Anti-wear optimization design method for turbine runner blade

Abstract

The invention relates to an anti-wear optimization design method for a turbine runner blade. The method comprises the following steps: acquiring environment characteristic information when the runner blade of the water turbine runs, and carrying out preprocessing and characteristic extraction on the environment characteristic information to obtain blade surface stress and initial distribution characteristics; then, the positions and stress peak values of high-pressure areas of the front edge and the blade back are calculated according to the characteristics, and a final material combination is determined by comparing various anti-abrasion materials; inputting the final material combination data into a wear prediction model, and analyzing the surface stress distribution and the wear rate of the blade by utilizing numerical simulation, so as to obtain a wear distribution estimated value of the blade in long-term operation; and finally, adjusting strength and anti-wear balance parameters in the wear prediction model by using an iterative optimization algorithm according to the pre-estimated value to obtain a final anti-wear design scheme. The optimal balance of strength and wear resistance is realized, the service life of the turbine runner blade is prolonged, the equipment maintenance cost and replacement frequency are reduced, and the operation stability and power generation efficiency of a turbine are improved.

Description

Technical Field

[0001] The invention belongs to the field of water conservancy and hydropower engineering, and in particular relates to an anti-wear optimization design method for turbine runner blades. Background Art

[0002] With the development of water conservancy and hydropower engineering technology, a wear-resistant optimization design technology for turbine runner blades has emerged. In the field of hydropower generation, turbines are core energy conversion equipment, and their operating stability and efficiency are crucial to power generation benefits. Turbine runner blades have been in a complex and harsh working environment for a long time, suffering from the continuous impact of water flow, particle erosion, and complex pressure and velocity changes in the flow field. Especially in waters with high sand content, the friction and impact of sediment particles on the blade surface aggravate the wear process. In the prior art, although anti-wear materials and improved structural design are used to alleviate the wear problem, there are still many deficiencies in accurately grasping the force and initial distribution characteristics of the blade surface and comprehensively and efficiently improving the wear resistance of the blade. This not only causes the operating efficiency of the turbine to decrease over time, but frequent equipment maintenance and replacement also greatly increases the cost of power generation, limiting the sustainable and efficient development of the hydropower industry. Summary of the invention

[0003] Based on this, it is necessary to provide an anti-wear optimization design method for turbine runner blades that can achieve an optimal balance between strength and wear resistance, extend the service life of turbine runner blades, and improve the operating stability and power generation efficiency of the turbine in response to the above technical problems.

[0004] In a first aspect, the present application provides a method for optimizing the wear resistance of a turbine runner blade, comprising: Acquire environmental characteristic information when turbine runner blades are running; preprocess and extract characteristics of the environmental characteristic information to obtain blade surface force and initial distribution characteristics; the environmental characteristic information includes at least one of water flow impact data, particle erosion parameters, flow field velocity, pressure distribution and sand content information.

[0005] According to the force and initial distribution characteristics of the blade surface, the position and peak force of the leading edge and back high-pressure areas are calculated, and a variety of anti-wear materials are compared to obtain the final material combination.

[0006] The final material combination data is input into the wear prediction model and numerical simulation is used to analyze the stress distribution and wear rate on the blade surface to obtain an estimated wear distribution of the blade during long-term operation.

[0007] According to the estimated wear distribution, the iterative optimization algorithm is used to adjust the strength and wear resistance balance parameters in the wear prediction model to obtain the final wear resistance design scheme.

[0008] In one embodiment, the environmental feature information is preprocessed and feature extracted to obtain the blade surface force and initial distribution characteristics, including: Based on the environmental feature information, wavelet transform and sliding window filtering technology are used to filter noise and remove outliers to obtain a stable feature data set.

[0009] A convolutional neural network is used to perform feature mining on a feature data set to obtain a potential feature subset related to the blade; the feature subset includes at least one of a water flow impact frequency feature, a particle size distribution feature, and a pressure fluctuation feature.

[0010] The feature subsets were analyzed by combining Pearson correlation coefficient and grey correlation analysis to obtain the key features reflecting the surface forces and initial distribution of the blades.

[0011] Principal component analysis, dimensionality reduction and information fusion are performed on key features to obtain fused feature parameters.

[0012] Based on the fusion characteristic parameter input and the trained fluid-solid coupling analysis model, the force and initial distribution characteristics of the blade surface are obtained.

[0013] In one embodiment, the positions and peak force values ​​of the leading edge and back high pressure areas are calculated based on the force and initial distribution characteristics of the blade surface, and a variety of wear-resistant materials are compared to obtain a final material combination, including: Based on the force and initial distribution characteristic data of the blade surface, the stress distribution of the blade under different working conditions is calculated to obtain the dynamic load spectrum of the blade surface; the dynamic load spectrum includes the location, size, and erosion risk level of the stress concentration area and the peak force sequence.

