Space combination elbow wear prediction method, system, equipment and medium

Abstract

The invention discloses a space combination elbow wear prediction method, system and device and a medium. The method comprises the steps that operation parameter information of an elbow is acquired; the operation parameter information of the elbow is input into the trained GPR model, the abrasion loss of the space combination elbow is obtained, the GPR model is trained based on a data set obtained through CFD-DEM coupling simulation, the method, the system, the equipment and the medium can predict abrasion of the space combination elbow, and the method, the system, the equipment and the medium have the advantages of being accurate, reliable and rapid.

Description

Technical Field

[0001] The invention belongs to the field of boiler flue gas and coal powder pipeline wear, and relates to a method, system, equipment and medium for predicting wear of a spatial combined elbow. Background Art

[0002] In order to reduce fuel costs, thermal power plants often burn low-quality coal, and a large amount of fly ash particles are produced after the combustion of pulverized coal. The high-temperature flue gas and fly ash particles in the tail flue of the boiler of a coal-fired power plant will form a gas-solid two-phase flow; as the flow velocity of the ash-containing flue gas and the fly ash particles increase, the wear effect on the flue and the wall of the heated tube, the denitration system guide plate, and the surface of the denitration system catalyst increases. . In the layout process of the smoke, air and pulverized coal pipeline of the coal-fired unit, it is often necessary to make a spatial turn. At this time, two elbows need to be connected to form a special spatial combination elbow structure. The spatial combination elbow structure has a limited connection length between the elbows, resulting in coupling between the elbows, making the gas-solid two-phase flow and wear characteristics in the spatial combination elbow different from those of a single elbow component. The wear of these components will affect the normal operation of the power station boiler and cause a significant increase in the boiler maintenance cost.

[0003] As the energy industry places increasing demands on equipment efficiency and safety, it is particularly important to accurately predict boiler tube wear. Existing wear prediction results mostly rely on operational experience models and flow field numerical simulation calculations, but the above prediction methods have certain limitations:

[0004] 1) When coal quality, component structure, and operating parameters change, a large number of experimental tests are required, and the establishment of different empirical models requires a large research cost;

[0005] 2) Research shows that computational fluid dynamics and discrete element method (CFD-DEM) coupling is usually used to calculate gas-solid two-phase flow. The CFD-DEM coupling method is more accurate for calculating the wear of flue and heating surface. However, the boiler flue and internal components are relatively complex, and the modeling and calculation cost of a large number of different structural parameters and operating parameters is high, and the calculation cycle is long.

[0006] 3) The wear empirical model obtained from the current experiment is relatively complete, the CFD-DEM numerical simulation calculation accuracy is good, and the machine learning prediction method is gradually applied to industrial production. Among them, the Gaussian process regression (GPR) algorithm can be used for modeling and prediction with a limited data set, but there are few related studies.

[0007] Therefore, it is necessary to establish an accurate, reliable and fast solid particle erosion prediction model by coupling gas-solid flow simulation data with the GPR algorithm. Developing a method to accurately predict the wear of boiler flue and internal components is of great significance for taking maintenance measures in advance, reducing downtime, reducing economic losses and preventing safety accidents. Summary of the invention

[0008] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method, system, device and medium for predicting the wear of a spatial combination elbow. The method, system, device and medium can predict the wear of a spatial combination elbow and are accurate, reliable and fast.

[0009] To achieve the above object, the present invention discloses a method for predicting wear of a spatial combined elbow, comprising:

[0010] Obtaining the operating parameter information of the elbow;

[0011] The operating parameter information of the elbow is input into the trained GPR model to obtain the wear amount of the spatial combined elbow, wherein the GPR model is trained based on a data set obtained by CFD-DEM coupling simulation.

[0012] Furthermore, before inputting the operation parameter information of the elbow into the trained GPR model, the method further includes:

[0013] Establish a three-dimensional model of a spatial combined elbow;

[0014] Perform simulation based on CFD-DEM coupling and construct a sample set based on the simulation results;

[0015] A GPR model is constructed, and the GPR model is trained based on the sample set to obtain a trained GPR model.

[0016] Furthermore, the operation parameters of the elbow may include but are not limited to: surface flow rates of the gas phase and liquid phase in the elbow, particle size and liquid viscosity.

[0017] Furthermore, based on the sample set, the GPR model is trained using a conjugate gradient method.

[0018] Furthermore, during the training of the GPR model based on the sample set, a five-fold cross validation method is used to evaluate the performance of the GPR model.

