Water turbine runner fatigue life prediction method based on multi-physics field coupling calculation

Abstract

The invention discloses a water turbine runner fatigue life prediction method based on multi-physics-field coupling calculation, and belongs to the technical field of water turbine runners, and the method comprises the steps: collecting the operation condition data of a water turbine set, carrying out the iterative solving of a residual vector based on a multi-physics-field coupling model, enabling the residual to approach zero, and outputting a time sequence result; extracting a vibration stress amplitude and a strain time history based on a time sequence result; accumulating the high-cycle fatigue damage and the low-cycle fatigue damage to obtain total accumulated damage of each runner unit, and performing weighted calculation by counting critical damage and operation duration to obtain the residual life of the hydraulic turbine set; and constructing a training data set containing historical working condition parameters, historical residual life and time sequence results of the hydraulic turbine set, training a Transform network, and generating a residual life prediction curve of the hydraulic turbine set under all working conditions. The method solves the problem that a traditional method is difficult to quantify coupling damage of the turbine runner temperature and the alternating load to predict the fatigue life of the turbine runner.

Description

Technical Field

[0001] The invention relates to a method for predicting fatigue life of a water turbine runner based on multi-physical field coupling calculation, and belongs to the technical field of water turbine runners. Background Art

[0002] Currently, fatigue life prediction for turbine runners primarily relies on single-field simulations or empirical formulas. Traditional methods typically decouple the temperature field from the mechanical load field, simplify calculations through static thermodynamic assumptions, or perform life assessments based solely on material fatigue properties at room temperature. For example, some studies use the finite element method combined with linear cumulative damage theory to predict fatigue damage, but these studies do not fully consider the multi-field coupling effects of fluid, structure, and temperature. Furthermore, damage calculations for high-cycle fatigue and low-cycle fatigue are often treated separately, lacking a unified quantitative model for combined loads.

[0003] However, traditional methods for predicting the fatigue life of turbine runners have many limitations: first, the coupling effect of temperature and mechanical load is seriously neglected, resulting in significant deviations between the fatigue parameters of materials and actual complex working conditions in traditional SN curves and Coffin-Manson equations due to the lack of temperature correction; second, existing single physical field simulation technologies cannot accurately simulate the transient coupling characteristics of fluid-structure-temperature fields, and the residual convergence accuracy is insufficient, which seriously affects the calculation authenticity of the time-series stress / strain field; finally, the composite damage accumulation method is too rough, using simple linear superposition without considering the nonlinear effect of temperature on the damage evolution rate, resulting in significant errors in the remaining life prediction. These limitations jointly restrict the accuracy and engineering applicability of turbine runner fatigue life prediction. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for predicting the fatigue life of a turbine runner based on multi-physics field coupling calculation. By iteratively solving multi-physics fields, temperature-corrected SN / Coffin-Manson equations, and integrating damage with the Miner criterion, and introducing Transformer to predict the full-operating life, the problem that traditional methods are difficult to quantify the coupled damage of turbine runner temperature and alternating load to predict the fatigue life of the turbine runner is solved.

[0005] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0006] The present invention provides a method for predicting the fatigue life of a turbine runner based on multi-physics field coupling calculation, comprising:

[0007] Collect operating data of the turbine unit, including speed, pressure, fluid domain temperature, structural domain temperature, and displacement;

[0008] According to the operating condition data, iteratively solve the residual vector based on the pre-built multi-physics field coupling model so that the residual approaches zero and outputs the time series result;

[0009] The time series results include temperature field, stress field, and strain field;

[0010] Finite element meshing is performed based on the geometric structure of the turbine unit to obtain discretized runner units, and the vibration stress amplitude and strain time history of each runner unit are extracted based on the time series results;

[0011] Calculate high cycle fatigue damage based on vibration stress amplitude and temperature-corrected SN curve;

[0012] The strain history is decomposed by the rain flow counting method, and the low cycle fatigue damage is calculated according to the temperature-corrected Coffin-Manson equation;

[0013] The Miner linear accumulation rule is used to accumulate high-cycle fatigue damage and low-cycle fatigue damage to obtain the total cumulative damage of each rotor unit.

