiment of the present application. Figure 2 As shown, the method in the embodiment of the present application may include:

[0068] Step 201: Obtain target visible light thermal radiation energy of each preset measuring point in the boiler furnace.

[0069] Optionally, in this embodiment, the target visible light thermal radiation energy of each preset measuring point is obtained by a visible light radiation energy sensor disposed in the boiler furnace.

[0070] Exemplarily, each preset measuring point can be set on the side wall of the boiler furnace, and the space formed by the positions of all the preset measuring points covers the high temperature area of ​​combustion heat release of the boiler. Figure 3 The combustion heat release high temperature area is the area where the heat is higher than the preset heat threshold, mainly the area of ​​the boiler combustion flame. The preset measuring points are set in the combustion heat release high temperature area, so that the calculated full-band thermal radiation energy of each preset measuring point can more accurately characterize the boiler combustion heat release.

[0071] Optional, see Figure 3 A first loop line and a second loop line are arranged on the side walls of the boiler furnace, and the space formed by the first loop line and the second loop line covers the high-temperature zone of combustion and heat release of the boiler. Four first preset measuring points are arranged on the first loop line, and the four first preset measuring points are respectively arranged on the four side walls of the furnace. Four second preset measuring points are arranged on the second loop line, and the four second preset measuring points are respectively arranged on the four side walls of the furnace.

[0072] In this way, the full-band thermal radiation energy corresponding to the eight preset measuring points arranged on the four side walls of the furnace and located on the first loop line or the second loop line can more accurately characterize the heat release of boiler combustion.

[0073] For example, taking the boiler furnace whose length, width and height are 14 meters, 14 meters and 34 meters respectively, it is determined that the height of the boiler combustion heat release high temperature zone is between 8 meters and 17 meters of the furnace, then the first loop is set at the 8-meter height of the furnace, and the second loop is set at the 17-meter height. Thus, four first preset measuring points are evenly set on the first loop, that is, the four first preset measuring points are respectively set on the four side walls of the furnace, and four second preset measuring points are evenly set on the second loop, that is, the four second preset measuring points are respectively set on the four side walls of the furnace.

[0074] Step 202: Input the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point.

[0075] The trained prediction model is based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions. The full band includes visible light band to infrared band. The visible light thermal radiation energy and full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

[0076] In this embodiment, the trained prediction model learns the mapping relationship between visible light thermal radiation energy and full-band thermal radiation energy, and it inputs visible light thermal radiation energy and outputs the corresponding full-band thermal radiation energy. The target visible light thermal radiation energy corresponding to each preset measuring point obtained by the visible light radiation energy sensor is input into the trained prediction model, and the target full-band thermal radiation energy corresponding to each preset measuring point that can well characterize the heat release of boiler combustion can be obtained, so that the variable load regulation of the boiler can be performed based on the above-mentioned target full-band thermal radiation energy, so as to realize the regulation and control of boiler combustion.

[0077] Here, it is only necessary to set up a visible light radiation energy sensor for obtaining the target visible light thermal radiation energy corresponding to each preset measuring point, and use the trained prediction model to learn to obtain the corresponding full-band thermal radiation energy. That is, there is no need to set up an additional infrared radiation energy sensor, and accurate online prediction of the thermal radiation energy in the furnace can be achieved, which saves costs and is easy to implement.

[0078] In addition, the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions used to train the prediction model are determined based on the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler. That is, when constructing the training set, there is no need to set up multiple sensors to detect the corresponding thermal radiation energy. Instead, the furnace temperature of the boiler obtained by combustion numerical simulation, as well as the particle concentration and gas concentration in the boiler are used to obtain the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions to form the training set. By setting different working conditions and different preset measuring points, the visible light thermal radiation energy and full-band thermal radiation energy under different circumstances can be obtained, so that the prediction model can be trained using the visible light thermal radiation energy and full-band thermal radiation energy of different preset measuring points under different working conditions, and a well-trained prediction model with good performance can be obtained.

[0079] The above embodiment describes inputting the target visible light thermal radiation energy of each preset measuring point into a trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point. The following embodiment describes in detail how to obtain the trained prediction model.

[0080] In a possible implementation manner, before the target visible light thermal radiation energy of each preset measuring point is input into the trained prediction model to obtain the target full-band thermal radiation energy, steps A1 to A3 may also be included.

