A high-load and wide-operating-condition turbine blade shaping method

Abstract

The present invention relates to a high-load wide-operating-condition turbine blade shaping method. Taking a high-load low-pressure turbine blade profile as a benchmark, trigonometric functions are used to define a number of wave equations for controlling the position of blade profiles along the span direction. The blade profiles on each span-wise section are completely consistent. Spline interpolation is used to generate a benchmark new-configuration bionic blade. A global optimization algorithm based on machine learning is used to optimize the key parameters of the benchmark new-configuration bionic blade, the pitch P, the amplitude A, and the blade leading edge radius r. The multi-objective function is set as the turbine blade cascade outlet total pressure loss coefficient under different attack angles. Optimization is carried out within a wide range of attack angles, a wide range of Mach numbers, and a wide range of Reynolds numbers. Sample points are automatically generated, and finally the optimal wide-operating-condition new-configuration turbine blade is output. The wide-operating-condition new-configuration turbine design method of the present invention has the characteristics of wide applicable attack angles, a wide range of Reynolds numbers and Mach numbers, and has a good flow control effect over the entire wide operating range.

Description

Technical Field

[0001] The invention belongs to the technical field of aeroengine / gas turbine turbines and relates to a high-load and wide-operating-condition turbine blade shaping method. Background Art

[0002] In the field of modern aircraft engines, the design and optimization of turbine components are the key to achieving high-performance aircraft. As one of the core components of the new generation of full-speed, large-flow variable-cycle aircraft engines, the design and manufacturing level of turbine components directly affects the overall performance and efficiency of the engine. In the development process of turbine components, designers face a long-standing challenge: how to improve efficiency while reducing losses and ensure that the turbine operates at high flow rates under a wide range of operating conditions.

[0003] The full-speed large-flow variable cycle turbine engine has higher requirements for turbine components due to its unique structure and working environment. This engine needs to maintain high aerodynamic performance under different flight conditions, such as from low altitude and low speed to high altitude and high speed. This requires that the turbine components must not only have the smallest outer diameter to achieve high flow, but also maintain high efficiency within a wide operating range. Such requirements lead to increased turbine load, enhanced flow capacity, and a significant increase in the blade angle of attack range, making traditional turbine aerodynamic design methods no longer applicable.

[0004] In order to solve this problem, scholars and researchers at home and abroad have conducted a lot of bionic research, drawing on the morphological characteristics of birds and fish in nature to explore new turbine design methods. The discovery of wavy bionic blades provides a new flow control method for improving aerodynamic performance. This blade design is inspired by organisms in nature. By imitating their specific morphology and structure, it can produce special flow effects on turbine blades, thereby improving their aerodynamic performance.

[0005] NASA's research found that applying the design method of bionic airfoils to the design of axial flow turbine blades can effectively delay the flow separation on the suction surface of the blades and improve their ability to maintain work after stalling. This design can not only improve the performance of the turbine under specific working conditions, but also broaden the effective operating angle of attack and working Mach number range of the blades, allowing turbine components to maintain efficient operation in a wider range of working conditions.

[0006] However, the traditional bionic airfoil is mainly aimed at the control of external flow noise, and adopts the design method of leading / trailing edge wave airfoil or wave shape on suction surface. For example, a bionic elliptical wave leading edge blade disclosed in CN114139306A is formed by stacking blade profiles with different chord lengths along the span direction, and the blade leading edge structure formed is a convex elliptical wave or a concave elliptical wave, and the average leading edge line of the elliptical wave leading edge blade is consistent with the leading edge line of the reference blade. Another example is a blade design method with a wavy leading edge and a serrated trailing edge disclosed in CN110701100A, which mainly includes the following steps: S1, determine the reference line; S2, generate a wavy leading edge line; S3, blade shaping of the wavy leading edge; S4, generate a serrated trailing edge line; S5, blade generation; modify the leading and trailing edge structure of the basic blade so that the blade can achieve a noise reduction effect. However, the control effect of this design on the internal flow of the impeller machinery is relatively limited. And CN110145370A discloses a low-pressure turbine blade with a wavy suction surface. By adjusting the cross-sectional shape of the blade, the blade suction surface curve produces circumferential periodic fluctuations, thereby constructing a suction surface with wavy characteristics. Due to the change in the distribution of the blade load in the span direction, the distance from the wave crest to the adjacent blade pressure surface is reduced, resulting in a reduction in the blade load of this section; the distance from the wave trough to the adjacent blade pressure surface is increased, resulting in an increase in the blade load of this section; resulting in low-energy fluid being more likely to accumulate in the wave trough section.