[0019] The present invention discloses a spatial combined elbow wear prediction system, which is characterized by comprising:

[0020] An acquisition module is used to obtain operating parameter information of the elbow;

[0021] The prediction module is used to input the operating parameter information of the elbow into the trained GPR model to obtain the wear amount of the spatial combined elbow, wherein the GPR model is trained based on the data set obtained by CFD-DEM coupling simulation.

[0022] Furthermore, it also includes:

[0023] Establish a module for establishing a three-dimensional model of a spatial combined elbow;

[0024] The simulation module is used to simulate based on CFD-DEM coupling and build a sample set according to the simulation results;

[0025] The training module is used to construct a GPR model and train the GPR model based on a sample set to obtain a trained GPR model.

[0026] Furthermore, the operation parameters of the elbow may include but are not limited to: surface flow rates of the gas phase and liquid phase in the elbow, particle size and liquid viscosity.

[0027] The present invention discloses a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the spatial combined elbow wear prediction method are implemented.

[0028] The present invention discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the spatial combined elbow wear prediction method are implemented.

[0029] The present invention has the following beneficial effects:

[0030] During specific operation, the method, system, device and medium for predicting wear of the spatial combination elbow described in the present invention input the operating parameter information of the elbow into the trained GPR model to obtain the wear amount of the spatial combination elbow, so as to realize the wear prediction of the spatial combination elbow with high accuracy and reliability. In addition, the GPR model described in the present invention is trained based on the data set obtained by CFD-DEM coupling simulation, and has a fast prediction speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is the flow chart of the CFD-DEM coupling calculation model;

[0032] Figure 2 This is the training flow chart of the GPR model;

[0033] Figure 3 It is a learning curve diagram in Example 2;

[0034] Figure 4 It is a scatter plot comparing the predicted value and the experimental value in Example 2. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0038] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0039] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0040] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0041] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0042] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0043] Embodiment 1

[0044] refer to Figure 1 and Figure 2 The wear prediction method of the spatial combined elbow of the present invention comprises the following steps:

[0045] 1) Calculate wall wear based on CFD-DEM method;

[0046] The commercial software ANSYS-FLUENT and Altair EDEM are used to solve the problem, and FLUENT and EDEM are used to iteratively solve the flow field and particle motion to achieve the momentum balance between the two phases. The information exchange between the coupled EDEM and Fluent software is realized through user-defined functions (UDF). First, the gas phase flow field is calculated within a time step. After convergence, the EDEM software obtains the flow field data, calculates the interaction force between the two phases, and then determines the position, force and velocity of the particles. At the same time, the particle information is updated to the gas phase calculation, and the energy transfer of the particles is represented by adding a momentum source term. Then, the flow field and particle motion are iteratively solved.

[0047] 2) Collect and preprocess the wear data;

[0048] Obtain wear test data of boiler flue gas duct and CFD-DEM simulation data, including key parameters such as flue gas flow rate, particle concentration, particle diameter and impact frequency. For gas-liquid-solid three-phase flow, it is also necessary to collect liquid phase related parameters such as liquid flow rate and viscosity. Normalize the collected data and screen out characteristic variables related to wear.

[0049] It is known that the particle components in the boiler flue and air duct space combination elbow mainly come from the incompletely burned coal ash or other fuel residues during the combustion process, such as silicates, aluminates and other inorganic minerals; carbon black and coke particles; sulfate particles; such as iron, vanadium and nickel oxides; some incompletely burned fuels will produce organic particles, so the main parameters affecting the pipe wear are particle size, particle aspect ratio, particle flow rate, particle hardness, particle pH, flue and air duct temperature, flue and air duct characteristic diameter, characteristic length and characteristic angle, etc.

[0050] 3) Construct a Gaussian process regression model;

[0051] Gaussian process regression assumes that the output variable is a function of some input variables. For the prediction of smoke and air duct wear, the input variable X and the output variable Y can be defined as:

[0052]

[0053] The core of GPR is the covariance function k(x,x').

[0054] The characteristics of the Gaussian process are completely determined by its mean function m(x) and covariance function k(x,x'). For simplicity, the mean function m(x) is usually set to 0. The covariance function describes the similarity between input features. In wear prediction, the covariance function helps the model understand the similarity of smoke and air duct wear under different conditions (such as flow rate, particle size, material type, etc.). The use of the covariance function is crucial to understanding the mutual influence between data points and is the key to model learning and prediction in GPR. The covariance function uses the widely used square exponential function:

[0055]

[0056] Each y i The value is considered as a function f(x i ) and the random variable ε representing the noise i The sum of i =f(x i )+ε i .