[0081] A1. Obtain the furnace temperature, particle concentration and gas concentration corresponding to each preset measuring point under different boiler operating conditions.

[0082] A2. For each preset measuring point under each working condition, the radiation characteristic parameters are obtained according to the particle concentration and gas concentration corresponding to the preset measuring point under the working condition; based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition, the visible light thermal radiation energy and the full-band thermal radiation energy are obtained.

[0083] A3. The visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions constitute a training set, and the preset prediction model is trained based on the training set to obtain a trained prediction model.

[0084] Among them, the radiation characteristic parameters include absorption coefficient, scattering coefficient and attenuation coefficient. As mentioned above, particulate matter includes soot, coke and fly ash particles, and gases include CO2 and water vapor.

[0085] Exemplarily, when obtaining the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler, this embodiment can divide the different operating conditions according to the design parameters and operating parameters of the boiler. For each operating condition, a combustion numerical simulation is performed according to the design parameters, operating parameters of the boiler under the operating condition, and the combustion mode of the boiler to obtain the overall furnace temperature, overall particulate matter concentration and overall gas concentration under the operating condition. Thereafter, based on the position of each preset measuring point, as well as the overall furnace temperature, overall particulate matter concentration and overall gas concentration under the operating condition, the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under the operating condition are determined.

[0086] Optionally, the boiler design parameters and operating parameters may include combustion load, main combustion air door opening, burnout air door opening, burnout air swing angle, and coal mill operation mode. For example, this embodiment considers three situations for each of the above parameters, and 243 operating conditions can be obtained. At the same time, this embodiment uses the boiler that actually performs full-band thermal radiation energy prediction as a reference for modeling and constructs a geometric model of the boiler furnace. For example, referring to Figure 3 In this embodiment, the length, width and height are 14 meters, 14 meters and 34 meters respectively (that is, the length, width and height of the actual furnace corresponding to the geometric model), and the shape is as follows Figure 3 In addition, in this embodiment, a coal-fired boiler of a 350 MW generator set in a thermal power plant is taken as an example, and the combustion mode of the boiler is determined to be four-corner tangential combustion.

[0087] Thus, for each working condition, according to the design parameters, operating parameters and combustion mode of the boiler under the working condition, this embodiment uses computational fluid dynamics (CFD) technology to perform combustion numerical simulation. Specifically, the geometric model of the constructed furnace can be divided 40 times along the three-dimensional directions of length, width and height to obtain 64,000 grid blocks. After that, the Fluent software can be used to perform combustion numerical simulation to obtain the overall furnace temperature, overall particle concentration and overall gas concentration under the working condition. Among them, CFD is a technology that uses numerical methods and computer simulation to study fluid flow and heat transfer, and the Fluent software is suitable for combustion numerical simulation in various complex situations.

[0088] Here, the overall furnace temperature is the temperature within the furnace area, and the 64,000 grid blocks in the furnace correspond to their respective furnace temperatures. Similarly, the overall particle concentration and the overall gas concentration are the particle concentration and gas concentration within the furnace area, and the 64,000 grid blocks in the furnace correspond to their respective particle concentrations and gas concentrations.

[0089] In this embodiment, the grid blocks corresponding to the preset measuring points can be determined according to the positions of the preset measuring points. For example, for a certain preset measuring point, the grid block closest to the preset measuring point is used as the grid block corresponding to the preset measuring point. Thus, the furnace temperature, particulate matter concentration and gas concentration corresponding to the grid block corresponding to the preset measuring point can be used as the furnace temperature, particulate matter concentration and gas concentration corresponding to the preset measuring point under the working condition.

[0090] In some embodiments, for each preset measuring point under each operating condition, when obtaining the radiation characteristic parameters, the absorption coefficient, scattering coefficient and attenuation coefficient of the particulate matter can be obtained based on the particulate matter concentration corresponding to the preset measuring point under the operating condition, and the absorption coefficient of the gas can be obtained based on the gas concentration corresponding to the preset measuring point under the operating condition.