[0007] These methods are mainly aimed at flow control under design conditions, and lack the ability to optimize the design of key parameters under a wide range of conditions, so their scope of application is narrow. It can be seen that the current design technology for high-efficiency, wide-operating-condition new-configuration turbines is insufficient, and there is a lack of new-configuration turbine loss prediction models and design methods, which makes it difficult to meet the design requirements of a new generation of high-efficiency, wide-operating-condition, high-load turbines. Summary of the invention

[0008] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a high-load wide-operating-condition turbine blade shaping method, which uses a seal-like wave-shaped concave-convex structure to generate counter-rotating vortices to suppress high-load turbine flow separation, thereby suppressing flow separation at a high-load turbine at a large angle of attack and widening the turbine's angle of attack range. The wide-operating-condition new turbine design method of the present invention has the characteristics of being applicable to a wide angle of attack, a wide range of Reynolds numbers and Mach numbers, and has a good flow control effect over the entire wide operating range.

[0009] The technical solution adopted by the present invention to solve its technical problems is:

[0010] A high-load wide-operating condition turbine blade shaping method comprises the following steps:

[0011] SS1, determine the reference blade profile: select the high-load low-pressure turbine blade profile as the reference blade profile, and use trigonometric functions to define several wave equations that control the position of the blade profile along the span direction;

[0012] SS2, interpolation to generate a benchmark new configuration bionic blade: perform spline interpolation on the blade profile on each spanwise section to generate a benchmark new configuration bionic blade;

[0013] SS3, design parameter determination: determine the key design parameters, including pitch P, amplitude A and blade leading edge radius r;

[0014] SS4, establish a multi-objective optimization model: minimize the total pressure loss coefficient at the turbine blade cascade outlet under different attack angles as the optimization goal, and conduct optimization in a wide range of attack angles, wide Mach numbers and wide Reynolds numbers;

[0015] SS5, machine learning global optimization: uses machine learning algorithms for global optimization, generates multiple new sample points in each iteration, and calculates the objective function value of each sample point;

[0016] SS6, sample point sorting and updating: sort the sample points according to the objective function value, select the best sample point as the starting point of the next iteration, and continuously update the model until the termination condition is met;

[0017] SS7, re-establishing the bionic blade: based on the optimization results, the optimization is fed back to the benchmark new configuration bionic blade, and finally the optimal wide-operating-condition new configuration turbine blade is output.

[0018] Among them, in step SS4, the blade profile loss of the blade is comprehensively calculated by weighted averaging. When weighted averaging, the blade profile loss of the design angle of attack is reduced in weight, and the comprehensive blade profile loss is The calculation formula is:

[0019]

[0020] Where Yp is the blade loss, and the calculation formula is:

[0021]

[0022] Where, P in,T , P in,s and They correspond to the total pressure, static pressure at the mid-section of the inlet and the average total pressure of a single periodic section at 1.4 times the axial chord length of the blade outlet.

[0023] Preferably, by changing the pitch P and amplitude A of the new turbine configuration, taking the blade comprehensive blade loss as the objective function, the key parameter selection criteria of the new turbine configuration are calculated. Specifically, the following specific steps are included:

[0024] SS51, initial setting: Set a reasonable range of variation for pitch P and amplitude A based on previous design experience, theoretical analysis or factors affecting turbine performance.

[0025] SS52, Create a parametric design model: Use computational fluid dynamics (CFD) software or other numerical simulation tools to create a parametric turbine blade model in which the blade geometry is automatically adjusted based on changes in pitch P and amplitude A.

[0026] SS53, Systematic variation of design parameters: The values ​​of pitch P and amplitude A are systematically varied according to a strategy predetermined by the optimization algorithm; a series of blade designs are generated.

[0027] SS54, Performance Evaluation and Loss Calculation: Numerical simulations are performed on each design to evaluate its aerodynamic performance. The purpose of the simulation is to calculate the flow characteristics and profile losses of the blade under different operating conditions. Based on the simulation results, the profile losses are weighted averaged at different angles of attack using formula (1) to obtain a comprehensive loss assessment.

[0028] SS55, Optimization process: Use an optimization algorithm to find the values ​​of pitch P and amplitude A that minimize the blade's comprehensive profile loss. The optimization algorithm iteratively updates the parameter values ​​based on the gradient of the loss function or through a search strategy, gradually approaching the optimal solution.

[0029] SS56, Determination of key parameter selection criteria: Analyze the data obtained during the optimization process to identify the values ​​of pitch P and amplitude A under which the blade comprehensive blade loss is minimized. Based on these data, establish key parameter selection criteria to quickly find the optimal blade design parameters.

[0030] SS57, Validation and Iteration: Select the optimal parameter combination for additional detailed simulation and / or experimental validation to ensure the effectiveness of these parameters in actual applications. Based on the validation results, iteratively adjust the parameter selection criteria to further improve the performance of the blade design.