[0014] The remaining life of the turbine unit is obtained by weighted calculation based on the total accumulated damage of each runner unit by statistically calculating the critical damage and the operating time;

[0015] A training dataset containing historical operating parameters, historical remaining life and time series results of the turbine unit is constructed, and the Transformer network is trained to generate the remaining life prediction curve of the unit under all operating conditions.

[0016] Furthermore, based on the interaction between flow field, temperature field and mechanical stress field, CFD model, FEA model and temperature field model are established respectively according to Navier-Stokes equations, nonlinear elastic dynamics equations, energy equations and heat transfer equations to construct a multi-physics field coupling model.

[0017] Furthermore, the multi-physics field coupling model is expressed as:

[0018] ;

[0019] Where, represents a multi-physics coupling model, Represents a vector of operating condition data of the turbine unit, where , Indicates speed, Indicates pressure, represents the fluid domain temperature, represents the fluid domain, represents the domain temperature, Represents the structural domain, represents displacement, represents the CFD model, represents the FEA model, represents the temperature field model, Indicates temperature.

[0020] Furthermore, according to the operating condition data, the residual vector is iteratively solved based on the pre-built multi-physics field coupling model so that the residual approaches zero and the time series result is output, including:

[0021] Initializing calculation parameters of the multi-physics field coupling model according to the operating condition data to obtain an initial calculation domain;

[0022] Flow iterative solution is performed based on the initial computational domain. Dynamic mesh technology is used to achieve bidirectional synchronization between flow field pressure load and structural deformation. A thermal-fluid coupling algorithm is used to calculate the heat flux density at the fluid-solid interface of the turbine unit, and intermediate coupled field results are obtained, including velocity field, pressure field, temperature field, and displacement field.

[0023] Calculate the residual vector based on the intermediate coupling field results and make a convergence judgment:

[0024] If the residual does not approach zero, the iteration is continued by adjusting the time step and relaxation factor until the convergence condition is met to obtain a converged time series result;

[0025] The spatiotemporal distribution data of temperature field, stress field and strain field are extracted according to the converged time series results, and the time series results are generated through the post-processing algorithm.

[0026] Furthermore, extracting the vibration stress amplitude and strain time history of each runner unit based on the time series results includes:

[0027] Perform dynamic analysis on each runner unit to obtain the vibration mode and natural frequency of each runner unit;

[0028] Perform harmonic response analysis based on the vibration mode and natural frequency of each runner unit to obtain the dynamic response of each runner unit under different working conditions;

[0029] Based on the time series results, the vibration stress amplitude and strain time history of each runner unit are extracted according to the dynamic response of each runner unit under different working conditions.

[0030] Furthermore, the temperature-corrected SN curve is expressed as:

[0031] ;

[0032] Where, Indicates the High cycle fatigue life of each runner unit, Indicates high cycle fatigue, Indicates the first The first The amplitude of the stress cycle, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength factors;

[0033] The high cycle fatigue damage is expressed as:

[0034] ;

[0035] Where, Indicates the High cycle fatigue damage of the runner unit, Indicates the number of high cycle fatigue cycles.

[0036] Furthermore, the temperature-corrected Coffin-Manson equation is expressed as:

[0037] ;

[0038] Where, Indicates the Low cycle fatigue life of each runner unit, Indicates solution, Indicates low cycle fatigue, Indicates the first The total strain amplitude of the wheel unit is Indicates low cycle fatigue The temperature of the rotor unit The relevant fatigue strength coefficient, Indicates low cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility coefficient, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility index, Indicates low cycle fatigue The temperature of the rotor unit The relevant elastic modulus, Indicates the first The number of failure cycles of each runner unit;

[0039] The low cycle fatigue damage is expressed as:

[0040] ;

[0041] Where, Indicates the Low cycle fatigue damage of the runner unit, Indicates the number of low-cycle fatigue cycles.