[0091] There is usually a dispersion system composed of multiple particles in the flame, which is also called a particle system or a dispersion medium. In this embodiment, based on the particle concentration, the Mie scattering theory is used to obtain the radiation characteristic parameters of the particle system, that is, the absorption coefficient, scattering coefficient and attenuation coefficient of the particle, which are calculated as follows:

[0092]

[0093] κ=β-σ s

[0094] In the formula, β, σ s and κ are the attenuation coefficient, scattering coefficient and absorption coefficient of particles, respectively. is the volume percentage of particle i, Di is the particle size of particle i, Ni is the particle number density of particle i, which can be expressed by the formula Where Ci is the particle concentration of particle i, V is the gas volume of particle i, and N A is the Avogadro constant, Qe,i and Qs,i are the attenuation factor and scattering factor of particle i respectively, and the expressions are as follows:

[0095]

[0096] Where an and bn are Mie scattering coefficients, x is a dimensionless parameter (x = πD / λ) determined by the particle size D and the incident light wavelength λ, n is the refractive index, and Re is the value in the above expression. It should be noted that Q e and Q s They are the attenuation factor and scattering factor of the particle respectively. Particle i is not specified so there is no subscript i.

[0097] It should be noted that since there are many formulas and parameters involved in this embodiment, the same letters may represent different parameters in different steps or parts. It is explained here that in the parts of the aforementioned embodiments for obtaining the absorption coefficient, scattering coefficient and attenuation coefficient of particulate matter, the same letters represent the same parameters, and different letters represent different parameters. In the parts for obtaining the absorption coefficient of the gas in the following embodiments, the same letters represent the same parameters, and different letters represent different parameters. And in the parts for obtaining the visible light thermal radiation energy and the full-band thermal radiation energy in the following embodiments, the same letters represent the same parameters, and different letters represent different parameters. Please note the distinction based on the annotations of the letters in each step or part.

[0098] Optionally, this embodiment uses a gas radiation model, such as a line-by-line (LBL) model, based on the gas concentration to obtain the gas absorption coefficient κ η , calculated as follows:

[0099]

[0100] Where N is the number density of gas molecules, that is, the gas concentration, and F j (η) is the spectral line shape of the jth spectral line, S j,η (T) is the line intensity of the jth spectral line at temperature T, and η is the wave number.

[0101] According to Lorenz theory, F j (η) can be expressed as:

[0102]

[0103] In the formula, γ j is the half width of the jth spectral line (HWHM, also known as the Lorentz half width), η 0j is the wave number of the jth spectral line in the database, and the Lorentz half-width γ can be calculated by the following formula:

[0104]

[0105] Where T0 is the database reference temperature. Here, the database can be HITRAN, HITEMP and CDSD-1000 databases. For HITRAN and HITEMP databases, T0 = 296K; for CDSD-1000 database, T0 = 1000K. n, P, P s , γ air and γ self They are temperature correlation coefficient, overall atmospheric pressure, gas partial pressure, air broadening and gas self broadening. All of the above parameters can be obtained by querying the corresponding database.

[0106] At temperature T, the line intensity Sj, η (T) can be extrapolated from the reference state line strength Sj(T0):

[0107]

[0108] In the formula, Q v () and Q r () are vibration partition function and rotation partition function respectively, Q v0 and Q r0 are the vibrational partition function and rotational partition function values ​​at T = T0, h is Planck's constant, E" is the low-state energy level of the absorption line, c is the speed of light in vacuum, η0 is the wave number of the spectral line in the database, k B is the Boltzmann constant.

[0109] In this way, the absorption coefficient, scattering coefficient and attenuation coefficient of the particles corresponding to each preset measuring point under each working condition, as well as the absorption coefficient of the gas can be obtained, which can be used to subsequently obtain the visible light thermal radiation energy and full-band thermal radiation energy.

[0110] In some embodiments, for each preset measuring point under each operating condition, when obtaining visible light thermal radiation energy and full-band thermal radiation energy, the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the operating condition can be input into a preset spectral imaging model to obtain spectral radiation energy. Thereafter, the spectral radiation energy is integrated along the wavelength in the visible light band to obtain visible light thermal radiation energy, and the spectral radiation energy is integrated along the wavelength in the full band to obtain full-band thermal radiation energy.

