Porous medium burner prediction and prediction model training method and device

Abstract

The invention provides a porous medium combustor prediction and prediction model training method and device, and the method comprises the steps: building a target working condition parameter combination space, and determining a lean-burn limit value; determining a target training element point from the target working condition parameter combination space based on a lean burn limit value, and obtaining real temperature field space distribution; inputting the parameter configuration into the model to generate a predicted temperature field spatial distribution; and adjusting the model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution until the training is finished, and outputting a target porous medium combustor prediction model. By determining the lean-burn limit value and the target training element point, the sampling deviation can be reduced, the accuracy of the model can be improved, and rapid and accurate prediction of the stable-combustion working condition parameters corresponding to the target temperature field under the design structure of the porous medium combustor can be realized; and meanwhile, the structure of the porous medium combustor is supported in the design space range, the engineering design is accelerated through iterative optimization of working condition parameters, and the design period is shortened.

Description

Technical Field

[0001] The present disclosure relates to the technical field of porous medium combustion, and in particular to a method and device for predicting a porous medium burner and training a prediction model. Background Art

[0002] In current technology, numerical simulation of ceramic burners is mainly focused on simulation under normal intake temperature (298K), normal pressure (1atm) and low intake velocity (<1m / s) conditions, while research on high-power combustion technology under near-gas turbine intake temperature (>500K) and high intake velocity (>20m / s) conditions is almost blank. The increase in airflow velocity will bring about problems such as flame stability. How to adapt to the actual gas turbine operating conditions, expand the lean combustion limit, and achieve flame stability combustion is an urgent problem to be solved. On the other hand, although high-fidelity numerical simulation can better simulate the spatial distribution of various physical fields in the combustion chamber, its computational consumption is large and can only be used for mechanistic research. Summary of the invention

[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, an object of the present disclosure is to propose a porous media burner prediction model training method.

[0005] The second objective of the present disclosure is to provide a porous media burner prediction method.

[0006] The third objective of the present disclosure is to provide a porous media burner prediction model training device.

[0007] The fourth objective of the present disclosure is to provide a porous media burner prediction device.

[0008] A fifth objective of the present disclosure is to provide an electronic device.

[0009] A sixth object of the present disclosure is to provide a non-transitory computer-readable storage medium.

[0010] A seventh objective of the present disclosure is to provide a computer program product.

[0011] To achieve the above-mentioned purpose, the first aspect of the present disclosure proposes a porous media burner prediction model training method, including: establishing a target operating parameter combination space of a porous media burner, and determining a lean burn limit value of the porous media burner, wherein the parameter space includes multiple element points, each element point includes a combination of operating parameters of the porous media burner, and the operating parameter combinations in the multiple element points are different; determining a target training element point from the target operating parameter combination space based on the lean burn limit value, and establishing an initial porous media burner prediction model of the porous media burner based on the lean burn limit value, and obtaining the real temperature field spatial distribution of the target training element point; inputting the parameter configuration in the target training element point into the initial porous media burner prediction model to generate a predicted temperature field spatial distribution; adjusting the initial porous media burner prediction model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, returning to execute the steps of determining the target training element point from the parameter space, obtaining the real temperature field spatial distribution of the target training element point and subsequent steps until the training is completed, and outputting the target porous media burner prediction model.

[0012] According to one embodiment of the present disclosure, establishing a target operating parameter combination space of a porous medium burner includes: obtaining a spatial range of the porous medium burner and a preset initial operating parameter combination space; sampling within the spatial range by a target sampling method based on the spatial range to obtain candidate training element points; and mapping the candidate training element points to the initial operating parameter combination space to generate a target operating parameter combination space.

[0013] According to one embodiment of the present disclosure, the target sampling method is one of the Latin hypercube sampling method, the Sobol method, the Kronecker method, and the randomness method.

[0014] According to one embodiment of the present disclosure, determining the target training element point from the target operating condition parameter combination space based on the lean fuel limit value includes: sampling within a limited range of the lean fuel limit value of the target operating condition parameter combination space based on a boundary resampling method to obtain the target training element point.

[0015] According to one embodiment of the present disclosure, the initial porous media burner prediction model of the porous media burner is established based on the lean burn limit value, including: obtaining the structural parameters of the porous media burner; establishing a gas phase mass conservation equation, a gas phase equation and a gas-solid heat transfer equation based on the structural parameters; and establishing a one-dimensional model based on the lean burn limit value, the gas phase mass conservation equation, the gas phase energy conservation equation and the gas-solid heat transfer equation, i.e., the initial porous media burner prediction model.

[0016] To achieve the above-mentioned purpose, the second aspect embodiment of the present disclosure proposes a porous media burner prediction method, including: obtaining the target operating parameters to be predicted of the porous media burner and a target porous media burner prediction model, wherein the target porous media burner prediction model is obtained by the porous media burner prediction model training method as described in the first aspect embodiment; inputting the target operating parameters into the target porous media burner prediction model to generate the spatial distribution of the temperature field of the porous media burner at the target operating parameters.