[0014] The clustering algorithm is used to classify the high-pressure area positions in the dynamic load spectrum, and the stress gradients in different areas are calculated using formulas to obtain the stress gradient distribution map associated with the material wear characteristics.

[0015] The stress gradient distribution diagram is matched and screened based on the preset fatigue threshold parameters of different material types to obtain candidate material combinations; the candidate material combinations must simultaneously meet the interface bonding strength constraints and the overall toughness constraints.

[0016] Multi-axial stress simulation tests are performed on candidate material combinations to generate stress-strain response curves of the mechanical behaviors of each combination of materials under different stress states.

[0017] The peak pressure decay rate index data is extracted according to the stress-strain response curve to obtain the final material combination that meets the peak pressure decay rate index.

[0018] In one embodiment, a clustering algorithm is used to classify the high pressure area positions in the dynamic load spectrum and a formula is used to calculate the stress gradients in different areas to obtain a stress gradient distribution diagram associated with the wear characteristics of the material, including: The data of dynamic load spectrum are processed by clustering algorithm to obtain the location classification results of high-pressure area.

[0019] The high-pressure area features are extracted from the location classification results to obtain the area division boundaries.

[0020] Based on the regional division boundary and using the stress gradient formula, the stress gradient value of each region is calculated.

[0021] The stress gradient value is calculated using the following stress gradient formula: ; in, represents the stress gradient value, represents the spatial dimension, represents stress, and represents the spatial coordinate component, represents the first-order stress derivative, represents the second-order stress derivative, and Represents the change in the spatial coordinate components of the calculated stress changes in different directions.

[0022] The correlation strength with the wear characteristics is determined based on the stress gradient value to obtain characteristic correlation data.

[0023] Based on the characteristic correlation data, the finite element method combined with high-order interpolation functions is used to fit and interpolate the stress data of different regions to generate a stress gradient distribution map associated with the material wear characteristics.

[0024] In one embodiment, the final material combination data is input into the wear prediction model and the stress distribution and wear rate on the blade surface are analyzed by numerical simulation to obtain an estimated wear distribution of the blade in long-term operation, including: The final material combination data is input into the wear prediction model to obtain material property parameters; the material property parameters include material composition and mechanical property data.

[0025] The blade surface stress field data is obtained by finite element simulation according to the material property parameters.

[0026] The surface stress field data is coupled with the dynamic wear coefficient and the parameters of the wear prediction model are updated to obtain a wear spatial distribution map; the wear spatial distribution map includes predicted values ​​of wear depth in different areas.

[0027] The wear spatial distribution map is matched and verified with the real-time monitoring data, and iterative calculations are performed based on the verification results to obtain the optimized stress field distribution characteristics.

[0028] The wear accumulation curves of multiple time nodes are constructed based on the stress field distribution characteristics, and the estimated wear distribution of the blade in long-term operation is calculated.

[0029] In one embodiment, the strength and anti-wear balance parameters in the wear prediction model are adjusted using an iterative optimization algorithm according to the wear distribution estimate to obtain a final anti-wear design solution, including: Obtain material performance indicators for wear distribution estimation; material performance indicators include strength parameters and wear resistance parameters.

[0030] The wear prediction model is adjusted according to material performance indicators to obtain a parameter adjustment strategy; the parameter adjustment strategy is used to balance the weight ratio of strength and wear resistance.

[0031] The parameter adjustment strategy is processed using an iterative optimization algorithm to obtain a parameter update sequence with dynamic feedback.

[0032] The parameter update sequence is input into the wear prediction model, and multiple sets of anti-wear performance simulation results are output.

[0033] According to the anti-wear performance simulation results, the environmental factor variables are extracted to correct the update sequence and generate the optimized anti-wear parameter combination.

[0034] The anti-wear parameter combination is compared with the preset model convergence conditions. If the model convergence conditions are met, the final anti-wear design solution is output; the final anti-wear design solution includes the matching relationship between the strength threshold and the anti-wear coefficient.

[0035] In a second aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0036] In a third aspect, the present application further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented.