[0111] The above-mentioned spectral imaging model inputs the furnace temperature and radiation characteristic parameters and outputs the spectral radiation energy. In this embodiment, the combustion medium in the furnace is assumed to be a gray medium that emits, absorbs and scatters, and the furnace wall is assumed to be a gray wall that emits, absorbs and scatters. For each preset measuring point under each working condition, the corresponding furnace temperature and radiation characteristic parameters are input into the preset spectral imaging model, and the spectral radiation energy can be obtained. The preset spectral imaging model is obtained based on the radiation transfer equation, and the radiation transfer equation is:

[0112]

[0113] Eλ=πIλ

[0114] In the formula, I λ is the radiation intensity, E λ is the spectral radiation energy, β, κ a and σ are the attenuation coefficient, absorption coefficient and scattering coefficient of the medium, ε is the wall emissivity, T(l) represents the temperature of the medium in the furnace located on the s-direction line passing through point O, where the medium in the furnace refers to the particulate matter and gas mentioned in the aforementioned embodiment, and the furnace boundary point O and the s-direction line can be calibrated manually. T(v) represents the temperature of all media located in the volume space area of ​​the furnace, including T(l). T(w) represents the temperature of the wall point that intersects the s-direction line passing through point O, and T(w') represents the temperature of all wall areas, including T(w). l and lw are the lengths of the reflection and scattering paths, respectively, and are related to the arrangement of visible light radiation energy sensors such as charge coupled devices (CCDs). and The distribution of radiation scattering shares between space and space units, space and wall units, and wall and wall units are given respectively, which can be obtained after the discrete grid division of the given furnace structure size, space and wall.

[0115] In this way, after obtaining the spectral radiation energy, the spectral radiation energy is integrated along the wavelength direction of the visible light band (380 nanometers to 780 nanometers) and the full band (380 nanometers to 14000 nanometers), respectively, to obtain the visible light thermal radiation energy and the full-band thermal radiation energy, respectively.

[0116]

[0117] In the formula, E VIS is the visible light thermal radiation energy, E VIS-IR is the full-band thermal radiation energy, and λ is the wavelength.

[0118] In this way, the visible light thermal radiation energy and full-band thermal radiation energy corresponding to each preset measuring point under each working condition can be obtained, and the visible light thermal radiation energy and full-band thermal radiation energy corresponding to each preset measuring point under each working condition constitute a data set, and the preset prediction model is subsequently trained.

[0119] In some embodiments, the preset prediction model includes a hidden layer. When the preset prediction model is trained based on the training set to obtain a trained prediction model, the visible light thermal radiation energy of a preset measuring point under a working condition in the training set can be input into the preset prediction model to obtain the corresponding full-band thermal radiation energy prediction value, and based on the difference between the full-band thermal radiation energy prediction value and the full-band thermal radiation energy, the loss value of the loss function is determined to determine whether the loss value is less than the preset loss value. If the loss value is less than the preset loss value, the training is completed to obtain a trained prediction model. If the loss value is greater than or equal to the preset loss value, the weights of all neurons in the hidden layer are updated, and the visible light thermal radiation energy of a preset measuring point under the next working condition in the training set is re-input into the preset prediction model until the training is completed.

[0120] In this embodiment, the constructed data set includes multiple groups of data, a part of the data can be extracted as a training set, and the remaining data can be extracted as a test set. The data in the training set and the test set can be standardized. Figure 4 The schematic diagram of the prediction model is shown in Figure 4 The preset prediction model can be a multilayer perceptron (MLP) neural network model, including an input layer, a hidden layer and an output layer. The activation function of the MLP neural network model can be a Leaky Relu function, the loss function can be a mean square error (MSE), and an L2 regularization algorithm is added to prevent the model from overfitting. The optimization algorithm can be an adaptive moment estimation (Adam) algorithm with an adaptive learning rate.

[0121] The visible light thermal radiation energy of a preset measuring point under a certain working condition in the training set is input into the input layer of the preset prediction model. After the nonlinear calculation of the 6 hidden layers, the corresponding full-band thermal radiation energy prediction value is output. Then, the loss value of the loss function is determined according to the difference between the full-band thermal radiation energy prediction value and the full-band thermal radiation energy corresponding to the preset measuring point under the working condition. When the loss value is less than the preset loss value, the training is completed to obtain a trained prediction model. When the loss value is greater than or equal to the preset loss value, the Adam algorithm is used to update the weights of all neurons in the hidden layer, and the visible light thermal radiation energy of a preset measuring point under the next working condition in the training set is re-input into the preset prediction model until the training is completed. For example, after more than 1,000 iterations of training, the loss value of the MLP neural network model is reduced to 0.06122, which is less than the preset loss value, and the training is completed to obtain a trained prediction model.