[0031] Preferably, in the optimization process, the gradient descent method is used to find the values ​​of the pitch P and amplitude A that minimize the comprehensive blade profile loss. This is an iterative process that requires continuous simulation and calculation. Specifically, the following steps are included:

[0032] SS551, Calculate the objective function and its gradient: Under the current parameter point, use computational fluid dynamics (CFD) simulation or other numerical methods to calculate the comprehensive blade profile loss, that is, the value of the objective function. Calculate the gradient (partial derivative) of the objective function with respect to pitch P and amplitude A. This gradient points to the direction where the loss increases fastest, so we hope to update the parameters in the opposite direction of the gradient to reduce the loss.

[0033] SS552, Determine step size and update parameters: Choose an appropriate step size (learning rate), which determines the magnitude of the parameter update at each iteration. The choice of step size has a great impact on the convergence speed and stability of the algorithm. Update the current parameter value based on the calculated gradient. The new value of the parameter is obtained by subtracting the gradient from the current value multiplied by the step size.

[0034] SS553, check convergence: After each iteration, check the change of the old and new parameter values ​​and the change of the objective function value. If the change is less than the preset threshold, or reaches a certain number of iterations, the algorithm is considered to have converged, and the parameter value at this time is the optimal solution. If the algorithm has not converged, continue the iteration step until the convergence condition is met.

[0035] SS554, iterative process: repeat steps 1 to 3 until convergence. In each iteration, the parameter value is updated in the opposite direction of the gradient, gradually approaching the minimum value of the objective function.

[0036] SS555, Result Verification: Once the optimal parameter values ​​are found, additional simulations are performed using these parameters to ensure that the blade's comprehensive profile losses are indeed minimized under actual operating conditions.

[0037] Preferably, the trigonometric function used in step SS1 is a sine function or a cosine function, which is used to accurately control the position change of the blade profile along the span direction.

[0038] Preferably, in step SS2, the spline interpolation uses a cubic or higher-order spline function to generate a smooth and continuous reference new-configuration bionic blade profile.

[0039] Preferably, a special coating is applied to the surface of the turbine blades of the new configuration to reduce the friction coefficient of the blade surface and further improve the aerodynamic efficiency of the turbine.

[0040] Preferably, internal cooling passages are designed for the blades to improve performance and durability in high temperature environments.

[0041] The advantages of the present invention compared with the prior art are as follows:

[0042] (1) The newly configured turbine of the present invention can generate counter-rotating vortices on the suction surface. These vortices can effectively accelerate the momentum exchange between the boundary layer and the mainstream, thereby significantly suppressing the flow separation phenomenon under the operating environment of large angle of attack and low Reynolds number. Flow separation is one of the main reasons for the decline in turbine performance. By reducing separation, flow losses can be significantly reduced and turbine efficiency can be improved. The design of the present invention is particularly suitable for those situations where it works at a large angle of attack, such as take-off and climb phases, which place extremely high requirements on the performance of the turbine. At the same time, at high Mach numbers and designed angles of attack, the newly configured turbine can still maintain excellent performance, and will not cause additional losses due to the generation of counter-rotating vortices, thereby improving adaptability and stability under a wide range of working conditions.

[0043] (2) Compared with the method in the prior art that adjusts the cross-sectional shape of the blade to make the suction surface curve of the blade produce circumferential periodic fluctuations and then construct a suction surface with wave-like characteristics, the present invention does not change the blade profile section, but makes the blade profile section slide periodically in the axial direction and expand in the span direction to realize the construction of the wave characteristics of the blade surface. Wave characteristics exist on both the leading edge and the trailing edge of the blade, and there is a conversion position for the wave characteristics on the suction surface, that is, before and after the conversion position, the crest characteristics will be converted into trough characteristics, and the trough characteristics will be converted into crest characteristics. Under the induction of the leading edge wave characteristics, the position where the suction surface generates counter-rotating vortices is more forward, and counter-rotating vortices are also generated on the pressure surface, which can suppress the separation of the pressure surface under negative angles of attack and broaden the range of angles of attack that the blade can adapt to. In addition, the wave characteristics of the trailing edge have the effect of accelerating the crushing and mixing of the Karman vortex street generated by the trailing edge of the blade, avoiding the adverse effects of the large vortex structure generated by the trailing edge of the blade on the downstream.