[0042] Furthermore, the total accumulated damage is expressed as:

[0043] ;

[0044] Where, Indicates the The total damage of the rotor unit.

[0045] Furthermore, the critical damage is expressed as:

[0046] ;

[0047] Where, Indicates critical damage, represents the maximum function, Represent the total damage of the first and second runner units respectively.

[0048] Furthermore, the remaining life of the hydro-turbine unit is expressed as:

[0049] ;

[0050] Where, Indicates the remaining life of the turbine unit, Indicates the turbine unit The critical damage of each working condition, The initial life of the turbine unit, ,in, represents the operating time of the turbine unit in the i-th working condition, and t represents the total operating time of the turbine unit.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. The present invention accurately captures the dynamic interaction of fluid, structure and temperature fields through a multi-physics field coupling model. Combined with the temperature-corrected SN curve and the Coffin-Manson equation, it quantifies the synergistic damage effect of temperature gradient and alternating load, overcoming the material parameter deviation problem caused by ignoring the temperature effect in traditional methods. At the same time, the life prediction model based on the Transformer network achieves dynamic and high-precision prediction of the remaining life under all operating conditions by learning the complex mapping relationship between historical operating conditions and time series results, providing a more reliable quantitative basis for maintenance decisions of turbine runners.

[0053] 2. The present invention establishes CFD models, FEA models and temperature field models respectively through the Navier-Stokes equations, nonlinear elastic dynamics equations and energy equations to construct a multi-physics field coupling model, realizing bidirectional synchronous calculation of flow field pressure loads and structural deformation, and ensuring the physical authenticity of temperature fields, stress fields and strain fields based on iterative residual convergence control, solving the problem that traditional single-field simulation cannot capture the transient interaction of fluid-structure-temperature, and improving the extraction accuracy of vibration stress amplitude and strain time history.

[0054] 3. The present invention innovatively embeds the temperature correction factor into the SN curve and the Coffin-Manson equation, dynamically characterizes the changes in material properties with temperature through the parameters of high-cycle fatigue and low-cycle fatigue, combines the rain flow counting method with the Miner criterion to achieve nonlinear accumulation of high-cycle and low-cycle damage, and calculates the remaining life through critical damage weighting, thereby reducing the life prediction error under high-temperature and high-load conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of a method for predicting fatigue life of a turbine runner based on multi-physics field coupling calculation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0057] Example 1

[0058] like Figure 1 As shown, this embodiment introduces a method for predicting the fatigue life of a turbine runner based on multi-physics field coupling calculation, including:

[0059] Step 1: Collect the operating data of the turbine unit, including speed, pressure, fluid domain temperature, structural domain temperature and displacement.

[0060] The present invention uses sensors to monitor the operating condition data of the turbine unit in real time, including speed, pressure, fluid domain temperature, structural domain temperature and displacement, providing initial boundary conditions for multi-physics field coupling calculations, ensuring that the simulation input data is consistent with the actual operating conditions, and solving the error problem caused by traditional methods relying on assumed boundary conditions.

[0061] Step 2: According to the operating condition data, based on the pre-built multi-physics field coupling model, iteratively solve the residual vector so that the residual approaches zero and outputs the time series results, which include temperature field, stress field, and strain field.

[0062] The present invention constructs a multi-physics field coupling model based on the Navier-Stokes equations, nonlinear elastic dynamics equations and energy equations, realizes bidirectional data transmission through dynamic mesh technology and thermal flow coupling algorithm, iteratively solves the residual vector until convergence, accurately captures the transient interaction effects of fluid-structure-temperature field, and generates high-fidelity time series field data, including temperature field, stress field and strain field.

[0063] Step 3: Perform finite element meshing according to the geometric structure of the turbine unit to obtain discretized runner units, and extract the vibration stress amplitude and strain time history of each runner unit based on the time series results.