[0122] Optionally, after the preset prediction model is trained using the training set to obtain the trained prediction model, the visible light thermal radiation energy corresponding to each prediction measurement point is quantitatively related to the full-band thermal radiation energy. At this time, the test set can be used to test the generalization ability and prediction accuracy of the trained prediction model. The visible light thermal radiation energy of each set of data in the test set is input into the trained prediction model, and the predicted full-band thermal radiation energy can be output. The predicted full-band thermal radiation energy is compared with the corresponding full-band thermal radiation energy, and finally the average relative error of the predicted full-band thermal radiation energy output by the trained prediction model is calculated to be 3.924%, indicating that the trained prediction model has high accuracy.

[0123] In this way, a well-trained and highly accurate prediction model can be obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions.

[0124] In this embodiment, the prediction model is trained by utilizing the intrinsic connection between the visible light thermal radiation energy and the full-band thermal radiation energy generated by the simulated combustion in the furnace, so that the prediction model learns the mapping relationship between the visible light thermal radiation energy and the full-band thermal radiation energy, and can input the visible light thermal radiation energy and accurately output the full-band thermal radiation energy. In this way, when subsequently predicting the thermal radiation energy in the furnace online, the target visible light thermal radiation energy of each preset measuring point can be input into the trained prediction model to accurately obtain the target full-band thermal radiation energy corresponding to each preset measuring point.

[0125] In addition, the visible light thermal radiation energy and full-band thermal radiation energy used to train the prediction model are determined based on the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler. In this way, when training the prediction model, there is no need to set up multiple sensors to detect thermal radiation energy to obtain a training set.

[0126] The method for predicting the full-band thermal radiation energy in a furnace provided in the embodiment of the present application inputs the obtained target visible light thermal radiation energy of each preset measuring point in the boiler furnace into the trained prediction model, thereby obtaining the target full-band thermal radiation energy corresponding to each preset measuring point, wherein the trained prediction model is obtained based on the visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions, and the visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, and the concentration of particulate matter and the concentration of gas in the boiler, so that when training the prediction model, it is not necessary to set up a variety of sensors to detect the thermal radiation energy to obtain the training set. The embodiment of the present application is based on the obtained target visible light thermal radiation energy, and the prediction model that has learned the mapping relationship between the visible light thermal radiation energy generated by the combustion in the furnace and the full-band thermal radiation energy, and can directly and accurately obtain the corresponding target full-band thermal radiation energy, that is, it realizes the online and accurate prediction of the thermal radiation energy in the furnace, and at the same time, because no additional infrared radiation energy sensor is required, it is low-cost and easy to implement.

[0127] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0128] Figure 5 Schematic diagram of the structure of a device for predicting full-band thermal radiation energy in a furnace provided in one embodiment of the present application. Figure 5 As shown, the device for predicting full-band thermal radiation energy in a furnace provided in this embodiment may include: a first acquisition module 501 and a first obtaining module 502 .

[0129] The first acquisition module 501 is used to acquire the target visible light thermal radiation energy of each preset measuring point in the boiler furnace.

[0130] The first obtaining module 502 is used to input the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point.

[0131] Among them, the trained prediction model is obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions; the full-band includes visible light band to infrared band.

[0132] The visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

[0133] Optionally, the full-band thermal radiation energy prediction device in the furnace may also include:

[0134] The second acquisition module is used to obtain the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler.

[0135] The second obtaining module is used to obtain radiation characteristic parameters for each preset measuring point under each operating condition according to the particle concentration and the gas concentration corresponding to the preset measuring point under the operating condition; wherein the radiation characteristic parameters include absorption coefficient, scattering coefficient and attenuation coefficient; based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the operating condition, obtain visible light thermal radiation energy and full-band thermal radiation energy.

[0136] The third obtaining module is used to form a training set with the visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions, and train the preset prediction model based on the training set to obtain a trained prediction model.