[0044] (3) The design method of the present invention is based on a deep understanding and mastery of the flow mechanism of the new turbine configuration, and can more accurately predict and control the flow characteristics inside the turbine. Through the understanding of the flow mechanism, we can better understand the complex flow phenomena inside the turbine, such as boundary layer development, flow separation and reattachment, which are key factors affecting turbine performance. On this basis, more effective flow control strategies can be developed, such as optimizing flow characteristics by adjusting parameters such as blade profile, blade installation angle and number of blades, thereby improving the accuracy of the design.

[0045] (4) The traditional turbine design system often relies on empirical formulas and simplified flow models, which to a certain extent limits the flexibility and innovation of the design. The present invention broadens the traditional turbine design system by introducing new design concepts and technical means, such as bionic principles, computational fluid dynamics simulation, and machine learning optimization algorithms. This method not only improves the accuracy of the design, but also provides new ideas and methods for the future development of turbine component design. For example, the application of bionic principles creates more efficient and reliable turbine blades; computational fluid dynamics simulation can simulate complex flow conditions in computers and optimize design parameters; machine learning optimization algorithms can process large amounts of data and quickly find the optimal solution. The parts not elaborated in detail in the present invention are common knowledge in the field. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a high-load, wide-operating-condition turbine blade shaping method of the present invention.

[0047] Figure 2 It is a structural schematic diagram of a reference high-load low-pressure turbine blade profile in the present invention;

[0048] Figure 3 is a schematic diagram of a wave equation for controlling the position of a blade profile along the span direction in the present invention;

[0049] Figure 4 It is a schematic structural diagram of a high-load and wide-operating-condition new turbine configuration in the present invention;

[0050] Figure 5 This is a criterion diagram for selecting key parameters of the new turbine configuration in the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limitations on the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The structure and technical scheme of the present invention are further described in detail below in conjunction with the drawings, and an embodiment of the present invention is given.

[0052] like Figure 1-5 As shown, a high-load wide-operating condition turbine blade shaping method of the present invention comprises the following steps:

[0053] SS1, determine the reference blade profile: select the high-load low-pressure turbine blade profile as the reference blade profile, and use trigonometric functions to define several wave equations that control the position of the blade profile along the span direction;

[0054] SS2, interpolation to generate a benchmark new configuration bionic blade: perform spline interpolation on the blade profile on each spanwise section to generate a benchmark new configuration bionic blade;

[0055] SS3, design parameter determination: determine the key design parameters, including pitch P, amplitude A and blade leading edge radius r;

[0056] SS4, establish a multi-objective optimization model: minimize the total pressure loss coefficient at the turbine blade cascade outlet under different attack angles as the optimization goal, and conduct optimization in a wide range of attack angles, wide Mach numbers and wide Reynolds numbers;

[0057] SS5, machine learning global optimization: uses machine learning algorithms for global optimization, generates multiple new sample points in each iteration, and calculates the objective function value of each sample point;

[0058] SS6, sample point sorting and updating: sort the sample points according to the objective function value, select the best sample point as the starting point of the next iteration, and continuously update the model until the termination condition is met;

[0059] SS7, re-establishing the bionic blade: based on the optimization results, the optimization is fed back to the benchmark new configuration bionic blade, and finally the optimal wide-operating-condition new configuration turbine blade is output.

[0060] The trigonometric function used in step SS1 is a sine function or a cosine function, which is used to accurately control the position change of the blade profile along the span direction.

[0061] In step SS2, the spline interpolation uses a cubic or higher-order spline function to generate a smooth and continuous reference new-configuration bionic blade profile.

[0062] Specifically, a high-load wide-operating condition turbine blade shaping method of the present invention is based on a high-load low-pressure turbine blade profile (such as Figure 2 As shown in the figure), using trigonometric functions, several wave equations that control the position of the blade profile along the span direction are defined (such as Figure 3 The blade profiles on each spanwise section are completely consistent, and the reference new configuration bionic blade is generated by spline interpolation (as shown in Figure 4 ).

[0063] The key parameters of the benchmark new configuration bionic blade, such as pitch P, amplitude A and blade leading edge radius r, are optimized using a global optimization algorithm based on machine learning (e.g. Figure 1 and Figure 2), taking the comprehensive blade profile loss of the blade as the multi-objective function, the multi-objective function is set as the total pressure loss coefficient of the turbine blade cascade outlet under different attack angles, and conducting optimization in the wide attack angle (-25°~+15°), wide Mach number and wide Reynolds number range, automatically generating sample points, and finally outputting the optimal new configuration turbine blade for a wide range of operating conditions.

