Method and device for predicting jet flow transition, electronic equipment and storage medium

Abstract

The application relates to the technical field of combustion, in particular to a jet flow transition prediction method and device, an electronic device and a storage medium, wherein the method comprises the following steps: acquiring a characteristic parameter of a jet flame; inputting the characteristic parameter into a preset transition prediction model to obtain a transition Reynolds number and a transition height of the jet flame, so as to perform transition prediction on the jet flame based on the transition Reynolds number and the transition height, wherein the preset transition prediction model is obtained by training a sample data set containing a transition state label. Thus, the problems that transition criteria are not unified, the prediction method has poor universality and is difficult to be popularized and applied under different working conditions in the related art are solved, and the accuracy and applicability of the transition prediction result under different experimental conditions and engineering application scenarios can be improved.

Description

Technical Field

[0001] This application relates to the field of combustion technology, and in particular to a jet transition prediction method, apparatus, electronic device, and storage medium. Background Technology

[0002] Jet flame transition is one of the key technical issues in combustion stability and its control, directly affecting the safe operation, combustion efficiency, and pollutant emission levels of combustion devices. In actual industrial combustion equipment, the fuel jet flame is often in a transition state from laminar to turbulent flow. This process easily induces combustion instability, adversely affecting flame stability, combustion efficiency, and equipment reliability. Establishing an accurate and rapid jet flame transition prediction model has significant engineering application value for combustion device operating status assessment, combustion stability control, and efficient utilization of low-carbon fuels.

[0003] The relevant technologies include the following transition prediction methods: (1) Empirical transition criteria: This method usually determines the transition based on experimental statistical laws or characteristic parameter thresholds. However, it fails to fully consider the physical mechanisms of jet flow instability and combustion coupling processes, has poor adaptability to different fuel compositions and operating conditions, and has limited prediction accuracy. (2) e N Method: This method introduces the theory of flow stability, which can describe the disturbance growth process in the early stage of transition. However, it is based on the assumption of parallel flow and linear stability analysis, which makes it difficult to characterize the nonlinear evolution characteristics in the transition process. Therefore, its applicability is limited under complex jet combustion conditions. (3) Model theory method: Although this type of prediction method based on model decomposition or low-order model can be used for transition simulation and analysis under complex flow conditions, its model parameters are highly dependent and it is difficult to fully reveal the coupling mechanism of flow instability and chemical reaction in the transition process of jet diffusion flame. (4) CFD-based numerical simulation method: This method can predict the flow field variables and transition-related parameters in the whole field. However, it fails to fully consider the influence of buoyancy effect and combustion reaction on flow instability, and it is difficult to accurately capture the large-scale vortex structure outside the transition flame. In addition, due to the limitations of computational cost and model assumptions, the coupling between chemical reaction and fluid instability mechanism is not sufficient, and it is difficult to achieve high-precision prediction of transition characteristics on an engineering scale. (5) Artificial intelligence-based transition prediction method: This method can be used in the fields of turbulent combustion modeling, instability analysis and transition prediction. However, since the transition behavior is affected by a variety of experimental parameters and physical quantities, its mechanism of action is extremely complex, which makes it challenging to achieve accurate and stable prediction. Summary of the Invention

[0004] This application provides a jet transition prediction method, device, electronic device, and storage medium to solve the problems of inconsistent transition criteria, poor universality of prediction methods, and difficulty in promoting and applying them under different working conditions in related technologies. This application can improve the accuracy and engineering applicability of transition prediction results under different experimental conditions and engineering application scenarios.

[0005] The first aspect of this application provides a jet transition prediction method, comprising the following steps: Obtain the characteristic parameters of the jet flame; The feature parameters are input into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and the transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

[0006] Optionally, in some embodiments, before inputting the feature parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the following steps are included: Obtain a sample dataset containing transition state labels; Based on a preset partitioning ratio, the sample dataset is divided into a training set, a validation set, and a test set. Construct a target neural network by inputting the training set into the target neural network for training to obtain initial model parameters; Based on the initial model parameters, the validation set is input into the target neural network for performance evaluation, and the initial model parameters are adjusted according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. Based on the optimal model parameters, the test set is input into the target neural network for model testing, and when the test results meet the preset requirements, the preset transition prediction model is obtained.