[0017] According to one embodiment of the present disclosure, the method further includes: determining the average temperature, maximum temperature and flue gas temperature of the porous medium burner at the target operating parameters based on the spatial distribution of the temperature field.

[0018] To achieve the above-mentioned purpose, the third aspect of the present disclosure proposes a porous medium burner prediction model training device, including: an establishment module, used to establish a target operating parameter combination space of a porous medium burner, wherein the parameter space includes multiple element points, each element point includes a combination of operating parameters of the porous medium burner, and the operating parameter combinations in multiple element points are different; a determination module, used to determine the target training element point from the target operating parameter combination space, and obtain the real temperature field spatial distribution of the target training element point; an input module, used to input the parameter configuration in the target training element point into the initial porous medium burner prediction model to generate a predicted temperature field spatial distribution; a training module, used to adjust the initial porous medium burner prediction model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, return to execute the determination of the target training element point from the parameter space, and obtain the real temperature field spatial distribution of the target training element point and its subsequent steps until the training is completed, and output the target porous medium burner prediction model.

[0019] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present disclosure proposes a porous media burner prediction device, including: an acquisition module, used to obtain the target operating parameters to be predicted of the porous media burner and a target porous media burner prediction model, wherein the target porous media burner prediction model is obtained by the porous media burner prediction model training method as described in the first aspect embodiment; an output module, used to input the target operating parameters into the target porous media burner prediction model to generate the spatial distribution of the temperature field of the porous media burner when the target operating parameters are.

[0020] To achieve the above-mentioned purpose, the fifth aspect embodiment of the present disclosure proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to implement the porous media burner prediction model training method as described in the first aspect embodiment of the present disclosure, or the porous media burner prediction method as described in the second aspect embodiment of the present disclosure.

[0021] To achieve the above-mentioned purpose, the sixth aspect embodiment of the present disclosure proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the porous media burner prediction model training method as described in the first aspect embodiment of the present disclosure, or the porous media burner prediction method as described in the second aspect embodiment of the present disclosure.

[0022] To achieve the above-mentioned purpose, the seventh aspect embodiment of the present disclosure proposes a computer program product, including a computer program, which, when executed by a processor, is used to implement the porous media burner prediction model training method as described in the first aspect embodiment of the present disclosure, or the porous media burner prediction method as described in the second aspect embodiment of the present disclosure.

[0023] Therefore, by determining the lean combustion limit value and determining the target training element point from the target operating condition parameter combination space based on the lean combustion limit value, the sampling deviation can be reduced, the accuracy of the model can be improved, and the rapid and accurate prediction of the steady combustion condition parameters corresponding to the target temperature field under the porous medium burner design structure can be achieved. At the same time, it supports the iterative optimization of the porous medium burner structure and operating parameters within the design space, accelerates engineering design, and shortens the design cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic diagram of a porous media burner prediction model training method according to an embodiment of the present disclosure;

[0025] Figure 2 This is a temperature distribution diagram under different flow rate conditions with an equivalence ratio of 0.8 according to an embodiment of the present disclosure;

[0026] Figure 3 is a temperature distribution diagram of different inlet air temperatures under the same equivalence ratio and flow rate conditions according to one embodiment of the present disclosure;

[0027] Figure 4 is a schematic diagram of another porous media burner prediction model training method according to an embodiment of the present disclosure;

[0028] Figure 5 It is a proxy model training set corresponding to the spatial distribution of sample points in one embodiment of the present disclosure;

[0029] Figure 6 It is the temperature distribution cloud map predicted by Gaussian process in the new sample space;

[0030] Figure 7 It is a line graph of temperature distribution predicted based on Gaussian process in the new sample space;

[0031] Figure 8 It is a line graph of temperature distribution predicted based on Gaussian process in the new sample space;

[0032] Fig. 9 It is a proxy model training set corresponding to the spatial distribution of sample points in one embodiment of the present disclosure;

[0033] Fig.10 is a schematic diagram of a one-dimensional model of a target porous media burner according to one embodiment of the present disclosure;

[0034] Fig.11 is a schematic diagram of a porous media burner prediction method according to an embodiment of the present disclosure;

[0035] Fig.12 is a schematic diagram of a porous media burner prediction model training device according to one embodiment of the present disclosure;

[0036] Fig.13 is a schematic diagram of a porous media burner prediction device according to one embodiment of the present disclosure;

[0037] Fig.14 is a schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0038] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0039] The acquisition, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of relevant laws and regulations.