[0037] The above-mentioned wear-resistant optimization design method for turbine runner blades obtains environmental characteristic information including water flow impact data, particle erosion parameters, flow field velocity, pressure distribution and sand content information when the turbine runner blade is running, and pre-processes and extracts characteristics to obtain the force and initial distribution characteristics of the blade surface; then calculates the position and force peak of the leading edge and blade back high pressure area based on the characteristics, and compares a variety of wear-resistant materials to determine the final material combination; then inputs the final material combination data into the wear prediction model, uses numerical simulation to analyze the stress distribution and wear rate of the blade surface, and then obtains the wear distribution estimate of the blade in long-term operation; finally, according to the estimate, the iterative optimization algorithm is used to adjust the strength and wear-resistant balance parameters in the wear prediction model to obtain the final wear-resistant design scheme. Achieve the best balance between strength and wear resistance, extend the service life of turbine runner blades, reduce equipment maintenance costs and replacement frequency, improve the operating stability and power generation efficiency of turbines, and effectively promote the sustainable development of the hydropower industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flow chart of a method for optimizing the wear resistance of a turbine runner blade provided by an embodiment of the present invention; Figure 2 A flowchart of preprocessing and feature extraction of environmental feature information to obtain blade surface force and initial distribution characteristics provided by an embodiment of the present invention; Figure 3 A flowchart of inputting the final material combination data into the wear prediction model and analyzing the blade surface stress distribution and wear rate by numerical simulation to obtain the wear distribution estimate of the blade in long-term operation provided by the embodiment of the present invention; Figure 4 The embodiment of the present invention provides a flowchart for adjusting the strength and wear resistance balance parameters in the wear prediction model using an iterative optimization algorithm according to the wear distribution estimate to obtain a final wear resistance design solution. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0041] First, as Figure 1 As shown, the present application provides a wear-resistant optimization design method for turbine runner blades, which may include: Step S101, obtaining environmental characteristic information of turbine runner blades during operation; preprocessing the environmental characteristic information and extracting characteristics to obtain blade surface force and initial distribution characteristics; the environmental characteristic information includes at least one of water flow impact data, particle erosion parameters, flow field velocity, pressure distribution and sand content information.

[0042] First, obtain the environmental characteristic information of the turbine runner blades during operation. The operating environment of the turbine is complex and changeable. The water flow impact data can intuitively reflect the key factors such as the force and frequency of the water flow impacting the blades, which directly affect the fatigue degree and wear condition of the blades; the particle erosion parameters involve the size, hardness, concentration and erosion angle of the particles carried in the water, which have a significant erosion effect on the blade surface; the flow field velocity and pressure distribution determine the flow pattern and pressure change of the water flow around the blades, which plays a key role in the force condition of the blades; the sand content information reflects the content of sediment in the water. As the main wear medium, the content of sediment is directly related to the rate of blade wear. After obtaining at least one of the above environmental characteristic information, it needs to be preprocessed to eliminate noise and outliers in the data and perform data standardization and other operations to improve data quality. Then, feature extraction is carried out. Through specific algorithms and technical means, key features are mined from the preprocessed data, and relevant information that can accurately describe the force condition and initial distribution characteristics of the blade surface is obtained.

[0043] Step S102, calculating the positions and peak force values ​​of the leading edge and back high pressure areas according to the blade surface force and initial distribution characteristics, and comparing various anti-wear materials to obtain a final material combination.

[0044] Based on the acquired and processed data of blade surface forces and initial distribution characteristics, the specific positions of the leading edge of the runner blade and the high-pressure area on the back of the blade can be accurately calculated by using a special mechanical model and calculation method. These two areas are prone to wear due to the large impact force and pressure of the water flow during the operation of the turbine. At the same time, the peak force values ​​borne by these areas are accurately calculated. The magnitude of the peak force directly reflects the extreme mechanical conditions faced by the blade at this position, which is extremely critical for assessing the wear risk of the blade. On this basis, in order to effectively improve the wear resistance of the blade, a comprehensive and detailed comparison of a variety of wear-resistant materials is required. Considering multi-dimensional indicators such as the material's hardness, toughness, corrosion resistance, fatigue resistance, and compatibility with the turbine operating environment, the advantages and disadvantages of different materials in dealing with the wear problems of different parts of the blade are comprehensively weighed. After rigorous comparative analysis, the optimal material combination for easily worn areas such as the leading edge of the blade and the high-pressure area on the back of the blade was finally determined to ensure that the turbine runner blades can have excellent wear resistance in complex and harsh operating environments, extend their service life, and improve the overall operating efficiency and reliability of the turbine.

[0045] Step S103, inputting the final material combination data into the wear prediction model and using numerical simulation to analyze the stress distribution and wear rate on the blade surface to obtain an estimated wear distribution value of the blade during long-term operation.

[0046] The final material combination data is fully input into the pre-built wear prediction model. The wear prediction model is built based on a large amount of experimental data, theoretical analysis and complex mathematical algorithms, and it can accurately simulate the wear process of turbine runner blades under actual operating conditions. With the help of advanced numerical simulation technology, the model is used to conduct in-depth analysis of the surface stress distribution of blades under different operating conditions. Numerical simulation can carefully simulate the dynamic changes of blade surface stress under complex effects such as water flow impact and particle erosion by fully incorporating factors such as blade geometry, material properties, and operating environment parameters into the calculation scope. At the same time, the model can also accurately calculate the wear rate of the blade, taking into account factors such as the wear characteristics of the material, the stress conditions, and the mechanism of action of water flow and particles. By continuously simulating the different stages of the blade during long-term operation, the wear distribution estimate of the blade during the entire operating cycle is obtained. This estimate is presented in the form of intuitive and detailed data and charts.

[0047] Step S104, adjusting the strength and anti-wear balance parameters in the wear prediction model using an iterative optimization algorithm according to the wear distribution estimate, to obtain a final anti-wear design solution.