[0007] Optionally, in some embodiments, obtaining the sample dataset containing transition state labels includes: Acquire the operational data of the jet flame; The target dataset containing the transition process of jet flame from laminar to turbulent flow is selected from the operational data, and the target dataset is labeled according to the preset transition labeling rules to obtain the sample dataset.

[0008] Optionally, in some embodiments, the characteristic parameters include at least one of the following: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number.

[0009] Optionally, in some embodiments, after inputting the feature parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the process includes: Based on the transition Reynolds number and the transition height, a transition probability range, a combination of transition critical parameters, or a transition risk level are generated to predict the transition of the jet flame based on the transition probability range, the combination of transition critical parameters, or the transition risk level.

[0010] A second aspect of this application provides a jet transition prediction device, comprising: The acquisition module is used to acquire the characteristic parameters of the jet flame; The prediction module is used to input the feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and the transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

[0011] Optionally, in some embodiments, before inputting the feature parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the prediction module includes: The first acquisition unit is used to acquire a sample dataset containing transition state labels; A partitioning unit is used to divide the sample dataset into a training set, a validation set, and a test set based on a preset partitioning ratio. The training unit is used to construct the target neural network and obtain the initial model parameters by inputting the training set into the target neural network for training. An optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. The generation unit is used to input the test set into the target neural network for model testing based on the optimal model parameters, and to obtain the preset transition prediction model when the test results meet the preset requirements.

[0012] Optionally, in some embodiments, the acquisition module includes: The second acquisition unit is used to acquire the operating data of the jet flame; The annotation unit is used to filter out target datasets containing the transition process of jet flame from laminar to turbulent flow from the running data, and to annotate the target datasets according to preset transition annotation rules to obtain the sample dataset.

[0013] Optionally, in some embodiments, the characteristic parameters include at least one of the following: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number.

[0014] Optionally, in some embodiments, after inputting the feature parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the prediction module includes: The prediction unit is used to generate a transition probability range, a combination of critical transition parameters, or a transition risk level based on the transition Reynolds number and the transition height, so as to predict the transition of the jet flame based on the transition probability range, the combination of critical transition parameters, or the transition risk level.

[0015] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the jet transition prediction method as described in the above embodiments.

[0016] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the jet transition prediction method as described in the above embodiments.

[0017] Therefore, by collecting experimental parameters and flame characteristics under different working conditions, a nonlinear mapping relationship between multiple parameters and transition states is constructed using a neural network-based method, forming a unified transition judgment and prediction framework. This enables rapid and accurate prediction of the transition behavior of jet diffusion flames under different working conditions and operating conditions, improving the practicality and reliability of transition identification results in combustion instability analysis and engineering applications.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of the jet transition prediction method provided according to an embodiment of this application; Figure 2 This is a block diagram of a jet transition prediction device provided according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0021] Before introducing the jet transition prediction method of the embodiments of this application, let's first introduce the difficulties in jet transition prediction.

[0022] Understandably, in the application of jet transition prediction, the transition process of jet diffusion flame is affected by a variety of experimental parameters and physical quantities, such as wake conditions, fuel outlet pipe diameter, fuel viscosity, and buoyancy effects, resulting in significant differences in transition behavior under different operating conditions. Model training relies on specific experimental conditions and datasets, and the transition determination process often requires the introduction of manually set criteria or empirical thresholds, making it difficult to achieve a unified representation across different experimental conditions.

[0023] This leads to a lack of comparability in transition prediction results obtained from different studies, making it difficult to form universally applicable prediction conclusions with engineering guidance. At the engineering application level, the inconsistency in transition criteria makes it difficult to establish a stable correspondence between the transition state of the jet flame and actual combustion instability, thus hindering the effective implementation of combustion device operation status assessment, stability control, and operating condition optimization decisions. Especially under multi-condition and wide-range operating conditions, the inconsistency in transition prediction results can easily lead to biases in judging combustion stability margins, making it difficult to coordinate combustion instability risk assessment and control strategies, and affecting the safety and reliability of the combustion system.