[0040] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0041] Figure 1is a schematic diagram of a porous media burner prediction model training method according to an embodiment of the present disclosure, such as Figure 1 As shown, the porous media burner prediction model training method includes the following steps:

[0042] S101, establishing a target operating parameter combination space of a porous medium burner, and determining a lean combustion limit value of the porous medium burner, wherein the parameter space includes a plurality of element points, each element point includes an operating parameter combination of the porous medium burner, and the operating parameter combinations in the plurality of element points are different.

[0043] The porous media burner prediction model training method of the embodiment of the present application can be applied to the scenario of parameter configuration of the porous media burner. The executor of the porous media burner prediction model training of the embodiment of the present application can be the porous media burner prediction model training device of the embodiment of the present application, and the porous media burner prediction model training device can be set on an electronic device.

[0044] It should be noted that the parameters in the operating condition parameter combination can be used to characterize the corresponding parameter values ​​of the porous medium burner under different operating conditions. The parameters in the operating condition parameter combination can be designed in advance and are parameters that affect the operation of the porous medium burner. Optionally, the parameters in the operating condition parameter combination can also be determined through experiments.

[0045] It should be noted that the operating condition parameter combination may include multiple parameters, which are not limited here. For example, the operating condition parameter combination may include inlet air temperature, equivalence ratio, ambient pressure, etc.

[0046] It should be noted that the lean burn limit refers to the minimum fuel / air ratio at which the flame is extinguished due to lack of oxygen during the fuel combustion process. Under oxygen-poor conditions, the fuel cannot burn completely and a large amount of smoke and harmful gases will be produced. The lean burn limit is usually measured by the air-fuel ratio (AFR) or the excess air coefficient (λ), where λ=1 indicates complete combustion, less than 1 indicates rich combustion (too much fuel), and greater than 1 indicates lean combustion (too much air). The lean burn limit is very important for the design and operation of the engine because it determines factors such as combustion efficiency, emission levels and power output.

[0047] In the disclosed embodiment, the lean burn limit value can be obtained by experimental measurement. By changing the ratio of the fuel supply and the air flow rate, it can be observed whether the flame can burn stably. When the flame begins to flicker or goes out, the fuel / air ratio at this time is considered to be the lean burn limit value. In addition, the lean burn limit value can also be determined by analyzing the chemical composition of the combustion products, because lean combustion will cause unburned fuel to remain in the exhaust gas. The lean burn limit value is crucial to optimizing engine performance and reducing emissions, so it is often studied and tested in the automotive industry and other related industries.

[0048] In another possible implementation method, the inlet boundary conditions can also be determined, and then a one-dimensional combustion characteristic prior analysis is carried out based on the inlet boundary conditions to determine the flammable and non-flammable points under different operating conditions. Combined with the actual gas turbine operating conditions, for conditions that meet the conditions, the lean limit boundary is determined by resampling in the area near the lean limit.

[0049] It should be noted that the one-dimensional combustion characteristics prior analysis is carried out based on the inlet boundary conditions. However, by combining the combustion and heat transfer theories, the corresponding laws between the temperature, pressure, flow rate and other parameters of different inlets and the outlet temperature, pressure, flow rate and other parameters are calculated, and then the combustion characteristics of the burner are analyzed a priori in combination with the existing data sets.

[0050] S102, determining a target training element point from a target operating condition parameter combination space based on the lean burn limit value, establishing an initial porous medium burner prediction model of the porous medium burner based on the lean burn limit value, and obtaining a real temperature field spatial distribution of the target training element point.

[0051] In the embodiments of the present disclosure, there are many methods for determining the target training element points from the target operating condition parameter combination space, which are not limited here.

[0052] In one possible implementation, a target training element point can be determined from a target operating condition parameter combination space by a screening model, where the screening model is pre-trained and used to determine the target training element point from the target operating condition parameter combination space. The screening model can be stored in a storage space of an electronic device to facilitate retrieval and use when needed.

[0053] Optionally, the target training element point can also be determined from the target operating condition parameter combination space according to user needs or training needs. The user needs or training needs are designed in advance and can be changed according to actual design needs, and no limitation is made here.

[0054] In one possible implementation, the user can control the accuracy, functional bias, etc. of the final generated model by changing user needs or training requirements.

[0055] S103, inputting the parameter configuration in the target training element point into the initial porous medium burner prediction model to generate a predicted temperature field spatial distribution.

[0056] It should be noted that the spatial distribution of temperature field refers to the variation of temperature with position in a given area. In many engineering and scientific studies, understanding the spatial distribution of temperature field is crucial for analyzing and optimizing system performance. This includes but is not limited to the study of heat conduction, heat convection, radiation heat transfer and other issues.

[0057] S104, adjusting the initial porous medium burner prediction model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, returning to execute determining the target training element point from the parameter space, and obtaining the real temperature field spatial distribution of the target training element point and its subsequent steps until the training is completed, and outputting the target porous medium burner prediction model.