[0057] Assume the noise has zero mean and Gaussian distribution of variance: Considering the noise distribution, the covariance function can be written as:

[0058]

[0059] in, It represents the maximum amplitude of the covariance, and l is the length parameter, which is used to determine the similarity of two input feature quantities.

[0060] Calculate all different combinations of training data points to get the covariance matrix:

[0061]

[0062] The covariance between the new input sample and the training sample and the covariance of the new input sample variable can be written as:

[0063] K'=[k(x * ,x1)k(x * ,x2)…k(x * ,x n )]

[0064] K″=k(x * ,x * )

[0065] The predicted value of the new input sample is: Y'=K'K -1 Y,

[0066] Assume the noise has zero mean and Gaussian distribution of variance.

[0067] The target value for the new input is predicted using the formula y * =K * K -1 y, where K is the covariance matrix of the new data input, K -1 is the inverse of the covariance matrix of the training data.

[0068] 4) Model training and optimization;

[0069] Use part of the data to train the GPR model and optimize the hyperparameters by maximizing the log-likelihood function:

[0070]

[0071]

[0072] Among them, Θ is the hyperparameter vector, which is optimized by the conjugate gradient method. The remaining data is used to test the prediction performance of the GPR model, and the evaluation indicators such as the accuracy, precision and recall of the model are calculated.

[0073] 5) Model evaluation;

[0074] The performance of the GPR model was evaluated using a five-fold cross validation method. The root mean square error (RMSE) of each test subset was calculated, and the average of the five RMSEs was taken as the model performance indicator.

[0075]

[0076] Among them, y i is the actual value, y' i is the predicted value, and n is the number of test data points.

[0077] 6) Learning curve analysis;

[0078] By adjusting the size of the training set (such as using 20%, 40%, 60%, and 80% of the data), drawing a learning curve, analyzing the impact of the training set size on model performance, and determining the optimal training set size, we can minimize the computational complexity while ensuring model performance.

[0079] 7) Parameter impact analysis;

[0080] The one-dimensional GPR model is used to analyze the impact of individual parameters on wear, such as flow rate and particle size. This helps to understand the contribution of each parameter to wear, provides a basis for equipment optimization and operating parameter adjustment, and uses the remaining data to test the predictive performance of the GPR model and calculate evaluation indicators such as the model's accuracy, precision, and recall.

[0081] 8) Dimensionless processing;

[0082] Dimensionless parameters such as Reynolds number (Re) and velocity ratio are introduced to establish a GPR model based on dimensionless parameters. The performance of the GPR model is compared with that of the original parameter model, and the influence of dimensionless processing on model performance and computational efficiency is discussed.

[0083] For gas-solid two-phase flow, the penetration number, Pnr, can be introduced:

[0084]

[0085] in, d pref is the reference particle size.

[0086] 9) Wear prediction;

[0087] A GPR model with multiple outputs was developed to predict the wear of different positions of the smoke and air duct. The input parameters of the model include the surface flow velocity of the gas and liquid phases, particle size, and liquid viscosity, and the output of the model is the wear rate of different positions of the smoke and air duct.

[0088] 10) Model application;

[0089] The verified GPR model is deployed in the actual boiler operation monitoring system to achieve real-time prediction and early warning of wear.

[0090] Embodiment 2

[0091] This embodiment includes the following steps:

[0092] 1) CFD-DEM coupled simulation data generation;

[0093] 11) Use ANSYS Design Modeler to construct the three-dimensional model of the spatial combined elbow mentioned in the "Technical Specifications for the Design of Smoke, Air and Pulverized Coal Pipes in Thermal Power Plants". Use ANSYS Meshing for meshing, use hexahedral structured meshes, and encrypt the mesh near the wall to ensure that the Yplus value is around 30. The total number of meshes needs to reach the million level.

[0094] 12) Set the turbulence model in ANSYS Fluent software, enable the discrete phase model, and select the enhanced wall function.

[0095] 13) DEM model setting: The particle shape, particle-particle contact model (Hertz-Mindlin with no slipoption) and particle-wall contact model (Hertz-Mindlin with JKRCohesion) were set in the Altair EDEM software.