[0048] The wear distribution estimate intuitively and in detail presents the expected wear conditions of various parts of the blade during long-term operation. The iterative optimization algorithm fine-tunes the strength and wear resistance balance parameters in the wear prediction model by continuously looping calculations. The strength parameter is related to the ability of the blade material to resist external force damage, and the wear resistance parameter determines the performance of the material to resist wear effects such as water flow erosion and particle erosion. During the adjustment process, the algorithm intelligently judges the deviation between the current parameter settings in the model and the actual expected wear resistance performance based on the wear distribution estimate. If it is found that the wear estimate in some areas is too high, it means that the current balance between strength and wear resistance tends to be insufficient strength or poor wear resistance. The algorithm will increase the weight of the wear resistance parameter or fine-tune the strength parameter accordingly, and vice versa. After multiple rounds of iterative calculations, the model parameters gradually approach the optimal solution, and finally obtain the final wear resistance design scheme that can accurately reflect the actual wear resistance requirements of the blade. This scheme comprehensively considers the wear characteristics of various parts of the blade under complex operating environments, ensuring that while meeting the strength requirements, the overall wear resistance of the blade is maximized, providing a solid guarantee for the efficient, stable and long-term operation of the turbine runner blades.

[0049] The above-mentioned wear-resistant optimization design method for turbine runner blades obtains environmental characteristic information including water flow impact data, particle erosion parameters, flow field velocity, pressure distribution and sand content information when the turbine runner blade is running, and pre-processes and extracts characteristics to obtain the force and initial distribution characteristics of the blade surface; then calculates the position and force peak of the leading edge and blade back high pressure area based on the characteristics, and compares a variety of wear-resistant materials to determine the final material combination; then inputs the final material combination data into the wear prediction model, uses numerical simulation to analyze the stress distribution and wear rate of the blade surface, and then obtains the wear distribution estimate of the blade in long-term operation; finally, according to the estimate, the iterative optimization algorithm is used to adjust the strength and wear-resistant balance parameters in the wear prediction model to obtain the final wear-resistant design scheme. Achieve the best balance between strength and wear resistance, extend the service life of turbine runner blades, reduce equipment maintenance costs and replacement frequency, improve the operating stability and power generation efficiency of turbines, and effectively promote the sustainable development of the hydropower industry.

[0050] In one embodiment, if Figure 2 As shown, preprocessing and feature extraction of environmental feature information to obtain blade surface force and initial distribution characteristics may include the following steps: Step S201, based on the environmental feature information, noise is filtered using wavelet transform and sliding window filtering technology, and outliers are removed to obtain a stable feature data set.

[0051] Step S202, using a convolutional neural network to perform feature mining on the feature data set to obtain a potential feature subset related to the blade; the feature subset includes at least one of a water flow impact frequency feature, a particle size distribution feature, and a pressure fluctuation feature.

[0052] Step S203, performing correlation analysis on the feature subset using a method combining the Pearson correlation coefficient and the grey correlation analysis to obtain key features reflecting the blade surface force and initial distribution.

[0053] Step S204, performing principal component analysis, dimensionality reduction and information fusion on the key features to obtain fused feature parameters.

[0054] Step S205, inputting the trained fluid-solid coupling analysis model based on the fusion characteristic parameters, and obtaining the blade surface force and initial distribution characteristics.

[0055] First, based on the acquired environmental feature information, the wavelet transform and sliding window filtering technology are used to filter the noise and remove outliers to obtain a stable feature data set. Next, the convolutional neural network is used to carry out feature mining on the data set to obtain at least one potential blade-related feature subset containing water flow impact frequency characteristics, particle size distribution characteristics, and pressure fluctuation characteristics. Subsequently, the Pearson correlation coefficient and grey correlation analysis are combined to perform correlation analysis on the feature subsets, and the key features reflecting the force and initial distribution of the blade surface are accurately extracted. After that, the key features are reduced in dimension and information fused through principal component analysis to generate fused feature parameters. Finally, the fused feature parameters are input into the trained fluid-solid coupling analysis model to successfully obtain the force and initial distribution characteristics of the blade surface.

[0056] Noise filtering and outlier removal are performed through wavelet transform and sliding window filtering technologies to ensure the reliability and stability of the data, laying a solid foundation for subsequent analysis. The use of convolutional neural networks efficiently mines potential features and broadens the cognitive dimension of blade-related characteristics. The combination of Pearson correlation coefficient and grey correlation analysis accurately locates key features and improves the accuracy of analysis. Principal component analysis dimensionality reduction and information fusion reduce data redundancy and improve computing efficiency, and the fused feature parameters can more comprehensively reflect the blade characteristics. Finally, the fluid-solid coupling analysis model provides a critical and accurate basis for the wear-resistant optimization design based on the blade surface force and initial distribution characteristics output by the fused feature parameters, which effectively guarantees the stable operation and wear-resistant performance improvement of turbine runner blades in complex environments.