[0064] The present invention uses the finite element method to discretize the wheel geometry into grids, combines time-series field data to perform harmonic response analysis, extracts the vibration stress amplitude and strain time history of each wheel unit, and provides a local refined load spectrum for fatigue damage calculation, overcoming the defect of traditional overall analysis method that ignores local stress concentration.

[0065] Step 4: Calculate high cycle fatigue damage based on the vibration stress amplitude and the temperature-corrected SN curve.

[0066] The present invention is based on a temperature-corrected SN curve, which correlates stress amplitude with temperature-related material parameters, quantifies high-cycle fatigue damage under alternating loads, and significantly improves the life prediction accuracy under high-temperature conditions.

[0067] Step 5: Decompose the strain history using the rain flow counting method and calculate the low-cycle fatigue damage based on the temperature-corrected Coffin-Manson equation.

[0068] The present invention uses the rain flow counting method to decompose the strain time history to obtain the strain amplitude spectrum, and combines the temperature-corrected Coffin-Manson equation to characterize the plastic strain damage through temperature-related parameters, thus overcoming the limitation of traditional methods that ignore the influence of temperature on material ductility.

[0069] Step 6: Use Miner's linear accumulation rule to accumulate high-cycle fatigue damage and low-cycle fatigue damage to obtain the total cumulative damage of each rotor unit.

[0070] Step 7: Based on the total cumulative damage of each runner unit, a weighted calculation is performed by statistically analyzing critical damage and operating time to obtain the remaining life of the turbine unit. This method uses the Miner criterion to linearly superimpose high-cycle fatigue damage and low-cycle fatigue damage, and calculates the remaining life by weighting critical damage with operating time, achieving quantitative assessment of composite damage and dynamic life prediction.

[0071] Step 8: Construct a training dataset containing the historical operating parameters, historical remaining life, and time series results of the turbine unit, and train the Transformer network to generate the remaining life prediction curve of the unit under all operating conditions.

[0072] The present invention is based on the Transformer network training historical working condition-life mapping relationship, and uses its self-attention mechanism to capture the nonlinear correlation between time series field data such as temperature field mutations and stress cycle characteristics and remaining life, to generate a full working condition adaptive prediction curve, breaking through the working condition limitations of traditional empirical formulas.

[0073] Example 2

[0074] Based on the same inventive concept as Example 1, this example introduces a method for predicting the fatigue life of a turbine runner based on multi-physics field coupling calculation, including the following steps:

[0075] Step 1: Collect the operating data of the turbine unit, including speed, pressure, fluid domain temperature, structural domain temperature and displacement.

[0076] Step 2: According to the operating condition data, based on the pre-built multi-physics field coupling model, iteratively solve the residual vector so that the residual approaches zero and outputs the time series results, which include temperature field, stress field, and strain field.

[0077] In this embodiment, according to the operating condition data, an iterative solution is performed based on a pre-built multi-physics field coupling model to make the residual vector approach zero and output a time series result, including:

[0078] Step 2.1: Initialize the calculation parameters of the multi-physics field coupling model according to the operating condition data to obtain an initial calculation domain.

[0079] Step 2.2: Perform flow iterative solution based on the initial computational domain. Use dynamic mesh technology to achieve bidirectional synchronization between flow field pressure load and structural deformation. Use thermal-fluid coupling algorithm to calculate the heat flux density at the fluid-solid interface of the turbine unit to obtain the intermediate coupled field results, including velocity field, pressure field, temperature field, and displacement field.

[0080] Step 2.3: Calculate the residual vector based on the intermediate coupling field results and make a convergence judgment:

[0081] If the residual does not approach zero, the iteration is continued by adjusting the time step and relaxation factor until the convergence condition is met to obtain a converged time series result.

[0082] Step 2.4: Extract the spatiotemporal distribution data of the temperature field, stress field, and strain field based on the converged time series results, and generate time series results through post-processing algorithms.