[0058] It can be understood that model training is an iterative process, which is carried out by continuously adjusting the network parameters of the model until the conditions for the end of training are met.

[0059] It should be noted that there may be multiple conditions for the end of training. For example, the end of training may be that the training reaches a preset number of times, or the training time reaches a preset time, or the loss value of the training result meets the loss threshold, etc.

[0060] In an embodiment of the present disclosure, a target operating parameter combination space of a porous medium burner is first established, wherein the parameter space includes multiple element points, each element point includes an operating parameter combination of the porous medium burner, and the operating parameter combinations in the multiple element points are different. Then, a target training element point is determined from the target operating parameter combination space, and the real temperature field spatial distribution of the target training element point is obtained. Then, the parameter configuration in the target training element point is input into the initial porous medium burner prediction model to generate a predicted temperature field spatial distribution. Finally, the initial porous medium burner prediction model is adjusted based on the real temperature field spatial distribution and the predicted temperature field spatial distribution. Return to execute the steps of determining the target training element point from the parameter space, obtaining the real temperature field spatial distribution of the target training element point and subsequent steps until the training is completed, and output the target porous medium burner prediction model. Therefore, by determining the lean combustion limit value and determining the target training element point from the target operating condition parameter combination space based on the lean combustion limit value, the sampling deviation can be reduced, the accuracy of the model can be improved, and the rapid and accurate prediction of the steady combustion condition parameters corresponding to the target temperature field under the porous medium burner design structure can be achieved. At the same time, it supports the iterative optimization of the porous medium burner structure and operating parameters within the design space, accelerates engineering design, and shortens the design cycle.

[0061] In the embodiment of the present disclosure, the existing data set can be used to obtain the influence of different operating parameters on the working characteristics of the burner, and realize the prior analysis of the one-dimensional combustion characteristics of the porous medium burner. Figure 2 The temperature distribution diagram under the conditions of equivalence ratio 0.8 and different flow rates is shown in Figure 2. Figure 3The temperature distribution diagram at different inlet air temperatures under the same equivalence ratio and flow rate conditions shows that under high flow rate conditions, with the increase of flow rate, the maximum flame temperature and the average flame temperature in the stable combustion zone decrease. This is because the increase in flow rate will enhance the convective heat transfer and take away more heat, resulting in a decrease in flame temperature and weakened flame stability. With the increase of inlet air temperature, the overall temperature of the stable combustion zone increases significantly, indicating that increasing the preheating temperature can enhance flame stability and is beneficial to flame stationary and immersion.

[0062] In the above embodiment, the target operating parameter combination space of the porous medium burner is established, and the target operating parameter combination space of the porous medium burner is established. Figure 4 Explaining further, the method includes:

[0063] S401, obtaining the spatial range of the porous medium burner and the preset initial operating condition parameter combination space.

[0064] It should be noted that the spatial range is the range for sampling, which can be determined manually or according to the actual working conditions and parameters of the porous medium burner.

[0065] The initial operating condition parameter combination space includes all element points in the target operating condition parameter combination space and other element points. Each element point includes an operating condition parameter combination of the porous medium burner, and the operating condition parameter combinations in multiple element points are different.

[0066] It should be noted that the spatial range and the initial operating condition parameter combination space are designed in advance and can be changed according to actual design needs, and no limitation is made here.

[0067] By setting the spatial range and the initial operating condition parameter combination space, the accuracy and performance of the final generated target porous media burner prediction model can be controlled.

[0068] S402: Sampling is performed in the spatial range by a target sampling method based on the spatial range to obtain candidate training element points.

[0069] It should be noted that the target sampling method may be one of the Latin hypercube sampling method, the Sobol method, the Kronecker method, and the randomness method.

[0070] In one embodiment of the present disclosure, the design parameter variables that can be confirmed based on DoE are inlet air velocity and fuel-air equivalence ratio, which range from 0.3 to 30 m / s and fuel-air equivalence ratio from 0.2 to 0.8, respectively. The spatial distribution of the design points in the design parameter space is obtained based on the Latin hypercube sampling method. Figure 5Given different operating parameters (i.e., air temperatures of room temperature, 400K, 550K, and 700K, and pressure of ambient pressure) and the spatial distribution of design points, the calculation of the premixed gas inlet temperature and inlet mass flux can be realized, where the inlet boundary conditions to be given are the premixed gas inlet temperature and mass flux and the design parameter variable of the equivalence ratio.

[0071] S403, mapping the candidate training element points to the initial operating condition parameter combination space to generate a target operating condition parameter combination space.

[0072] In an embodiment of the present disclosure, the spatial range of the porous media burner and the preset initial operating parameter combination space are first obtained, and then sampling is performed in the spatial range through a target sampling method based on the spatial range to obtain candidate training element points, and finally the candidate training element points are mapped to the initial operating parameter combination space to generate a target operating parameter combination space.