[0096] 14) Set up the coupling between fluid and particle simulation, and realize the bidirectional coupling between ANSYS Fluent software and Altair EDEM software by embedding user-defined functions (UDF) in Fluent software. Select an appropriate time step to exchange the DEM model calculation data with the CFD fluid calculation data.

[0097] 15) Set boundary conditions and physical parameters according to different combustion operation conditions of the boiler. Use velocity inlet to set gas phase velocity and particle velocity respectively; give pressure data of elbow outlet according to operation measurement data, set no-slip boundary conditions of the wall, set gas phase density viscosity and particle density and particle size range.

[0098] 16) Wear calculation settings, using the Lagrangian method to track and solve each particle, so as to obtain relevant information such as the speed, angle and position when the particle collides with the wall of the curved pipe, and then combined with the wear calculation interface provided by the Fluent software, the wear amount is solved according to the wear calculation model.

[0099] 17) Perform numerical simulation optimization based on the benchmark test conditions and results, select appropriate models and simulation settings, ensure that the error between the numerical simulation calculation results and the benchmark test condition results is within ±5%, and select typical conditions for simulation.

[0100] 2) Data collection and preprocessing;

[0101] 21) Extract the following data from the CFD-DEM simulation results. The input variables include d p is the particle size, λ p is the particle aspect ratio, V p is the particle velocity, H p is particle hardness, pH p is the particle pH value, T is the flue gas temperature, D is the characteristic diameter of the flue gas duct elbow, L is the characteristic length of the combined elbow, θ is the characteristic angle, Reynolds number Re, Stokes number St, and penetration rate number Pnr. The output variable is defined as wear rate / wear amount W.

[0102] 22) Data preprocessing: Perform outlier detection, use the 3σ criterion to identify and process outliers; perform data normalization, use the Min-Max normalization method to scale all features to the [0,1] interval. Use the recursive feature elimination (RFE) algorithm to select the most relevant features.

[0103] 3) Gaussian process regression model construction;

[0104] 31) Covariance function selection:

[0105]

[0106] 32) Considering the measurement noise, the output is modeled as: y i =f(x i )+ε i ,ε~N(0,σ n 2).

[0107] 33) Prior distribution setting, assuming that the function f(x i ) follows a Gaussian process prior with zero mean:

[0108] f(x)~GP(0,k(x,x'))

[0109] 34) Posterior distribution inference: Given training data D and test input x*, infer the posterior distribution of the test output y*: y|D,x * ~N(μ * ,σ *2 ).

[0110] Where μ*=k T (K+σ n2I)-1y;σ*2=k(x,x*)-k T (K+σ n 2I)-1k.

[0111] 4) Model training and optimization;

[0112] 41) Dataset division: The dataset is divided into training set and test set with a ratio of 80% and 20% respectively.

[0113] 42) Hyperparameter optimization, using Bayesian optimization method to optimize the hyperparameters of the GPR model, including: optimizing the hyperparameters by maximizing the log-likelihood function:

[0114]

[0115] Θ={σ f ,σ n ,l}

[0116] 43) Carry out model training and use the optimized hyperparameters to train the GPR model on the training set.

[0117] 5) Model evaluation and validation;

[0118] 51) Calculate the following performance indicators: root mean square error (RMSE); coefficient of determination (R 2 ); mean absolute error (MAE); maximum absolute error (MaxAE). 2) Cross-validation: use the five-fold cross-validation method to evaluate the generalization ability of the model. Draw a learning curve to analyze the impact of the training set size on the model performance.

[0119] 6) Parameter sensitivity analysis;

[0120] 61) Global sensitivity analysis, using the Sobol method to calculate the global sensitivity index of each input parameter to the wear rate. 62) Local sensitivity analysis, calculating the partial derivatives , analyze the local influence of each parameter on the wear rate.

[0121] 7) Local wear distribution prediction;

[0122] 71) Multi-output GPR model, develop a GPR model with multiple outputs to predict the wear rate at different positions of the spatial combination elbow. 72) Wear hotspot identification, based on the prediction results, identify the wear hotspot area and calculate its area and average wear rate.

[0123] 8) Model application and system integration;

[0124] 81) Real-time prediction module, encapsulate the trained GPR model into a dynamic link library (DLL) and integrate it into the boiler control system; 82) Data interface design, design a standardized data interface for real-time acquisition of operating parameters and input into the GPR model; 83) Visual interface, develop a Qt-based graphical user interface to display wear prediction results and 3D wear distribution maps in real time; 84) Set multi-level wear warning thresholds, and trigger an alarm when the predicted wear rate exceeds the threshold. 5) Based on the wear prediction results and combined with the equipment historical data, develop a maintenance decision support algorithm to provide suggestions for formulating the optimal maintenance strategy.