[0024] The reason for this is that the transition process of jet diffusion flames is itself a complex physical phenomenon involving multi-parameter coupling, nonlinear evolution, and multi-scale characteristics. However, when determining the transition state, related technologies mostly rely on local flow characteristics, single physical quantities, or empirical rules under specific operating conditions. A unified criterion that can universally describe the transition behavior under different experimental conditions has not yet been established. This mismatch between the extreme complexity of transition physics and the relative simplicity of the determination methods is the fundamental problem that leads to the insufficient generalization ability of existing prediction models and makes it difficult to promote their application in engineering and industrial scenarios.

[0025] In summary, jet transition prediction methods lack unified and clear transition judgment criteria and generalizable modeling basis. The transition identification process is highly dependent on specific experimental conditions and empirical thresholds, making it difficult to accurately and stably reflect the transition behavior of jet diffusion flames under different conditions, which restricts the promotion and application of prediction methods in engineering and industrial scenarios.

[0026] To address the aforementioned issues, this application provides a jet transition prediction method. In this method, by collecting experimental operating parameters and flame characteristic quantities, a nonlinear mapping relationship between multiple parameters and transition states is constructed, forming a unified transition prediction and judgment framework. This method can achieve stable prediction of the transition behavior of jet diffusion flames under different experimental conditions and operating conditions, providing reliable technical support for combustion instability analysis and control.

[0027] Specifically, Figure 1 This is a schematic flowchart of a jet transition prediction method provided in an embodiment of this application.

[0028] like Figure 1 As shown, the jet transition prediction method includes the following steps: In step S101, the characteristic parameters of the jet flame are obtained.

[0029] The characteristic parameters include at least one of the following: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number.

[0030] Specifically, by comprehensively considering the fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number, an input feature set capable of reversing the transition mechanism of the wake diffusion flame can be constructed, which can improve the stability and generalization ability of transition prediction under different operating conditions and engineering application scenarios.

[0031] It should be noted that, in addition to using the fuel outlet Reynolds number, nozzle diameter, fuel physical properties, wake conditions and Froude number, characteristic parameters such as Richardson number, Grashof number, jet momentum ratio or equivalent dimensionless combination parameters can also be introduced to construct characteristic parameters for characterizing jet diffusion.

[0032] In step S102, the feature parameters are input into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

[0033] Understandably, the prediction of jet flame transition is difficult to apply in engineering scenarios. One of the core reasons is that the transition process of jet diffusion flames is influenced by a variety of experimental parameters and physical quantities. For example, the mechanisms of action of factors such as wake conditions, fuel outlet pipe diameter, fuel viscosity, and buoyancy effects are complex, making it difficult to achieve accurate and stable prediction of transition behavior. To address this issue, different studies often differ significantly in the selection of experimental conditions and transition judgment criteria, resulting in a lack of comparability and universality in the prediction methods. This inconsistency in criteria has become a major obstacle to the widespread application of jet diffusion flame transition prediction methods in engineering and industrial scenarios.

[0034] To address the aforementioned issues, this application proposes a jet transition prediction method to overcome the shortcomings of inconsistent transition criteria and the difficulty in applying prediction methods across different operating conditions. Based on experimental data, a machine learning model is used to train the nonlinear mapping relationship between experimental operating condition characteristic parameters and transition states, thereby predicting the transition behavior of jet diffusion flames. Compared with transition prediction methods based on a single operating condition, this method can more accurately capture transition criteria under different operating conditions, improve the accuracy and engineering applicability of transition prediction results under different experimental conditions and engineering application scenarios, and thus enhance the guiding significance of transition identification results for actual combustion instability.

[0035] Optionally, in some embodiments, before inputting the feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the method includes: acquiring a sample dataset containing transition state labels; dividing the sample dataset into a training set, a validation set, and a test set based on a preset partitioning ratio; constructing a target neural network, inputting the training set into the target neural network for training to obtain initial model parameters; based on the initial model parameters, inputting the validation set into the target neural network for performance evaluation, and adjusting the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters; based on the optimal model parameters, inputting the test set into the target neural network for model testing, and obtaining the preset transition prediction model when the test results meet preset requirements.

[0036] Furthermore, in some embodiments, obtaining a sample dataset containing transition state labels includes: obtaining the operation data of the jet flame; filtering out a target dataset containing the transition process of the jet flame from laminar to turbulent flow from the operation data; and labeling the target dataset according to a preset transition labeling rule to obtain a sample dataset.

[0037] The preset transition annotation rules include: manual unified annotation, automatic annotation method based on statistical feature mutation, threshold determination method or semi-supervised learning method, etc.