[0064] In order to comprehensively evaluate the effect of the amplitude and pitch of the new blade configuration on the blade loss under the condition of variable inlet angle of attack, the blade loss of the blade is comprehensively calculated by weighted average. Considering that it is difficult to ensure that the turbine blade operates at the design angle of attack in actual operation, the blade loss at the design angle of attack is reduced in weighted average. The calculation formula is:

[0065]

[0066] In turbine blade design, blade loss Yp is a key performance indicator, which reflects the performance of the blade under specific working conditions. Blade loss usually consists of two parts: one part is the loss caused by blade surface friction and shape, and the other part is the loss caused by flow separation near the leading and trailing edges of the blade. In practical applications, turbine blades need to work at different angles of attack, including positive angle of attack, negative angle of attack and 0 angle of attack (i.e., design angle of attack). Therefore, in order to comprehensively evaluate the performance of the blade, it is necessary to comprehensively calculate the blade loss at these different angles of attack. The positive angle of attack usually refers to the positive angle of the blade relative to the incoming flow, while the negative angle of attack refers to the negative angle of the blade relative to the incoming flow. In actual flight, the aircraft may need to fly at various angles of attack, including take-off, climb, cruise and descent. These stages have different performance requirements for the turbine, so it is necessary to ensure that the blade can maintain good performance at both positive and negative angles of attack. In the present invention, the calculation of the comprehensive blade loss adopts a weighted average method. This method takes into account the different effects of blade loss at different angles of attack and integrates these losses by assigning different weights.

[0067] Giving equal weight to positive and negative angles of attack, the performance of the blades in both cases is equally important. This is because in actual flight, the aircraft will frequently switch between positive and negative angles of attack, so the performance of the blades at both angles of attack needs to be fully considered and optimized. 0 angle of attack, also known as the design angle of attack, refers to the angle of the blade under design conditions. The performance of the blades at this angle of attack is crucial to the performance of the entire turbine because it is directly related to the efficiency and stability of the turbine under standard operating conditions. Because in actual flight, it is unlikely that the blades will remain at the full design angle of attack for a long time. Therefore, the performance at 0 angle of attack is important, but its impact is relatively small during the entire flight process compared to positive and negative angles of attack. Through this weighted average calculation method, it is ensured that the turbine blades can perform well at various angles of attack, while also taking into account the various operating conditions that may be encountered in actual flight. This method helps to improve the adaptability and reliability of the turbine and ensure that the turbine can still operate efficiently under variable flight conditions. In addition, this weight distribution takes into account the comprehensive consideration of blade performance and the pursuit of improving the overall performance of the turbine. By optimizing the blade design, reducing flow separation and reducing losses, the efficiency of the turbine can be improved.

[0068] Where Yp is the blade loss, and the calculation formula is:

[0069]

[0070] Where, P in,T , P in,s and They correspond to the total pressure and static pressure at the inlet midsection and the average total pressure of a single periodic section at 1.4 times the axial chord length at the cascade outlet. The blade profile loss Yp is the ratio of the difference between the average value of the inlet total pressure and the outlet total pressure to the difference between the inlet total pressure and the inlet static pressure. The smaller this ratio is, the lower the blade profile loss is and the better the performance of the turbine blade is. The total pressure averaging at 1.4 times the axial chord length at the cascade outlet is chosen to capture the average state of the flow near the cascade outlet, so as to more accurately evaluate the performance of the blade. This is because the position of 1.4 times the axial chord length is considered to be able to better represent the average flow state at the cascade outlet. At this position, the flow has basically developed and stabilized, and can reflect the overall effect of the blade on the airflow. By taking the average value on a single periodic section, the influence caused by local flow structures (such as vortices, flow separation, etc.) is reduced, thereby obtaining more stable and representative data. By selecting a representative position at the cascade outlet for total pressure measurement and averaging, the effect of blade design can be better evaluated, providing a basis for further optimization.

[0071] By changing the new turbine pitch P and amplitude A, taking the blade comprehensive blade loss as the objective function, we can calculate Figure 2The selection criteria of key parameters of the new turbine configuration include the following specific steps:

[0072] SS51, initial setting: Set a reasonable range of variation for pitch P and amplitude A based on previous design experience, theoretical analysis or factors affecting turbine performance.

[0073] SS52, Create a parametric design model: Use computational fluid dynamics (CFD) software or other numerical simulation tools to create a parametric turbine blade model in which the blade geometry is automatically adjusted based on changes in pitch P and amplitude A.

[0074] SS53, Systematic variation of design parameters: The values ​​of pitch P and amplitude A are systematically varied according to a strategy predetermined by the optimization algorithm; a series of blade designs are generated.

[0075] SS54, Performance Evaluation and Loss Calculation: Numerical simulations are performed on each design to evaluate its aerodynamic performance. The purpose of the simulation is to calculate the flow characteristics and profile losses of the blade under different operating conditions. Based on the simulation results, the profile losses are weighted averaged at different angles of attack using formula (1) to obtain a comprehensive loss assessment.