[0057] In one embodiment, the positions and peak force values ​​of the leading edge and back high pressure areas are calculated based on the force and initial distribution characteristics of the blade surface, and a plurality of wear-resistant materials are compared to obtain a final material combination, which may include the following steps: Step S301, based on the blade surface force and initial distribution characteristic data, calculate the stress distribution of the blade under different working conditions to obtain a dynamic load spectrum of the blade surface; the dynamic load spectrum includes the location, size, and erosion risk level of the stress concentration area.

[0058] Step S302, using a clustering algorithm to classify the high pressure area positions in the dynamic load spectrum and using a formula to calculate the stress gradients of different areas to obtain a stress gradient distribution diagram associated with the material wear characteristics.

[0059] Step S303, matching and screening the stress gradient distribution map based on the fatigue threshold parameters preset for different material types to obtain a candidate material combination; the candidate material combination must satisfy both the interface bonding strength constraint and the overall toughness constraint.

[0060] Step S304: Perform a multi-axial stress simulation test on the candidate material combinations to generate stress-strain response curves of the mechanical behaviors of each combination of materials under different stress states.

[0061] Step S305, extracting peak pressure decay rate index data according to the stress-strain response curve to obtain a final material combination that meets the peak pressure decay rate index.

[0062] Specifically, relying on the force and initial distribution characteristics of the blade surface, in-depth calculations are carried out on the stress distribution of the blade under various working conditions, and then a detailed dynamic load map of the blade surface is generated. The map clearly presents key information such as the location, range, peak sequence of stress concentration areas, and erosion risk level. Subsequently, a clustering algorithm is used to classify the location of the high-pressure area in the dynamic load spectrum, and the stress gradients in different areas are calculated with the help of a specific formula, so as to draw a stress gradient distribution map closely related to the wear characteristics of the material. According to the fatigue threshold parameters pre-set for different material types, the stress gradient distribution map is accurately matched and screened, and the interface bonding strength constraints and overall toughness constraints are strictly followed to obtain candidate material combinations. Then, multi-axial stress simulation tests are carried out on the candidate material combinations to generate stress-strain response curves corresponding to the mechanical behavior of each combination material under different stress states. Finally, the peak pressure decay rate index data is extracted from the stress-strain response curve, and after careful comparison and analysis, the final material combination that meets the peak pressure decay rate index requirements is determined.

[0063] This embodiment can comprehensively and intuitively understand the stress condition of the blade in a complex operating environment by calculating the stress distribution under different working conditions and generating a dynamic load spectrum, providing basic data support for subsequent analysis. Clustering algorithms and stress gradient calculations help to clarify the relationship between stress characteristics and material wear in different regions, and point out the direction for material screening. Matching screening based on fatigue threshold parameters, combined with dual constraints, greatly narrows the range of candidate materials and improves screening efficiency and accuracy. The stress-strain response curve generated by the multi-axis stress simulation test truly shows the mechanical behavior of the material under different stresses, providing an intuitive basis for material performance evaluation. The final material combination is determined based on the peak pressure decay rate index to ensure that the selected material can effectively respond to pressure changes in actual operation, significantly improve the wear resistance and service life of the turbine runner blades, and ensure the long-term stable and efficient operation of the turbine.

[0064] In one embodiment, using a clustering algorithm to classify the high pressure area positions in the dynamic load spectrum and using a formula to calculate the stress gradients of different areas to obtain a stress gradient distribution map associated with the wear characteristics of the material may include the following steps: Step S401, using a clustering algorithm to process the data of the dynamic load spectrum to obtain a position classification result of the high-pressure area.

[0065] Step S402: extract high-voltage area features from the position classification results to obtain area division boundaries.

[0066] Step S403, using the stress gradient formula based on the region division boundary, calculate the stress gradient value of each region.

[0067] The stress gradient value is calculated using the following stress gradient formula: ; in, represents the stress gradient value, represents the spatial dimension, represents stress, and represents the spatial coordinate component, represents the first-order stress derivative, represents the second-order stress derivative, and Represents the change in the spatial coordinate components of the calculated stress changes in different directions.

[0068] Step S404: judging the correlation strength with the wear characteristic according to the stress gradient value, and obtaining characteristic correlation data.

[0069] Step S405, based on the characteristic correlation data, the stress data of different regions are fitted and interpolated using the finite element method combined with a high-order interpolation function to generate a stress gradient distribution diagram associated with the material wear characteristics.