[0073] It should be noted that the target training element point is determined from the target operating condition parameter combination space based on the lean burn limit value, and the target operating condition parameter combination space can be sampled within a limited range of the lean burn limit value based on the boundary resampling method to obtain the target training element point.

[0074] In one possible implementation, a boundary resampling method can be used to uniformly select points at the boundary to cover the boundary of the input space, reduce sampling bias, and improve the accuracy of the model, such as Figure 5 Rectangular scatter points are shown, based on Figure 5 The spatial distribution of the sample points shown in the figure is used to build the corresponding proxy model training set. The training set is used to build a proxy model based on Gaussian process (GP), in which the parameterized input parameter is the spatial distribution of the design point, and the output parameter is the spatial distribution of the temperature field. Further, a response surface model of the parameterized input and the average temperature, maximum temperature and flue gas temperature of the stable combustion zone is constructed, as shown in Figure 6 , 7, 8, where Figure 6 is the temperature distribution cloud map predicted by Gaussian process in the new sample space, Figure 7 is the temperature distribution line diagram (average temperature, maximum temperature and flue gas temperature, constant flow rate) predicted based on Gaussian process in the new sample space. Figure 8The temperature distribution line diagram (fixed equivalence ratio) predicted based on Gaussian process in the new sample space is shown. The results show that: 1. The prediction results of the proxy model based on the boundary resampling method for the temperature field are in line with the physical laws and have high prediction accuracy, indicating the effectiveness of the boundary resampling method in improving the prediction accuracy of the proxy model; 2. Under high flow rate conditions, the maximum temperature, average temperature and flue gas temperature in the stable combustion zone are slightly different, which is because the high flow rate strengthens the heat exchange process; 3. The temperature near the equivalence ratio of 0.4 and the inlet air velocity of 20m / s is relatively low, which is because the area is close to the lean combustion limit and close to cold state regression; 4. The equivalence ratio has a greater impact on the temperature. With the increase of the equivalence ratio, the overall temperature of the stable combustion zone increases; Under low flow rate conditions, with the increase of flow rate, the overall temperature of the stable combustion zone increases. Under high flow rate conditions, with the increase of flow rate, the temperature change is not obvious. This is because under high flow rate conditions, with the increase of flow rate, on the one hand, the fuel flow increases, the total heat released by the fuel increases, and on the other hand, more heat is taken away with the increase of flow rate.

[0075] In the above embodiment, the initial porous medium burner prediction model of the porous medium burner is established based on the lean combustion limit value, and the prediction model of the porous medium burner can also be Fig. 9 Explaining further, the method includes:

[0076] S901, obtaining structural parameters of the porous medium burner.

[0077] It should be noted that the structural parameters of the porous medium burner may include various ones, which are not limited here, and may be limited according to actual design requirements. For example, the structural parameters may include inlet temperature, inlet equivalence ratio, inlet mass flow rate, etc.

[0078] S902, establishing the gas phase mass conservation equation, the gas phase equation and the gas-solid heat transfer equation based on the structural parameters.

[0079] It should be noted that the gas phase mass conservation equation, gas phase equation and gas-solid heat transfer equation established based on the structural parameters can be shown as follows:

[0080] 1. Gas phase mass conservation equation: In a fluidized bed reactor, the gas phase mass conservation equation describes that the mass inflow and outflow rate of the gas is equal to the difference between the generation and consumption rates. The mathematical expression is as follows:

[0081] Where ρg is the gas density, vg is the gas velocity, t is the time, and S_m is the material source term, which represents the gas mass flux produced by the reaction.

[0082] 2. Gas phase energy conservation equation: The gas phase energy conservation equation describes that the rate of energy inflow and outflow of gas is equal to the rate of internal energy change plus the rate of pressure work and heat exchange. The mathematical expression is as follows: Among them, u_g is the internal energy of the gas, Q_v is the volume heat source, and W_p is the pressure work.

[0083] 3. Gas-solid heat transfer equation: The gas-solid heat transfer equation describes the heat transfer process between gas and solid. In a fluidized bed reactor, this is usually achieved through heat conduction and convection heat transfer. The mathematical expression is as follows: hA(T_s-T_g)=kA(T_s-T_w) where h is the heat transfer coefficient, A is the contact area, T_s is the solid temperature, T_g is the gas temperature, k is the thermal conductivity, and T_w is the wall temperature.

[0084] It should be noted that the gas-solid heat transfer equation obtains the one-dimensional distribution of the temperature field by solving the solid-environment radiation heat loss, solid-solid conduction / radiation heat transfer and solid-gas volume heat exchange.

[0085] In the disclosed embodiments, the material thermal conductivity data is 1TPRC data series, the thermodynamic properties are calculated using NASAReport, the radiation thermal conductivity is calculated using the Roeesland model, the chemical reaction mechanism is calculated using GRI 3.0, the chemical reaction network consists of 20 reactors, and it is assumed that the combustion process reaches stability at 3900 seconds.