[0038] Specifically, the embodiments of this application can establish a preset transition prediction model through the following steps.

[0039] (1) Data collection.

[0040] This application embodiment can collect jet diffusion flame related data from publicly available experimental literature, databases or technical reports, including: flame operation data under different fuel types, nozzle geometry parameters, fuel physical property parameters, wake conditions and environmental parameters, and screen out effective data samples containing the transition process of jet diffusion flame from laminar to turbulent flow to construct the original dataset of jet diffusion flame transition.

[0041] Training and validating models based on existing literature and publicly available experimental data can reduce the implementation cost and application threshold of transition prediction methods, and improve their scalability in engineering design, operation analysis and combustion instability assessment.

[0042] (2) Data is labeled uniformly.

[0043] Based on the flame state descriptions, transition criteria, or experimental conclusions given in the literature, this application embodiment organizes and classifies the corresponding flame states in the original dataset, and standardizes the annotation of data from different literature sources according to the preset unified transition annotation rules, forming a sample dataset containing transition state labels.

[0044] It should be noted that, in addition to manual unified annotation based on literature conclusions, the embodiments of this application may also use automatic annotation methods based on statistical feature mutations, threshold determination methods, or semi-supervised learning methods to annotate the flame transition state. The common point is that they all use unified rules to standardize the transition state of different data sources.

[0045] (3) Feature parameter extraction and standardization processing.

[0046] This application embodiment can extract characteristic parameters related to the transition of jet diffusion flame from the sample dataset, including: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number, etc. The parameter forms used in different literatures are converted and standardized, and the characteristic parameters are dimensionless or normalized to eliminate the influence of differences in data sources on model training.

[0047] (4) Sample selection and dataset partitioning.

[0048] The embodiments of this application can perform validity screening on the processed sample data, removing data that is missing key parameters or does not meet the transition recognition requirements; then the sample data is divided into training set, validation set and test set according to a preset ratio of 8:1:1 to ensure the generalization ability of the model under different data source conditions.

[0049] (5) Neural network model construction.

[0050] A neural network model for predicting the transition of jet-diffused flames is constructed. The neural network model includes an input layer, at least one hidden layer, and an output layer. The nodes in the input layer correspond to feature parameters, and the output layer is used to output the transition Reynolds number and transition height of the jet-diffused flame.

[0051] In the embodiments of this application, in addition to using a feedforward neural network, machine learning models such as convolutional neural networks, recurrent neural networks, support vector machines, random forests, or gradient boosting trees can also be used. The common feature of these models is that they can all establish a nonlinear mapping relationship between operating parameters and the transition characteristics of jet diffusion flame.

[0052] (6) Model training and optimization.

[0053] The training set samples are input into the neural network model, and the model parameters are trained using the backpropagation algorithm. The model hyperparameters are then adjusted using the validation set to reduce prediction errors and improve the model's adaptability to different literature data.

[0054] In the embodiments of this application, in addition to training based on the backpropagation algorithm, genetic algorithms, particle swarm optimization, or Bayesian optimization methods can also be used to optimize the model parameters. The common point is that they all aim to reduce prediction errors and improve the model's generalization ability.

[0055] (7) Turning point prediction and judgment.

[0056] The test set data or the data of the working conditions to be predicted are input into the trained neural network model, which outputs the prediction results of the transition of the jet diffusion flame, and judges whether the flame has transitioned according to the unified transition judgment criteria.

[0057] (8) Model applicability verification steps.

[0058] The embodiments of this application evaluate the prediction results through cross-validation between data from different literature sources, in order to verify the accuracy, stability and engineering applicability of the method under multiple working conditions and multiple data sources.

[0059] Therefore, this application embodiment constructs a multi-parameter feature system by uniformly screening, standardizing parameters, and annotating transition data from multiple sources of literature and public data. Based on a neural network, it establishes a nonlinear mapping relationship between experimental working condition parameters and the transition Reynolds number and transition height of the jet diffusion flame, forming a unified and comparable transition prediction and judgment framework.

[0060] Optionally, in some embodiments, after inputting the feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the method includes: generating a transition probability range, a combination of transition critical parameters, or a transition risk level based on the transition Reynolds number and transition height, so as to perform transition prediction on the jet flame based on the transition probability range, the combination of transition critical parameters, or the transition risk level.