[0070] First, a clustering algorithm is used to systematically process a large amount of complex data contained in the dynamic load spectrum, and data points with similar characteristics are clustered and summarized, so as to accurately obtain the position classification results of the high-pressure area. Subsequently, based on the position classification results, the high-pressure area feature extraction work is carried out in depth. Through specific algorithms and technical means, the regional division boundaries that can clearly define the scope of the high-pressure area are extracted from the complex data. Next, relying on the determined regional division boundaries, the established stress gradient formula is used for rigorous calculation. This formula comprehensively considers factors such as spatial dimensions, spatial coordinate components for calculating stress changes in different directions, and first-order and second-order stress derivatives, and obtains the stress gradient value corresponding to each high-pressure area through precise calculation. Based on the obtained stress gradient value, the correlation strength between it and the wear characteristics is deeply analyzed, and the possibility and degree of blade wear under different stress gradients are evaluated in detail, so as to obtain characteristic correlation data. Finally, based on the characteristic correlation data, with the help of the finite element method and combined with high-order interpolation functions, the stress data of different regions are finely fitted and interpolated, and the discrete stress data are converted into a continuous and intuitive graph form, and a stress gradient distribution map closely related to the material wear characteristics is successfully generated.

[0071] The clustering algorithm processes the dynamic load spectrum data, efficiently and accurately identifies the location of the high-pressure area, greatly improving the pertinence and efficiency of data processing. The feature extraction of the high-pressure area clearly divides the boundaries, providing an accurate range definition for subsequent stress calculations. The stress gradient calculation based on professional formulas comprehensively considers multiple factors to ensure the scientificity and reliability of the obtained stress gradient value. The characteristic correlation data obtained by judging the correlation strength between the stress gradient and the wear characteristics provides a key basis for in-depth understanding of the blade wear mechanism. The stress gradient distribution map generated by the finite element method combined with the high-order interpolation function presents the intrinsic connection between stress distribution and material wear characteristics in an intuitive and visual way, which provides an extremely important reference for the selection of wear-resistant materials, structural optimization design and operation and maintenance strategy formulation of turbine runner blades, effectively ensuring the stable operation of the turbine under complex working conditions and effectively improving the wear resistance and service life of the blades.

[0072] In one embodiment, if Figure 3 As shown, the final material combination data is input into the wear prediction model and the stress distribution and wear rate of the blade surface are analyzed by numerical simulation to obtain the estimated wear distribution of the blade in long-term operation, which may include the following steps: Step S501, inputting the final material combination data into the wear prediction model to obtain material attribute parameters; the material attribute parameters include material composition and mechanical property data.

[0073] Step S502, obtaining blade surface stress field data by finite element simulation calculation according to material property parameters.

[0074] Step S503, coupling the surface stress field data with the dynamic wear coefficient and updating the parameters of the wear prediction model to obtain a wear spatial distribution map; the wear spatial distribution map includes wear depth prediction values ​​of different regions.

[0075] Step S504, matching and verifying the wear spatial distribution map with the real-time monitoring data and performing iterative calculations according to the verification results to obtain optimized stress field distribution characteristics.

[0076] Step S505, constructing a wear accumulation curve of multiple time nodes based on the stress field distribution characteristics, and calculating an estimated wear distribution value of the blade in long-term operation.

[0077] The final material combination data determined after multiple rounds of screening is accurately input into the pre-built wear prediction model. The model outputs material property parameters including material composition and mechanical performance data based on material-related algorithms and data reserves. Next, using these material property parameters, the mechanical behavior of the blade under actual working conditions is simulated through finite element simulation technology, so as to calculate accurate blade surface stress field data. Subsequently, the surface stress field data is deeply coupled with the dynamic wear coefficient, and the material properties and actual wear influencing factors are comprehensively considered. The parameters of the wear prediction model are updated accordingly, and a detailed wear spatial distribution map is generated, in which the wear depth prediction values ​​of different regions are clearly presented. The wear spatial distribution map is rigorously matched and verified with the real-time monitoring data, and the differences between the two are analyzed. Iterative calculations are performed based on the verification results to continuously optimize the simulation accuracy of the model for the blade stress field, and the optimized stress field distribution characteristics are obtained. Finally, based on the optimized stress field distribution characteristics, the wear accumulation curve of multiple time nodes is constructed. By simulating and calculating the wear conditions at different time nodes, the wear distribution estimate of the blade in long-term operation is finally obtained.

[0078] By inputting the final material combination data into the wear prediction model to obtain the material property parameters, accurate basic material information is provided for subsequent analysis. Finite element simulation calculates the surface stress field data to truly simulate the actual force conditions of the blades. The surface stress field data is coupled with the dynamic wear coefficient to update the model parameters and generate a wear spatial distribution map, which comprehensively considers a variety of wear-related factors and improves the accuracy of wear prediction. Matching verification and iterative calculations with real-time monitoring data ensure that the model can continuously adapt to actual operation changes and optimize the stress field distribution characteristics. Constructing a multi-time node wear accumulation curve to obtain a long-term wear distribution estimate provides a key basis for preventive maintenance, replacement cycle determination, and wear-resistant design optimization of turbine runner blades, effectively ensuring the long-term stable operation of turbines, reducing maintenance costs, and extending the service life of blades.