[0086] S903, establishing a one-dimensional model, namely, an initial porous medium burner prediction model, based on the lean combustion limit value, the gas phase mass conservation equation, the gas phase energy conservation equation, and the gas-solid heat transfer equation.

[0087] It should be noted that the one-dimensional model simplifies the problem to changes in one direction and ignores the effects in other directions. For the one-dimensional model of the fluidized bed reactor, we can assume that the reactor changes along the height direction (usually the axial direction) and is uniform in the radial direction (cross section).

[0088] In the embodiment of the present disclosure, the combustion process of the target structure can be modeled by using Cantera to establish an initial one-dimensional model, and then the model is trained according to the method in the above embodiment to finally generate a target porous media burner prediction model. For example, the one-dimensional model in the embodiment of the present disclosure can be as follows: Fig.10 As shown, the total length of the combustion chamber is 9.4 inches, PPI stands for void density, which characterizes the average number of pores per unit inch of length, ε represents emissivity, and the cavity part is assumed to be a ZrO2 ring with a porosity of 0.5 and an emissivity of 0.001. Cantera is used to model the combustion process of the target structure, including the gas phase mass conservation equation, the gas phase energy conservation equation, and the gas-solid heat transfer equation to obtain the spatial distribution of the temperature field.

[0089] By establishing a one-dimensional model, the prediction of porous media burners can be simplified and the calculation efficiency can be improved.

[0090] It should be noted that the one-dimensional model can be represented by a multivariate Gaussian process (GP), as follows:

[0091] Y (i) ~GP(μ;∑);

[0092] where Y (i) represents the temperature value corresponding to the i-th reactor, is the covariance function, which uses the square exponential form and the formula is as follows:

[0093]

[0094] where θ k It represents the correlation between inputs and determines the smoothness and oscillation of the Gaussian process.

[0095] In the training phase, known input and output samples are used to learn the parameters of the Gaussian process, including the parameters of the mean function and covariance function. In the prediction phase, new data points are input and the learned model is used to predict the output and evaluate the uncertainty of the model. It should be noted that uncertainty can be used to evaluate whether the model has been satisfied.

[0096] Temperature prediction vector at the new target training element point It can be expressed as follows:

[0097]

[0098] In the formula, Where n is the number of target training element points, r is the dimension of the temperature vector, which is equivalent to the number of reactors, represents the Kronecker product, 1 n is a unit vector, c and C measure the correlation between the new target training element point and the training set and the training set respectively.

[0099] In the embodiment of the present disclosure, the structural parameters of the porous medium burner are first obtained, and then the gas phase mass conservation equation, gas phase equation and gas-solid heat transfer equation are established based on the structural parameters. Finally, a one-dimensional model is established based on the lean combustion limit, the gas phase mass conservation equation, the gas phase energy conservation equation and the gas-solid heat transfer equation, i.e., the initial porous medium burner prediction model. Thus, by establishing the gas phase mass conservation equation, the gas phase equation and the gas-solid heat transfer equation based on the structural parameters and establishing a one-dimensional model, the porous medium burner prediction can be simplified, the calculation efficiency can be improved, and the prediction cost can be reduced.

[0100] Fig.11 is a schematic diagram of a porous media burner prediction method according to an embodiment of the present disclosure, such as Fig.11As shown, the porous media burner prediction method includes the following steps:

[0101] S1101, obtaining target operating parameters of a porous medium burner to be predicted and a target porous medium burner prediction model.

[0102] It should be noted that the target porous media burner prediction model in the embodiment of the present disclosure is as follows: Figure 1-Figure 3 The porous media burner prediction model training method of the embodiment is obtained.

[0103] S1102, inputting the target operating condition parameters into the target porous medium burner prediction model to generate the temperature field spatial distribution of the porous medium burner under the target operating condition parameters.

[0104] In the embodiment of the present disclosure, after obtaining the spatial distribution of the temperature field of the porous medium burner at the target operating parameters, the average temperature, maximum temperature and flue gas temperature of the porous medium burner at the target operating parameters can also be determined based on the spatial distribution of the temperature field.

[0105] It should be noted that the average temperature, maximum temperature and flue gas temperature may be defined as the average value of the temperature of a specific reactor. For example, the average temperature may be defined as the average value of the temperature of the 11th reactor to the 20th reactor, the maximum temperature may be defined as the highest value of the temperature of the 11th reactor to the 20th reactor, and the flue gas temperature may be defined as the temperature of the 20th reactor.