[0076] SS55, Optimization process: Use an optimization algorithm to find the values ​​of pitch P and amplitude A that minimize the blade's comprehensive profile loss. The optimization algorithm iteratively updates the parameter values ​​based on the gradient of the loss function or through a search strategy, gradually approaching the optimal solution.

[0077] SS56, Determination of key parameter selection criteria: Analyze the data obtained during the optimization process to identify the values ​​of pitch P and amplitude A under which the blade comprehensive blade loss is minimized. Based on these data, establish key parameter selection criteria to quickly find the optimal blade design parameters.

[0078] SS57, Validation and Iteration: Select the optimal parameter combination for additional detailed simulation and / or experimental validation to ensure the effectiveness of these parameters in actual applications. Based on the validation results, iteratively adjust the parameter selection criteria to further improve the performance of the blade design.

[0079] In the optimization process, the gradient descent method is used to find the values ​​of pitch P and amplitude A that minimize the comprehensive blade profile loss. This is an iterative process that requires continuous simulation and calculation. The specific steps include:

[0080] SS551, Calculate the objective function and its gradient: Under the current parameter point, use computational fluid dynamics (CFD) simulation or other numerical methods to calculate the comprehensive blade profile loss, that is, the value of the objective function. Calculate the gradient (partial derivative) of the objective function with respect to pitch P and amplitude A. This gradient points to the direction where the loss increases fastest, so we hope to update the parameters in the opposite direction of the gradient to reduce the loss.

[0081] SS552, Determine step size and update parameters: Choose an appropriate step size (learning rate), which determines the magnitude of the parameter update at each iteration. The choice of step size has a great impact on the convergence speed and stability of the algorithm. Update the current parameter value based on the calculated gradient. The new value of the parameter is obtained by subtracting the gradient from the current value multiplied by the step size.

[0082] SS553, check convergence: After each iteration, check the change of the old and new parameter values ​​and the change of the objective function value. If the change is less than the preset threshold, or reaches a certain number of iterations, the algorithm is considered to have converged, and the parameter value at this time is the optimal solution. If the algorithm has not converged, continue the iteration step until the convergence condition is met.

[0083] SS554, iterative process: repeat steps 1 to 3 until convergence. In each iteration, the parameter value is updated in the opposite direction of the gradient, gradually approaching the minimum value of the objective function.

[0084] SS555, Result Verification: Once the optimal parameter values ​​are found, additional simulations are performed using these parameters to ensure that the blade's comprehensive profile losses are indeed minimized under actual operating conditions.

[0085] like Figure 4 As shown in the figure, when the pitch size is a to b times the leading edge radius and c to d times the leading edge radius, the new configuration blade has a reducing effect on the comprehensive blade loss. When the pitch size is a times the leading edge radius and the amplitude size is below e times the leading edge radius, the new configuration blade has the best effect on reducing the comprehensive blade loss. The optimal design parameters of the new configuration blade are determined.

[0086] Preferably, a special coating is applied to the surface of the new configuration turbine blade to reduce the friction coefficient of the blade surface and further improve the aerodynamic efficiency of the turbine. The application of this special coating can not only significantly reduce the friction resistance between the blade and the airflow during high-speed rotation, but also improve the wear resistance and corrosion resistance of the blade, thereby extending the service life of the blade. The low friction characteristics of the coating help to reduce energy loss and improve the output power and overall efficiency of the turbine. In addition, the high wear resistance of the coating can reduce the damage to the blade surface caused by friction, maintain the geometric shape and aerodynamic performance of the blade, and ensure the reliability and stability of the turbine in long-term operation.

[0087] Preferably, internal cooling channels are designed for the blades to improve performance and durability in high temperature environments. The design of the internal cooling channels allows a cooling medium, such as air or coolant, to pass through the inside of the blades, effectively absorbing the heat generated by the blades under high temperature conditions. This design can significantly reduce the operating temperature of the blades, reduce thermal stress and thermal deformation, and improve the strength and toughness of the blade materials, thereby enhancing the performance of the blades under extreme temperature conditions. At the same time, the presence of cooling channels also helps prevent fatigue cracks and fractures in the blades caused by long-term high-temperature operation, further improving the durability and safety of the blades. By optimizing the layout and design of the cooling channels, more efficient heat transfer and more uniform temperature distribution can be achieved, thereby improving the cooling effect of the entire turbine blade.