[0079] In one embodiment, if Figure 4 As shown, according to the wear distribution estimation, the strength and wear resistance balance parameters in the wear prediction model are adjusted by using an iterative optimization algorithm to obtain the final wear resistance design scheme, which may include the following steps: Step S601, obtaining material performance indicators of wear distribution estimation values; the material performance indicators include strength parameters and wear resistance parameters.

[0080] Step S602, adjusting the wear prediction model according to the material performance index to obtain a parameter adjustment strategy; the parameter adjustment strategy is used to balance the weight ratio of strength and wear resistance.

[0081] Step S603: Process the parameter adjustment strategy using an iterative optimization algorithm to obtain a dynamic feedback parameter update sequence.

[0082] Step S604: input the parameter update sequence into the wear prediction model, and output multiple sets of anti-wear performance simulation results.

[0083] Step S605 , extracting the environmental factor variable correction update sequence according to the anti-wear performance simulation result, and generating an optimized anti-wear parameter combination.

[0084] Step S606, comparing the anti-wear parameter combination with the preset model convergence condition, and outputting the final anti-wear design solution if the model convergence condition is met; the final anti-wear design solution includes the matching relationship between the strength threshold and the anti-wear coefficient.

[0085] Specifically, the wear prediction model is adjusted according to the material performance indicators. By comprehensively considering the relationship between the multi-dimensional performance of the material, such as strength, toughness, hardness, and the wear resistance requirements of the blade, a parameter adjustment strategy is formulated. The core of this strategy is to reasonably balance the weight ratio of strength and wear resistance in the model to ensure that the model can more accurately reflect the actual operating conditions. Subsequently, an iterative optimization algorithm is introduced to deeply process the established parameter adjustment strategy. The iterative optimization algorithm dynamically adjusts the parameters according to the results of the previous round of calculations through continuous loop calculations, thereby obtaining a series of dynamic feedback parameter update sequences, which gradually approach the optimal solution as the number of iterations increases. The generated parameter update sequences are sequentially input into the wear prediction model. Based on these updated parameters, the model simulates the wear resistance of the blade under different parameter combinations and outputs multiple sets of wear resistance simulation results. From these simulation results, environmental factor variables, such as water flow velocity, sediment content, water temperature, etc., which have a significant impact on the wear resistance of the blade, are carefully extracted. Based on this, the update sequence is corrected to generate an optimized wear resistance parameter combination. Finally, the optimized anti-wear parameter combination is strictly compared with the pre-set model convergence conditions. If the convergence conditions are met, it means that the model has reached a stable and accurate state. At this time, the final anti-wear design scheme is output, which clearly shows the precise matching relationship between the strength threshold and the anti-wear coefficient.

[0086] This embodiment adjusts the wear prediction model according to the material performance indicators, so that the model is more suitable for the actual application scenario and the prediction accuracy is improved. The iterative optimization algorithm generates a parameter update sequence with dynamic feedback, effectively explores the parameter space, and improves the optimization efficiency. Multiple sets of anti-wear performance simulation results provide rich data for subsequent analysis, which is convenient for comprehensive evaluation of the effects of different parameter combinations. Based on the environmental factor variable correction update sequence, the actual operating environment The impact of the blade wear is fully considered to enhance the practicality of the solution. The anti-wear parameter combination is compared with the model convergence condition to ensure that the output of the final anti-wear design solution is stable and reliable. The matching relationship between the strength threshold and the anti-wear coefficient provides key guidance for the material selection, structural design, and operation and maintenance of the turbine runner blades, effectively ensuring that the blades have good wear resistance in complex environments, extending the service life, and improving the stability and economy of turbine operation.

[0087] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0088] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the aforementioned method for optimizing the wear resistance of a turbine runner blade are implemented.

[0089] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0090] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can refer to the partial description of the method embodiments. The device embodiments described above are only schematic, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0091] The above-mentioned embodiments only express several implementation methods of the embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the embodiments of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the embodiments of the present application, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. A wear-resistant optimization design method for turbine runner blades, characterized in that: The method comprises: Acquire environmental characteristic information when turbine runner blades are running; preprocess and feature extract the environmental characteristic information to obtain blade surface force and initial distribution characteristics; the environmental characteristic information includes at least one of water flow impact data, particle erosion parameters, flow field velocity, pressure distribution and sand content information; Calculating the positions and peak force values ​​of the leading edge and the back high pressure areas according to the blade surface force and initial distribution characteristics and comparing various anti-wear materials to obtain a final material combination; Inputting the final material combination data into a wear prediction model and using numerical simulation to analyze the stress distribution and wear rate on the blade surface to obtain an estimated wear distribution value of the blade during long-term operation; The strength and wear resistance balance parameters in the wear prediction model are adjusted using an iterative optimization algorithm according to the wear distribution estimate to obtain a final wear resistance design solution.