[0061] Specifically, in addition to outputting the transition Reynolds number and transition height, the embodiments of this application can also output the transition probability range, the combination of transition critical parameters, or the transition risk level. The common point is that they are all used to quantitatively or semi-quantitatively characterize the transition behavior of the jet diffusion flame.

[0062] In summary, the embodiments of this application introduce a jet diffusion flame transition prediction method based on multi-document data fusion and unified annotation, which solves the problems of inconsistent transition judgment criteria and incomparable prediction results under different operating conditions and data sources in existing studies. Combined with the systematic extraction and unification of multiple parameters such as fuel properties, nozzle geometry, wake conditions and experimental conditions, the model can more comprehensively characterize the key influencing factors in the jet diffusion flame transition process, and improve the stability and accuracy of transition prediction results under different operating conditions.

[0063] Furthermore, this application embodiment achieves unified modeling of the complex relationship between experimental operating condition characteristic parameters and transition Reynolds number and transition height by constructing a nonlinear mapping model with neural network as its core. Compared with prediction methods that rely on a single empirical criterion or local operating conditions, it can more accurately capture the transition characteristics under different operating conditions, enhancing the guiding significance of the prediction results for actual combustion instability analysis. In addition, training samples are constructed based on existing literature and publicly available data, eliminating the need for additional experimental equipment and online measurement systems, thus reducing implementation costs and application barriers. Moreover, cross-validation through multiple data sources improves the model's generalization ability and engineering applicability under multiple operating conditions and scales, which is conducive to its widespread application in actual combustion equipment design, operation, and safety assessment scenarios.

[0064] According to the jet transition prediction method proposed in this application, characteristic parameters of the jet flame are obtained; the characteristic parameters are input into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and transition height. The preset transition prediction model is trained from a sample dataset containing transition state labels. This solves the problems of inconsistent transition criteria, poor universality of prediction methods, and difficulty in promoting application under different working conditions in related technologies. This application can improve the accuracy and applicability of transition prediction results under different experimental conditions and engineering application scenarios.

[0065] Next, the jet transition prediction device according to the embodiments of this application is described with reference to the accompanying drawings.

[0066] Figure 2 This is a block diagram of the jet transition prediction device according to an embodiment of this application.

[0067] like Figure 2 As shown, the jet transition prediction device 10 includes an acquisition module 100 and a prediction module 200.

[0068] The acquisition module 100 is used to acquire the characteristic parameters of the jet flame.

[0069] The prediction module 200 is used to input feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

[0070] Optionally, in some embodiments, before inputting the feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the prediction module 200 includes: a first acquisition unit, a division unit, a training unit, an optimization unit, and a generation unit.

[0071] The first acquisition unit is used to acquire a sample dataset containing transition state labels.

[0072] The partitioning unit is used to divide the sample dataset into training, validation, and test sets based on a preset partitioning ratio.

[0073] The training unit is used to build the target neural network. It obtains the initial model parameters by inputting the training set into the target neural network for training.

[0074] The optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters.

[0075] The generation unit is used to input the test set into the target neural network based on the optimal model parameters to test the model, and when the test results meet the preset requirements, obtain the preset transition prediction model.

[0076] Optionally, in some embodiments, the acquisition module 100 includes a second acquisition unit and a labeling unit.

[0077] The second acquisition unit is used to acquire the operating data of the jet flame.

[0078] The annotation unit is used to filter out target datasets containing the transition process of jet flame from laminar to turbulent flow from the running data, and to annotate the target datasets according to the preset transition annotation rules to obtain sample datasets.

[0079] Optionally, in some embodiments, the characteristic parameters include at least one of the following: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number.

[0080] Optionally, in some embodiments, after inputting the characteristic parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the prediction module 200 includes: a prediction unit.

[0081] The prediction unit is used to generate a transition probability range, a combination of critical transition parameters, or a transition risk level based on the transition Reynolds number and the transition height, so as to predict the transition of the jet flame based on the transition probability range, the combination of critical transition parameters, or the transition risk level.

[0082] It should be noted that the foregoing explanation of the jet transition prediction method embodiment also applies to the jet transition prediction device of this embodiment, and will not be repeated here.