[0083] This embodiment is based on the interaction between flow field, temperature field, and mechanical stress field. According to the Navier-Stokes equations, nonlinear elastic dynamics equations, energy equations, and heat transfer equations, a CFD model, an FEA model, and a temperature field model are established respectively to construct a multi-physics field coupling model.

[0084] In this embodiment, the multi-physics field coupling model is expressed as:

[0085] ;

[0086] Where, represents a multi-physics coupling model, Represents a vector of operating condition data of the turbine unit, where , Indicates speed, Indicates pressure, represents the fluid domain temperature, represents the fluid domain, represents the domain temperature, Represents the structural domain, represents displacement, represents the CFD model, represents the FEA model, represents the temperature field model, Indicates temperature.

[0087] Step 3: Perform finite element meshing according to the geometric structure of the turbine unit to obtain discretized runner units, and extract the vibration stress amplitude and strain time history of each runner unit based on the time series results.

[0088] In this embodiment, extracting the vibration stress amplitude and strain time history of each runner unit based on the time series results includes:

[0089] Step 3.1: Perform dynamic analysis on each rotor unit to obtain the vibration mode and natural frequency of each rotor unit.

[0090] Step 3.2: Perform harmonic response analysis based on the vibration mode and natural frequency of each runner unit to obtain the dynamic response of each runner unit under different working conditions.

[0091] Step 3.3: Based on the timing results, extract the vibration stress amplitude and strain time history of each runner unit according to the dynamic response of each runner unit under different working conditions.

[0092] Step 4: Calculate the high cycle fatigue damage based on the vibration stress amplitude and the temperature-corrected SN curve.

[0093] In this embodiment, the temperature-corrected SN curve is expressed as:

[0094] ;

[0095] Where, Indicates the High cycle fatigue life of each runner unit, Indicates high cycle fatigue, Indicates the first The first The amplitude of the stress cycle, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength factors;

[0096] In this embodiment, the high cycle fatigue damage is expressed as:

[0097] ;

[0098] Where, Indicates the High cycle fatigue damage of the runner unit, Indicates the number of high cycle fatigue cycles.

[0099] Step 5: Decompose the strain history by the rain flow counting method and calculate the low-cycle fatigue damage according to the temperature-corrected Coffin-Manson equation.

[0100] In this embodiment, the temperature-corrected Coffin-Manson equation is expressed as:

[0101] ;

[0102] Where, Indicates the Low cycle fatigue life of each runner unit, Indicates solution, Indicates low cycle fatigue, Indicates the first The total strain amplitude of the wheel unit is Indicates low cycle fatigue The temperature of the rotor unit The relevant fatigue strength coefficient, Indicates low cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility coefficient, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility index, Indicates low cycle fatigue The temperature of the rotor unit The relevant elastic modulus, Indicates the first The number of failure cycles of each runner unit;

[0103] In this embodiment, the low cycle fatigue damage is expressed as:

[0104] ;

[0105] Where, Indicates the Low cycle fatigue damage of the runner unit, Indicates the number of low-cycle fatigue cycles.

[0106] Step 6: Use Miner's linear accumulation rule to accumulate high-cycle fatigue damage and low-cycle fatigue damage to obtain the total cumulative damage of each rotor unit.

[0107] In this embodiment, the total accumulated damage is expressed as:

[0108] ;

[0109] Where, Indicates the The total damage of the rotor unit.

[0110] Step 7: The total accumulated damage of each runner unit is weighted by statistical critical damage and operating time to obtain the remaining life of the turbine unit.

[0111] In this embodiment, the critical damage is expressed as:

[0112] ;

[0113] Where, Indicates critical damage, represents the maximum function, Represent the total damage of the first and second runner units respectively.

[0114] In this embodiment, the remaining life of the hydro-turbine unit is expressed as:

[0115] ;

[0116] Where, Indicates the remaining life of the turbine unit, Indicates the turbine unit The critical damage of each working condition, The initial life of the turbine unit, ,in, represents the operating time of the turbine unit in the i-th working condition, and t represents the total operating time of the turbine unit.