[0106] Corresponding to the porous medium burner prediction model training methods provided in the above-mentioned embodiments, an embodiment of the present disclosure also provides a porous medium burner prediction model training device. Since the porous medium burner prediction model training device provided in the embodiment of the present disclosure corresponds to the porous medium burner prediction model training methods provided in the above-mentioned embodiments, the implementation method of the above-mentioned porous medium burner prediction model training method is also applicable to the porous medium burner prediction model training device provided in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0107] Fig.12 is a schematic diagram of a porous media burner prediction model training device according to an embodiment of the present disclosure, such as Fig.12 As shown, the porous medium burner prediction model training device 1200 includes: an establishment module 1210, a determination module 1220, an input module 1230 and a training module 1240.

[0108] Establish module 1210, which is used to establish a target operating parameter combination space of the porous medium burner and determine the lean combustion limit value of the porous medium burner, wherein the parameter space includes multiple element points, each element point includes an operating parameter combination of the porous medium burner, and the operating parameter combinations in the multiple element points are different.

[0109] The determination module 1220 is used to determine the target training element point from the target operating condition parameter combination space based on the lean burn limit value, establish an initial porous medium burner prediction model of the porous medium burner based on the lean burn limit value, and obtain the real temperature field spatial distribution of the target training element point.

[0110] The input module 1230 is used to input the parameter configuration in the target training element point into the initial porous media burner prediction model to generate a predicted temperature field spatial distribution.

[0111] The training module 1240 is used to adjust the initial porous medium burner prediction model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, return to execute the determination of the target training element point from the parameter space, and obtain the real temperature field spatial distribution of the target training element point and its subsequent steps until the training is completed, and output the target porous medium burner prediction model.

[0112] In one embodiment of the present disclosure, module 1210 is established, which is also used to: obtain the spatial range of the porous medium burner and the preset initial operating parameter combination space; sample in the spatial range through a target sampling method based on the spatial range to obtain candidate training element points; map the candidate training element points to the initial operating parameter combination space to generate a target operating parameter combination space.

[0113] In one embodiment of the present disclosure, the target sampling method is one of the Latin hypercube sampling method, the Sobol method, the Kronecker method, and the randomness method.

[0114] In one embodiment of the present disclosure, the determination module 1220 is further configured to: perform sampling within a limited range of the lean limit value of the target operating condition parameter combination space based on a boundary resampling method to obtain a target training element point.

[0115] In one embodiment of the present disclosure, the determination module 1220 is also used to: obtain the structural parameters of the porous medium burner; establish the gas phase mass conservation equation, the gas phase equation and the gas-solid heat transfer equation based on the structural parameters; and establish a one-dimensional model based on the lean combustion limit value, the gas phase mass conservation equation, the gas phase energy conservation equation and the gas-solid heat transfer equation, i.e., the initial porous medium burner prediction model.

[0116] Therefore, by determining the lean combustion limit value and determining the target training element point from the target operating condition parameter combination space based on the lean combustion limit value, the sampling deviation can be reduced, the accuracy of the model can be improved, and the rapid and accurate prediction of the steady combustion condition parameters corresponding to the target temperature field under the porous medium burner design structure can be achieved. At the same time, it supports the iterative optimization of the porous medium burner structure and operating parameters within the design space, accelerates engineering design, and shortens the design cycle.

[0117] Corresponding to the porous media burner prediction methods provided in the above-mentioned embodiments, an embodiment of the present disclosure also provides a porous media burner prediction device. Since the porous media burner prediction device provided in the embodiment of the present disclosure corresponds to the porous media burner prediction methods provided in the above-mentioned embodiments, the implementation method of the above-mentioned porous media burner prediction method is also applicable to the porous media burner prediction device provided in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0118] Fig.13 is a schematic diagram of a porous media burner prediction device according to one embodiment of the present disclosure, such as Fig.13 As shown, the porous medium burner prediction device 1300 includes: an acquisition module 1310 and an output module 1320.

[0119] The acquisition module 1310 is used to acquire target operating parameters of the porous medium burner to be predicted and a target porous medium burner prediction model.

[0120] The output module 1320 is used to input the target operating parameters into the target porous medium burner prediction model to generate the temperature field spatial distribution of the porous medium burner under the target operating parameters.

[0121] In one embodiment of the present disclosure, the average temperature, maximum temperature and flue gas temperature of the porous medium burner at the target operating parameters are determined based on the spatial distribution of the temperature field.

[0122] In order to implement the above embodiment, the present disclosure further provides an electronic device 1400, Fig.14 is a schematic diagram of an electronic device according to an embodiment of the present disclosure, such as Fig.14 As shown, the electronic device 1400 includes: a processor 1401 and a memory 1402 connected to the processor in communication, the memory 1402 stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor 1401 to implement the present disclosure. Figure 1-Figure 12 The porous media burner prediction model training method of the embodiment, or Fig.13 A porous media burner prediction method according to an embodiment.

[0123] In order to implement the above embodiments, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to implement the above embodiments. Figure 1-Figure 12 The porous media burner prediction model training method of the embodiment, or Fig.13 A porous media burner prediction method according to an embodiment.