[0088] The advantages of the present invention compared with the prior art are as follows:

[0089] (1) The newly configured turbine of the present invention can generate counter-rotating vortices on the suction surface. These vortices can effectively accelerate the momentum exchange between the boundary layer and the mainstream, thereby significantly suppressing the flow separation phenomenon under the operating environment of large angle of attack and low Reynolds number. Flow separation is one of the main reasons for the decline in turbine performance. By reducing separation, flow losses can be significantly reduced and turbine efficiency can be improved. The design of the present invention is particularly suitable for those situations where it works at a large angle of attack, such as take-off and climb phases, which place extremely high requirements on the performance of the turbine. At the same time, at high Mach numbers and designed angles of attack, the newly configured turbine can still maintain excellent performance, and will not cause additional losses due to the generation of counter-rotating vortices, thereby improving adaptability and stability under a wide range of working conditions.

[0090] (2) Compared with the method in the prior art that adjusts the cross-sectional shape of the blade to make the suction surface curve of the blade produce circumferential periodic fluctuations and then construct a suction surface with wave-like characteristics, the present invention does not change the blade profile section, but makes the blade profile section slide periodically in the axial direction and expand in the span direction to realize the construction of the wave characteristics of the blade surface. Wave characteristics exist on both the leading edge and the trailing edge of the blade, and there is a conversion position for the wave characteristics on the suction surface, that is, before and after the conversion position, the crest characteristics will be converted into trough characteristics, and the trough characteristics will be converted into crest characteristics. Under the induction of the leading edge wave characteristics, the position where the suction surface generates counter-rotating vortices is more forward, and counter-rotating vortices are also generated on the pressure surface, which can suppress the separation of the pressure surface under negative angles of attack and broaden the range of angles of attack that the blade can adapt to. In addition, the wave characteristics of the trailing edge have the effect of accelerating the crushing and mixing of the Karman vortex street generated by the trailing edge of the blade, avoiding the adverse effects of the large vortex structure generated by the trailing edge of the blade on the downstream.

[0091] (3) The design method of the present invention is based on a deep understanding and mastery of the flow mechanism of the new turbine configuration, and can more accurately predict and control the flow characteristics inside the turbine. Through the understanding of the flow mechanism, we can better understand the complex flow phenomena inside the turbine, such as boundary layer development, flow separation and reattachment, which are key factors affecting turbine performance. On this basis, more effective flow control strategies can be developed, such as optimizing flow characteristics by adjusting parameters such as blade profile, blade installation angle and number of blades, thereby improving the accuracy of the design.

[0092] (4) The traditional turbine design system often relies on empirical formulas and simplified flow models, which to a certain extent limits the flexibility and innovation of the design. The present invention broadens the traditional turbine design system by introducing new design concepts and technical means, such as bionic principles, computational fluid dynamics simulation, and machine learning optimization algorithms. This method not only improves the accuracy of the design, but also provides new ideas and methods for the future development of turbine component design. For example, the application of bionic principles creates more efficient and reliable turbine blades; computational fluid dynamics simulation can simulate complex flow conditions in computers and optimize design parameters; machine learning optimization algorithms can process large amounts of data and quickly find the optimal solution. The parts not elaborated in detail in the present invention are common knowledge in the field.

[0093] Through the above embodiments, the purpose of the present invention is fully and effectively achieved. Those skilled in the art can understand that the present invention includes but is not limited to the contents described in the drawings and the above specific embodiments. Although the present invention has been described with respect to the most practical and preferred embodiments currently considered, it should be understood that the present invention is not limited to the disclosed embodiments, and any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. A high-load wide-operating condition turbine blade shaping method, characterized in that: The following steps are involved: SS1, determine the reference blade profile: select the high-load low-pressure turbine blade profile as the reference blade profile, and use trigonometric functions to define several wave equations that control the position of the blade profile along the span direction; SS2, interpolation to generate a benchmark new configuration bionic blade: perform spline interpolation on the blade profile on each spanwise section to generate a benchmark new configuration bionic blade; SS3, design parameter determination: determine the key design parameters, including pitch P, amplitude A and blade leading edge radius r; SS4, establish a multi-objective optimization model: minimize the total pressure loss coefficient at the turbine blade cascade outlet under different attack angles as the optimization goal, and conduct optimization in a wide range of attack angles, wide Mach numbers and wide Reynolds numbers; SS5, machine learning global optimization: uses machine learning algorithms for global optimization, generates multiple new sample points in each iteration, and calculates the objective function value of each sample point; SS6, sample point sorting and updating: sort the sample points according to the objective function value, select the best sample point as the starting point of the next iteration, and continuously update the model until the termination condition is met; SS7, re-establishing the bionic blade: feedback the optimization results to the benchmark new configuration bionic blade, and finally output the optimal wide-operating condition new configuration turbine blade; Among them, in step SS4, the blade profile loss of the blade is comprehensively calculated by weighted averaging. When weighted averaging, the blade profile loss of the design angle of attack is reduced in weight. The calculation formula is: Where Yp is the blade loss, and the calculation formula is: Where, P in,T , P in,s and They correspond to the total pressure, static pressure at the mid-section of the inlet and the average total pressure of a single periodic section at 1.4 times the axial chord length of the blade outlet.