[0083] According to the jet transition prediction device proposed in this application, characteristic parameters of the jet flame are obtained; the characteristic parameters are input into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and transition height. The preset transition prediction model is trained from a sample dataset containing transition state labels. This solves the problems of inconsistent transition criteria, poor universality of prediction methods, and difficulty in promoting application under different working conditions in related technologies. This application can improve the accuracy and applicability of transition prediction results under different experimental conditions and engineering application scenarios.

[0084] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 301, the processor 302, and the computer program stored on the memory 301 and capable of running on the processor 302.

[0085] When the processor 302 executes the program, it implements the jet transition prediction method provided in the above embodiments.

[0086] Furthermore, electronic devices also include: Communication interface 303 is used for communication between memory 301 and processor 302.

[0087] The memory 301 is used to store computer programs that can run on the processor 302.

[0088] The memory 301 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0089] If the memory 301, processor 302, and communication interface 303 are implemented independently, then the communication interface 303, memory 301, and processor 302 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0090] Optionally, in a specific implementation, if the memory 301, processor 302, and communication interface 303 are integrated on a single chip, then the memory 301, processor 302, and communication interface 303 can communicate with each other through an internal interface.

[0091] Processor 302 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0092] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described jet transition prediction method.

[0093] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0094] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0095] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0096] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0097] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0098] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for predicting jet transition, characterized in that, Includes the following steps: Obtain the characteristic parameters of the jet flame; The feature parameters are input into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and the transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

2. The method according to claim 1, characterized in that, Before inputting the characteristic parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the process includes: Obtain a sample dataset containing transition state labels; Based on a preset partitioning ratio, the sample dataset is divided into a training set, a validation set, and a test set. Construct a target neural network by inputting the training set into the target neural network for training to obtain initial model parameters; Based on the initial model parameters, the validation set is input into the target neural network for performance evaluation, and the initial model parameters are adjusted according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. Based on the optimal model parameters, the test set is input into the target neural network for model testing, and when the test results meet the preset requirements, the preset transition prediction model is obtained.

3. The method according to claim 1, characterized in that, The process of obtaining a sample dataset containing transition state labels includes: Acquire the operational data of the jet flame; The target dataset containing the transition process of jet flame from laminar to turbulent flow is selected from the operational data, and the target dataset is labeled according to the preset transition labeling rules to obtain the sample dataset.

4. The method according to claim 1, characterized in that, The characteristic parameters include at least one of the following: fuel outlet Reynolds number, nozzle diameter, fuel viscosity, density, wake velocity ratio, wake temperature, pressure, combustion rate, and Froude number.

5. The method according to claim 1, characterized in that, After inputting the characteristic parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the process includes: Based on the transition Reynolds number and the transition height, a transition probability range, a combination of transition critical parameters, or a transition risk level are generated to predict the transition of the jet flame based on the transition probability range, the combination of transition critical parameters, or the transition risk level.

6. A jet transition prediction device, characterized in that, include: The acquisition module is used to acquire the characteristic parameters of the jet flame; The prediction module is used to input the feature parameters into a preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, so as to predict the transition of the jet flame based on the transition Reynolds number and the transition height. The preset transition prediction model is trained by a sample dataset containing transition state labels.

7. The apparatus according to claim 6, characterized in that, Before inputting the characteristic parameters into the preset transition prediction model to obtain the transition Reynolds number and transition height of the jet flame, the prediction module includes: The first acquisition unit is used to acquire a sample dataset containing transition state labels; A partitioning unit is used to divide the sample dataset into a training set, a validation set, and a test set based on a preset partitioning ratio. A training unit is used to construct a target neural network, and to obtain initial model parameters by inputting the training set into the target neural network for training. An optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. The generation unit is used to input the test set into the target neural network for model testing based on the optimal model parameters, and to obtain the preset transition prediction model when the test results meet the preset requirements.

8. The apparatus according to claim 6, characterized in that, The acquisition module includes: The second acquisition unit is used to acquire the operating data of the jet flame; The annotation unit is used to filter out target datasets containing the transition process of jet flame from laminar to turbulent flow from the running data, and to annotate the target datasets according to preset transition annotation rules to obtain the sample dataset.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the jet transition prediction method as described in any one of claims 1-5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the jet transition prediction method as described in any one of claims 1-5.

Images