[0137] Optionally, the second acquisition module is further used for:

[0138] According to the design parameters and operating parameters of the boiler, different working conditions are divided;

[0139] For each operating condition, a combustion numerical simulation is performed according to the design parameters, the operating parameters, and the combustion mode of the boiler under the operating condition to obtain the overall furnace temperature, the overall particulate matter concentration, and the overall gas concentration under the operating condition;

[0140] Based on the position of each preset measuring point, as well as the overall furnace temperature, overall particle concentration and overall gas concentration under the operating condition, the furnace temperature, particle concentration and gas concentration corresponding to each preset measuring point under the operating condition are determined.

[0141] Optionally, the second module is further used to:

[0142] The furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition are input into a preset spectral imaging model to obtain spectral radiation energy; wherein the spectral imaging model inputs the furnace temperature and radiation characteristic parameters to output spectral radiation energy;

[0143] Integrating the spectral radiation energy along the wavelength in the visible light band to obtain visible light thermal radiation energy;

[0144] The spectral radiation energy is fully integrated along the wavelength to obtain the full-band thermal radiation energy.

[0145] Optionally, the second module is further used to:

[0146] According to the particle concentration corresponding to the preset measuring point under the working condition, the absorption coefficient, scattering coefficient and attenuation coefficient of the particles are obtained;

[0147] The gas absorption coefficient is obtained according to the gas concentration corresponding to the preset measuring point under the working condition.

[0148] Optionally, the preset prediction model includes a hidden layer; the third obtaining module is further used for:

[0149] Input the visible light thermal radiation energy of a preset measuring point under a working condition in the training set into the preset prediction model to obtain the corresponding full-band thermal radiation energy prediction value;

[0150] Determine the loss value of the loss function according to the difference between the predicted value of the full-band thermal radiation energy and the corresponding full-band thermal radiation energy, and judge whether the loss value is less than a preset loss value;

[0151] If the loss value is less than the preset loss value, the training is completed and a trained prediction model is obtained;

[0152] If the loss value is greater than or equal to the preset loss value, the weights of all neurons in the hidden layer are updated, and the visible light thermal radiation energy of a preset measuring point under the next working condition in the training set is re-input into the preset prediction model until the training is completed.

[0153] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0154] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0155] Those of ordinary skill in the art will appreciate that the templates, units, and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0156] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium.

[0157] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting full-band thermal radiation energy in a furnace, characterized in that: include: Obtain the target visible light thermal radiation energy of each preset measuring point in the boiler furnace; The target visible light thermal radiation energy of each preset measuring point is input into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point; Among them, the trained prediction model is obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions; the full band includes the visible light band to the infrared band; The visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

2. The method for predicting full-band thermal radiation energy in a furnace according to claim 1, characterized in that: Before inputting the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy, the method further includes: Obtaining the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler; For each preset measuring point under each working condition, a radiation characteristic parameter is obtained according to the particle concentration and the gas concentration corresponding to the preset measuring point under the working condition; wherein the radiation characteristic parameter includes an absorption coefficient, a scattering coefficient and an attenuation coefficient; based on the furnace temperature and the radiation characteristic parameter corresponding to the preset measuring point under the working condition, the visible light thermal radiation energy and the full-band thermal radiation energy are obtained; The visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions constitute a training set, and the preset prediction model is trained based on the training set to obtain a trained prediction model.

3. The method for predicting full-band thermal radiation energy in a furnace according to claim 2, characterized in that: The obtaining of the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler includes: According to the design parameters and operating parameters of the boiler, different working conditions are divided; For each operating condition, a combustion numerical simulation is performed according to the design parameters, the operating parameters, and the combustion mode of the boiler under the operating condition to obtain the overall furnace temperature, the overall particulate matter concentration, and the overall gas concentration under the operating condition; Based on the position of each preset measuring point, as well as the overall furnace temperature, overall particle concentration and overall gas concentration under the operating condition, the furnace temperature, particle concentration and gas concentration corresponding to each preset measuring point under the operating condition are determined.