2. A high-load wide-operating condition turbine blade shaping method as claimed in claim 1, characterized in that: By changing the pitch P and amplitude A of the new turbine configuration and taking the comprehensive blade profile loss as the objective function, the key parameter selection criteria of the new turbine configuration are calculated; the specific steps are as follows: SS51, initial setting: Set a reasonable range of variation for pitch P and amplitude A based on previous design experience, theoretical analysis or factors affecting turbine performance; SS52, Create a parametric design model: Use numerical simulation tools to create a parametric turbine blade model in which the blade geometry is automatically adjusted based on the changes in pitch P and amplitude A. SS53, Systematic variation of design parameters: The values ​​of pitch P and amplitude A are systematically varied according to a strategy predetermined by the optimization algorithm; a series of blade designs are generated; SS54, Performance Evaluation and Loss Calculation: Numerical simulations are performed on each design to evaluate its aerodynamic performance. The purpose of the simulations is to calculate the flow characteristics and profile losses of the blade under different operating conditions. Based on the simulation results, the profile losses are weighted averaged at different angles of attack using formula (1) to obtain a comprehensive loss evaluation. SS55, Optimization process: Use an optimization algorithm to find the values ​​of pitch P and amplitude A that minimize the blade's comprehensive blade profile loss; the optimization algorithm iteratively updates the parameter values ​​based on the gradient of the loss function or through a search strategy, gradually approaching the optimal solution; SS56, determination of key parameter selection criteria: Analyze the data obtained during the optimization process to identify the values ​​of pitch P and amplitude A under which the blade comprehensive blade loss is the smallest; based on these data, establish key parameter selection criteria to quickly find the optimal blade design parameters; SS57, Validation and Iteration: Select the optimal parameter combination for additional detailed simulation and / or experimental validation to ensure the effectiveness of these parameters in actual applications; based on the validation results, iteratively adjust the parameter selection criteria to further improve the performance of the blade design.

3. A high-load wide-operating condition turbine blade shaping method as claimed in claim 2, characterized in that: In the optimization process, the gradient descent method is used to find the values ​​of the pitch P and amplitude A that minimize the comprehensive blade profile loss of the blade; specifically, the following steps are included: SS551, calculate the objective function and its gradient: Under the current parameter point, use computational fluid dynamics simulation or other numerical methods to calculate the comprehensive blade profile loss, that is, the value of the objective function; calculate the gradient of the objective function with respect to pitch P and amplitude A; this gradient points to the direction where the loss increases fastest, and update the parameters in the opposite direction of the gradient to reduce the loss; SS552, determine the step size and update parameters: choose an appropriate step size, which determines the magnitude of the parameter update in each iteration; update the current parameter value based on the calculated gradient; the new value of the parameter is obtained by subtracting the gradient from the current value multiplied by the step size; SS553, check convergence: after each iteration, check the change of old and new parameter values ​​and the change of objective function value; if the change is less than the preset threshold, or reaches a certain number of iterations, the algorithm is considered to have converged, and the parameter value at this time is the optimal solution; if the algorithm has not converged, continue the iteration step until the convergence condition is met; SS554, iterative process: repeat steps SS551 to SS553 until convergence; in each iteration, the parameter value is updated in the opposite direction of the gradient, gradually approaching the minimum value of the objective function; SS555, Result Verification: Once the optimal parameter values ​​are found, additional simulations are performed using these parameters to ensure that the blade's comprehensive profile losses are indeed minimized under actual operating conditions.

4. A high-load wide-operating condition turbine blade shaping method as claimed in claim 1, characterized in that: The trigonometric function used in step SS1 is a sine function or a cosine function, which is used to accurately control the position change of the blade profile along the span direction.

5. A high-load wide-operating condition turbine blade shaping method as claimed in claim 1, characterized in that: In step SS2 , spline interpolation uses a cubic or higher-order spline function to generate a smooth and continuous reference new-configuration bionic blade profile.

6. A high-load wide-operating condition turbine blade shaping method as claimed in claim 1, characterized in that: The following steps are also included: A special coating is applied to the surface of the newly configured turbine blades to reduce the friction coefficient of the blade surface and further improve the aerodynamic efficiency of the turbine.

7. A high-load wide-operating condition turbine blade shaping method as claimed in claim 1, characterized in that: The following steps are also included: Internal cooling channels are designed for the blades to improve performance and durability in high temperature environments.

Images