[0124] In order to implement the above embodiments, the present disclosure also provides a computer program product, including a computer program. When the computer program is executed by a processor, the computer program implements the above embodiments. Figure 1-Figure 12 The porous media burner prediction model training method of the embodiment, or Fig.13 A porous media burner prediction method according to an embodiment.

[0125] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0126] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0127] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0128] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0129] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0130] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that contains, stores, communicates, propagates or transmits a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0131] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0132] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0133] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0134] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A porous media burner prediction model training method, characterized in that: include: Establishing a target operating parameter combination space of a porous medium burner, and determining a lean combustion limit value of the porous medium burner, wherein the parameter space includes a plurality of element points, each element point includes a combination of operating parameters of the porous medium burner, and the operating parameter combinations in the plurality of element points are different; Determining a target training element point from the target operating condition parameter combination space based on the lean burn limit value, establishing an initial porous medium burner prediction model of the porous medium burner based on the lean burn limit value, and obtaining a real temperature field spatial distribution of the target training element point; Inputting the parameter configuration in the target training element point into the initial porous media burner prediction model to generate a predicted temperature field spatial distribution; The initial porous media burner prediction model is adjusted based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, and the process of determining the target training element point from the parameter space and obtaining the real temperature field spatial distribution of the target training element point and subsequent steps is returned to execute until the training is completed, and the target porous media burner prediction model is output.

2. The method according to claim 1, characterized in that The target operating parameter combination space of the porous medium burner is established, including: Obtaining the spatial range of the porous medium burner and the preset initial operating condition parameter combination space; Based on the spatial range, sampling is performed in the spatial range by a target sampling method to obtain candidate training element points; The candidate training element points are mapped to the initial operating condition parameter combination space to generate a target operating condition parameter combination space.

3. The method according to claim 2, characterized in that The target sampling method is one of a Latin hypercube sampling method, a Sobol method, a Kronecker method, and a randomness method.

4. The method according to any one of claims 1 to 3, characterized in that The determining a target training element point from the target operating condition parameter combination space based on the lean combustion limit value comprises: Based on the boundary resampling method, sampling is performed within a limited range of the lean limit value of the target operating condition parameter combination space to obtain a target training element point.

5. The method according to claim 1, characterized in that The method of establishing an initial porous medium burner prediction model of the porous medium burner based on the lean combustion limit value comprises: Obtaining structural parameters of the porous medium burner; Based on the structural parameters, a gas phase mass conservation equation, a gas phase equation and a gas-solid heat transfer equation are established; A one-dimensional model, namely, the initial porous medium burner prediction model, is established based on the lean combustion limit value, the gas phase mass conservation equation, the gas phase energy conservation equation, and the gas-solid heat transfer equation.

6. A porous media burner prediction method, characterized in that: include: Obtaining target operating parameters to be predicted for a porous medium burner and a target porous medium burner prediction model, wherein the target porous medium burner prediction model is obtained by the porous medium burner prediction model training method according to any one of claims 1 to 5; The target operating parameters are input into the target porous medium burner prediction model to generate a temperature field spatial distribution of the porous medium burner at the target operating parameters.

7. The method according to claim 6, characterized in that The method further comprises: The average temperature, maximum temperature and flue gas temperature of the porous medium burner at the target operating parameters are determined based on the spatial distribution of the temperature field.

8. A porous media burner prediction model training device, characterized in that: include: An establishment module is used to establish a target operating parameter combination space of a porous medium burner, wherein the parameter space includes a plurality of element points, each element point includes a combination of operating parameters of the porous medium burner, and the operating parameter combinations in the plurality of element points are different; A determination module, used to determine a target training element point from the target operating condition parameter combination space, and obtain the real temperature field spatial distribution of the target training element point; An input module, used for inputting the parameter configuration in the target training element point into the initial porous media burner prediction model to generate a predicted temperature field spatial distribution; A training module is used to adjust the initial porous medium burner prediction model based on the real temperature field spatial distribution and the predicted temperature field spatial distribution, return to execute the determination of the target training element point from the parameter space, and obtain the real temperature field spatial distribution of the target training element point and its subsequent steps until the training is completed, and output the target porous medium burner prediction model.

9. A porous media burner prediction device, characterized in that: include: An acquisition module, used for acquiring target operating parameters to be predicted of a porous medium burner and a target porous medium burner prediction model, wherein the target porous medium burner prediction model is obtained by the porous medium burner prediction model training method according to any one of claims 1 to 5; An output module is used to input the target operating parameters into the target porous medium burner prediction model to generate a temperature field spatial distribution of the porous medium burner at the target operating parameters.

10. An electronic device, characterized in that: Including memory and processor; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the porous media burner prediction model training method as described in any one of claims 1 to 5, or the porous media burner prediction method as described in claim 6 or 7.

Images