4. The method for predicting full-band thermal radiation energy in a furnace according to claim 2, characterized in that: For each preset measuring point under each working condition, the visible light thermal radiation energy and the full-band thermal radiation energy are obtained based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition, including: The furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition are input into a preset spectral imaging model to obtain spectral radiation energy; wherein the spectral imaging model inputs the furnace temperature and radiation characteristic parameters to output spectral radiation energy; Integrating the spectral radiation energy along the wavelength in the visible light band to obtain visible light thermal radiation energy; The spectral radiation energy is fully integrated along the wavelength to obtain the full-band thermal radiation energy.

5. The method for predicting full-band thermal radiation energy in a furnace according to any one of claims 1 to 4, characterized in that: The preset measuring points are arranged on the side wall of the boiler furnace, and the space formed by the positions of all the preset measuring points covers the combustion and heat release high temperature zone of the boiler; the combustion and heat release high temperature zone is an area where the heat is higher than the preset heat threshold.

6. The method for predicting full-band thermal radiation energy in a furnace according to claim 2, characterized in that: For each preset measuring point under each working condition, obtaining the radiation characteristic parameter according to the particle concentration and the gas concentration corresponding to the preset measuring point under the working condition includes: According to the particle concentration corresponding to the preset measuring point under the working condition, the absorption coefficient, scattering coefficient and attenuation coefficient of the particles are obtained; The gas absorption coefficient is obtained according to the gas concentration corresponding to the preset measuring point under the working condition.

7. The method for predicting full-band thermal radiation energy in a furnace according to claim 2, characterized in that: The preset prediction model includes a hidden layer; The step of training a preset prediction model based on the training set to obtain a trained prediction model includes: Input the visible light thermal radiation energy of a preset measuring point under a working condition in the training set into the preset prediction model to obtain the corresponding full-band thermal radiation energy prediction value; Determine the loss value of the loss function according to the difference between the predicted value of the full-band thermal radiation energy and the corresponding full-band thermal radiation energy, and judge whether the loss value is less than a preset loss value; If the loss value is less than the preset loss value, the training is completed and a trained prediction model is obtained; If the loss value is greater than or equal to the preset loss value, the weights of all neurons in the hidden layer are updated, and the visible light thermal radiation energy of a preset measuring point under the next working condition in the training set is re-input into the preset prediction model until the training is completed.

8. The method for predicting full-band thermal radiation energy in a furnace according to claim 5, characterized in that: The side wall of the boiler furnace is provided with a first loop line and a second loop line around four side walls of the furnace, and the space formed by the first loop line and the second loop line covers the combustion heat release high temperature area of ​​the boiler; Four first preset measuring points are arranged on the first loop line, and the four first preset measuring points are arranged on four side walls of the furnace respectively; Four second preset measuring points are arranged on the second loop line, and the four second preset measuring points are arranged on four side walls of the furnace respectively.

9. A device for predicting full-band thermal radiation energy in a furnace, characterized in that: include: The first acquisition module is used to acquire the target visible light thermal radiation energy of each preset measuring point in the boiler furnace; The first obtaining module is used to input the target visible light thermal radiation energy of each preset measuring point into the trained prediction model to obtain the target full-band thermal radiation energy corresponding to each preset measuring point; Among them, the trained prediction model is obtained based on the visible light thermal radiation energy and full-band thermal radiation energy of each preset measuring point under different working conditions; the full band includes the visible light band to the infrared band; The visible light thermal radiation energy and the full-band thermal radiation energy are determined according to the furnace temperature of the boiler, as well as the particle concentration and gas concentration in the boiler.

10. The device for predicting full-band thermal radiation energy in a furnace according to claim 9, characterized in that: The device also includes: The second acquisition module is used to obtain the furnace temperature, particulate matter concentration and gas concentration corresponding to each preset measuring point under different operating conditions of the boiler; The second obtaining module is used to obtain radiation characteristic parameters for each preset measuring point under each working condition according to the particle concentration and the gas concentration corresponding to the preset measuring point under the working condition; wherein the radiation characteristic parameters include absorption coefficient, scattering coefficient and attenuation coefficient; based on the furnace temperature and radiation characteristic parameters corresponding to the preset measuring point under the working condition, obtain visible light thermal radiation energy and full-band thermal radiation energy; The third obtaining module is used to form a training set with the visible light thermal radiation energy and the full-band thermal radiation energy of each preset measuring point under different working conditions, and train the preset prediction model based on the training set to obtain a trained prediction model.

Images