[0117] Step 8: Construct a training dataset containing the historical operating parameters, historical remaining life, and time series results of the turbine unit, and train the Transformer network to generate the remaining life prediction curve of the unit under all operating conditions.

[0118] Example 3

[0119] Based on the same inventive concept as other embodiments, this embodiment introduces a computer-readable storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of the method of the above-mentioned embodiment 1 or 2 are implemented.

[0120] Example 4

[0121] Based on the same inventive concept as other embodiments, this embodiment introduces a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the method in the above-mentioned embodiment 1 or 2 are implemented.

[0122] In summary, the present invention accurately captures the dynamic interaction of fluid, structure and temperature field through a multi-physics field coupling model, and combines the temperature-corrected SN curve and the Coffin-Manson equation to quantify the synergistic damage effect of temperature gradient and alternating load, overcoming the problem of material parameter deviation caused by ignoring the influence of temperature in traditional methods; at the same time, the life prediction model based on the Transformer network achieves dynamic and high-precision prediction of the remaining life under all working conditions by learning the complex mapping relationship between historical working conditions and time series results, providing a more reliable quantitative basis for maintenance decisions of turbine runners.

[0123] The present invention establishes CFD models, FEA models and temperature field models respectively through the Navier-Stokes equations, nonlinear elastic dynamics equations and energy equations to construct a multi-physics field coupling model, realizing bidirectional synchronous calculation of flow field pressure loads and structural deformation, and ensuring the physical authenticity of temperature fields, stress fields and strain fields based on iterative residual convergence control, solving the problem that traditional single-field simulation cannot capture the transient interaction of fluid-structure-temperature, and improving the extraction accuracy of vibration stress amplitude and strain time history.

[0124] The present invention innovatively embeds the temperature correction factor into the SN curve and the Coffin-Manson equation, dynamically characterizes the changes in material properties with temperature through the parameters of high-cycle fatigue and low-cycle fatigue, combines the rain flow counting method with the Miner criterion to realize the nonlinear accumulation of high-cycle and low-cycle damage, and calculates the remaining life through critical damage weighting, thereby reducing the life prediction error under high-temperature and high-load conditions.

[0125] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, 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 produce 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 flowcharts and / or block diagrams. 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.

[0127] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 The function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device 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.

[0129] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A method for predicting fatigue life of a turbine runner based on multi-physics field coupling calculation, characterized in that: include: Collect operating data of the turbine unit, including speed, pressure, fluid domain temperature, structural domain temperature, and displacement; According to the operating condition data, iteratively solve the residual vector based on the pre-built multi-physics field coupling model so that the residual approaches zero and outputs the time series result; The time series results include temperature field, stress field, and strain field; Finite element meshing is performed based on the geometric structure of the turbine unit to obtain discretized runner units, and the vibration stress amplitude and strain time history of each runner unit are extracted based on the time series results; Calculate high cycle fatigue damage based on vibration stress amplitude and temperature-corrected SN curve; The strain history is decomposed by the rain flow counting method, and the low cycle fatigue damage is calculated according to the temperature-corrected Coffin-Manson equation; The Miner linear accumulation rule is used to accumulate high-cycle fatigue damage and low-cycle fatigue damage to obtain the total cumulative damage of each rotor unit. The remaining life of the turbine unit is obtained by weighted calculation based on the total accumulated damage of each runner unit by statistically calculating the critical damage and the operating time; A training dataset containing historical operating parameters, historical remaining life and time series results of the turbine unit is constructed, and the Transformer network is trained to generate the remaining life prediction curve of the unit under all operating conditions.

2. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 1, characterized in that: It also includes establishing CFD models, FEA models and temperature field models based on the interaction between flow fields, temperature fields and mechanical stress fields, according to the Navier-Stokes equations, nonlinear elastic dynamics equations, energy equations and heat transfer equations to construct a multi-physics field coupling model.

3. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 2, characterized in that: The multi-physics coupling model is expressed as: ； Where, represents a multi-physics coupling model, Represents a vector of operating condition data of the turbine unit, where , Indicates speed, Indicates pressure, represents the fluid domain temperature, represents the fluid domain, represents the domain temperature, Represents the structural domain, represents displacement, represents the CFD model, represents the FEA model, represents the temperature field model, Indicates temperature.

4. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 1, characterized in that: According to the operating condition data, the residual vector is iteratively solved based on the pre-built multi-physics field coupling model so that the residual approaches zero and the timing results are output, including: Initializing calculation parameters of the multi-physics field coupling model according to the operating condition data to obtain an initial calculation domain; Flow iterative solution is performed based on the initial computational domain. Dynamic mesh technology is used to achieve bidirectional synchronization between flow field pressure load and structural deformation. A thermal-fluid coupling algorithm is used to calculate the heat flux density at the fluid-solid interface of the turbine unit, and intermediate coupled field results are obtained, including velocity field, pressure field, temperature field, and displacement field. Calculate the residual vector based on the intermediate coupling field results and make a convergence judgment: If the residual does not approach zero, the iteration is continued by adjusting the time step and relaxation factor until the convergence condition is met to obtain a converged time series result; The spatiotemporal distribution data of temperature field, stress field and strain field are extracted according to the converged time series results, and the time series results are generated through the post-processing algorithm.

5. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 1, characterized in that: The step of extracting the vibration stress amplitude and strain time history of each runner unit based on the time series results includes: Perform dynamic analysis on each runner unit to obtain the vibration mode and natural frequency of each runner unit; Perform harmonic response analysis based on the vibration mode and natural frequency of each runner unit to obtain the dynamic response of each runner unit under different working conditions; Based on the time series results, the vibration stress amplitude and strain time history of each runner unit are extracted according to the dynamic response of each runner unit under different working conditions.

6. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 1, characterized in that: The temperature-corrected SN curve is expressed as: ； Where, Indicates the High cycle fatigue life of each runner unit, Indicates high cycle fatigue, Indicates the first The first The amplitude of the stress cycle, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates high cycle fatigue The temperature of the rotor unit Related fatigue strength factors; The high cycle fatigue damage is expressed as: ； Where, Indicates the High cycle fatigue damage of the runner unit, Indicates the number of high cycle fatigue cycles.

7. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 6, characterized in that: The temperature-corrected Coffin-Manson equation is expressed as: ； Where, Indicates the Low cycle fatigue life of each runner unit, Indicates solution, Indicates low cycle fatigue, Indicates the first The total strain amplitude of the wheel unit is Indicates low cycle fatigue The temperature of the rotor unit The relevant fatigue strength coefficient, Indicates low cycle fatigue The temperature of the rotor unit Related fatigue strength index, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility coefficient, Indicates low cycle fatigue The temperature of the rotor unit The related fatigue ductility index, Indicates low cycle fatigue The temperature of the rotor unit The relevant elastic modulus, Indicates the first The number of failure cycles of each runner unit; The low cycle fatigue damage is expressed as: ； Where, Indicates the Low cycle fatigue damage of the runner unit, Indicates the number of low-cycle fatigue cycles.

8. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 7, characterized in that: The total cumulative damage is expressed as: ； Where, Indicates the The total damage of the rotor unit.

9. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 8, characterized in that: The critical damage is expressed as: ； Where, Indicates critical damage, represents the maximum function, Represent the total damage of the first and second runner units respectively.

10. The method for predicting fatigue life of a hydraulic turbine runner based on multi-physics field coupling calculation according to claim 9, characterized in that: The remaining life of the turbine set is expressed as: ； Where, Indicates the remaining life of the turbine unit, Indicates the turbine unit The critical damage of each working condition, The initial life of the turbine unit, ,in, represents the operating time of the turbine unit in the i-th working condition, and t represents the total operating time of the turbine unit.

Images