2. The method according to claim 1, characterized in that The preprocessing and feature extraction of the environmental feature information to obtain the blade surface force and initial distribution characteristics includes: Based on the environmental feature information, wavelet transform and sliding window filtering technology are used to filter noise and remove outliers to obtain a stable feature data set; Using a convolutional neural network to perform feature mining on the feature data set to obtain a potential feature subset related to the blade; the feature subset includes at least one of a water flow impact frequency feature, a particle size distribution feature, and a pressure fluctuation feature; The feature subsets are analyzed by combining the Pearson correlation coefficient and the grey correlation analysis method to obtain the key features reflecting the surface force and initial distribution of the blade. Performing principal component analysis, dimensionality reduction and information fusion on the key features to obtain fusion feature parameters; Based on the fusion characteristic parameters, the trained fluid-solid coupling analysis model is input to obtain the blade surface force and initial distribution characteristics.

3. The method according to claim 1, characterized in that: The position and peak force of the high-pressure area at the leading edge and the back of the blade are calculated according to the force and initial distribution characteristics of the blade surface, and a plurality of wear-resistant materials are compared to obtain a final material combination, including: Based on the blade surface force and initial distribution characteristic data, the stress distribution of the blade under different working conditions is calculated to obtain a dynamic load spectrum of the blade surface; the dynamic load spectrum includes the location, size, and erosion risk level of the stress peak sequence of the stress concentration area; Using a clustering algorithm to classify the high-pressure area positions in the dynamic load spectrum and using a formula to calculate the stress gradients of different areas, to obtain a stress gradient distribution map associated with the wear characteristics of the material; The stress gradient distribution diagram is matched and screened based on fatigue threshold parameters preset for different material types to obtain a candidate material combination; the candidate material combination must simultaneously satisfy an interface bonding strength constraint and an overall toughness constraint; Performing a multi-axial stress simulation test on the candidate material combination to generate a stress-strain response curve of the mechanical behavior of each combination material under different stress states; Peak pressure decay rate index data is extracted according to the stress-strain response curve to obtain a final material combination that meets the peak pressure decay rate index.

4. The method according to claim 3, characterized in that The clustering algorithm is used to classify the high pressure area positions in the dynamic load spectrum and the stress gradients of different areas are calculated using a formula to obtain a stress gradient distribution diagram associated with the material wear characteristics, including: Processing the data of the dynamic load spectrum by using a clustering algorithm to obtain a position classification result of the high-pressure area; Extracting high-voltage area features from the position classification results to obtain area division boundaries; Based on the regional division boundary, the stress gradient value of each region is calculated using the stress gradient formula; The stress gradient value is calculated using the following stress gradient formula: ； in, represents the stress gradient value, represents the spatial dimension, represents stress, and represents the spatial coordinate component, represents the first-order stress derivative, represents the second-order stress derivative, and Indicates the change in the spatial coordinate components of the calculated stress changes in different directions; Determining the correlation strength with the wear characteristic according to the stress gradient value to obtain characteristic correlation data; Based on the characteristic-related data, the stress data of different regions are fitted and interpolated using the finite element method combined with a high-order interpolation function to generate a stress gradient distribution diagram associated with the material wear characteristics.

5. The method according to claim 1, characterized in that The method of inputting the final material combination data into the wear prediction model and analyzing the blade surface stress distribution and wear rate by numerical simulation to obtain an estimated wear distribution value of the blade in long-term operation includes: Inputting the final material combination data into the wear prediction model to obtain material property parameters; the material property parameters include material composition and mechanical property data; According to the material property parameters, the blade surface stress field data is obtained by finite element simulation calculation; The surface stress field data is coupled with the dynamic wear coefficient and the parameters of the wear prediction model are updated to obtain a wear spatial distribution map; the wear spatial distribution map includes wear depth prediction values ​​of different regions; Matching and verifying the wear spatial distribution map with the real-time monitoring data and performing iterative calculations according to the verification results to obtain optimized stress field distribution characteristics; Based on the stress field distribution characteristics, the wear accumulation curves of multiple time nodes are constructed, and the wear distribution estimation of the blade in long-term operation is calculated.

6. The method according to claim 1, characterized in that The method of adjusting the strength and wear resistance balance parameters in the wear prediction model by using an iterative optimization algorithm according to the wear distribution estimation value to obtain a final wear resistance design scheme includes: Obtain material performance indicators of the wear distribution estimation value; the material performance indicators include strength parameters and wear resistance parameters; The wear prediction model is adjusted according to the material performance index to obtain a parameter adjustment strategy; the parameter adjustment strategy is used to balance the weight ratio of strength and wear resistance; Processing the parameter adjustment strategy using an iterative optimization algorithm to obtain a parameter update sequence of dynamic feedback; Inputting the parameter update sequence into the wear prediction model, and outputting multiple sets of anti-wear performance simulation results; Extracting the environmental factor variable correction update sequence according to the anti-wear performance simulation result to generate an optimized anti-wear parameter combination; The anti-wear parameter combination is compared with a preset model convergence condition, and if the model convergence condition is met, a final anti-wear design solution is output; the final anti-wear design solution includes a matching relationship between a strength threshold and an anti-wear coefficient.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images