[0125] Through the above steps, a spatial combined elbow wear prediction system based on CFD-DEM coupling simulation and Gaussian process regression was realized. This system can not only accurately predict the wear under different working conditions, but also provide strong support for equipment maintenance decisions, effectively improving boiler operation efficiency and safety.

[0126] Embodiment 3

[0127] The spatial combined elbow wear prediction system of the present invention is characterized by comprising:

[0128] An acquisition module is used to obtain operating parameter information of the elbow;

[0129] The prediction module is used to input the operating parameter information of the elbow into the trained GPR model to obtain the wear amount of the spatial combined elbow, wherein the GPR model is trained based on the data set obtained by CFD-DEM coupling simulation.

[0130] Furthermore, it also includes:

[0131] Establish a module for establishing a three-dimensional model of a spatial combined elbow;

[0132] The simulation module is used to simulate based on CFD-DEM coupling and build a sample set according to the simulation results;

[0133] The training module is used to construct a GPR model and train the GPR model based on a sample set to obtain a trained GPR model.

[0134] Furthermore, the operation parameters of the elbow may include but are not limited to: surface flow rates of the gas phase and liquid phase in the elbow, particle size and liquid viscosity.

[0135] The division of modules in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present application may be integrated into a processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0136] Embodiment 4

[0137] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the spatial combination elbow wear prediction method are implemented. The memory may include a memory, such as a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk memory, etc. The processor, the network interface, and the memory are interconnected through an internal bus. The internal bus may be an industrial standard architecture bus, a peripheral component interconnection standard bus, an extended industrial standard structure bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program may include a program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0138] Embodiment 5

[0139] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the spatial combination elbow wear prediction method are implemented. Specifically, the computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include a read-only memory (ROM), a hard disk, a flash memory, an optical disk, a magnetic disk, etc.

[0140] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting wear of a spatial combined elbow, characterized in that: include: Obtaining the operating parameter information of the elbow; The operating parameter information of the elbow is input into the trained GPR model to obtain the wear amount of the spatial combined elbow, wherein the GPR model is trained based on a data set obtained by CFD-DEM coupling simulation.

2. The wear prediction method for spatial combined elbow according to claim 1 is characterized in that: Before inputting the operation parameter information of the elbow into the trained GPR model, the method further includes: Establish a three-dimensional model of a spatial combined elbow; Perform simulation based on CFD-DEM coupling and construct a sample set based on the simulation results; A GPR model is constructed, and the GPR model is trained based on the sample set to obtain a trained GPR model.

3. The wear prediction method of the spatial combination elbow according to claim 1 is characterized in that: The operation parameters of the elbow may include, but are not limited to: surface flow rates of the gas phase and liquid phase in the elbow, particle size and liquid viscosity.

4. The wear prediction method for spatial combined elbow according to claim 1 is characterized in that: Based on the sample set, the GPR model is trained using the conjugate gradient method.

5. The wear prediction method of spatial combined elbow according to claim 1 is characterized in that: During the training of the GPR model based on the sample set, a five-fold cross validation method is used to evaluate the performance of the GPR model.

6. A spatial combination elbow wear prediction system, characterized in that: include: An acquisition module is used to obtain operating parameter information of the elbow; The prediction module is used to input the operating parameter information of the elbow into the trained GPR model to obtain the wear amount of the spatial combined elbow, wherein the GPR model is trained based on the data set obtained by CFD-DEM coupling simulation.

7. The spatial combination elbow wear prediction system according to claim 6, characterized in that: Also includes: Establishing a module for establishing a three-dimensional model of a spatial combined elbow; The simulation module is used to simulate based on CFD-DEM coupling and build a sample set according to the simulation results; The training module is used to construct a GPR model and train the GPR model based on a sample set to obtain a trained GPR model.

8. The spatial combination elbow wear prediction system according to claim 7, characterized in that: The operation parameters of the elbow may include, but are not limited to: surface flow rates of the gas phase and liquid phase in the elbow, particle size and liquid viscosity.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the spatial combined elbow wear prediction method as described in any one of claims 1-5 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the spatial combined elbow wear prediction method as described in any one of claims 1 to 5 are implemented.

Images