Multi-working-condition multi-objective optimization design method for bucket structure of impulse turbine

Abstract

The invention discloses a multi-working-condition multi-objective optimization design method for a water bucket structure of an impulse water turbine, which comprises the following steps: constructing a three-dimensional numerical model of an original water bucket structure of the impulse water turbine, and obtaining parameterized optimization design variables; modeling is conducted through the parameterized optimization design variables, a three-dimensional model of the impulse turbine runner bucket subjected to preliminary optimization design is obtained, the coordinate change direction and range of a control point of the three-dimensional model are restrained, and the restrained optimization design variables are obtained; according to the parameterized optimization design variable and the constrained optimization design variable, constructing a sample space of optimization design, and constructing a response surface model between the constrained optimization design variable and the optimization objective function; and performing global optimization on the response surface model to obtain an optimal solution set, and obtaining an optimal solution from the optimal solution set. According to the method, actual engineering requirements are considered when the optimization parameter range is set, and the phenomenon that the bucket blade is excessively deformed in the multi-working-condition optimization design process is effectively avoided.

Description

Technical Field

[0001] The invention relates to the technical field of structural optimization design of key flow-through components of hydraulic machinery, and in particular to a multi-operating-condition and multi-objective optimization design method for a bucket structure of an impulse turbine. Background Art

[0002] Due to its simple structure, high applied head and rapid operating condition conversion, the impulse turbine has been widely used in the development of high-head hydraulic resources in my country, and is constantly developing in the direction of high head, large capacity and gigantism.

[0003] Previous studies have found that in areas with fragile rock structures, affected by the scouring of high-energy water flow, the riverbed erosion is strong and the river sediment concentration is high. When the high-speed sand-containing jet impacts the bucket and completes the energy conversion process, the surface of the turbine flow components has a strong erosion effect, and the water-dividing edge line and working surface curvature of the bucket blade change significantly, which not only reduces the turbine operation efficiency, but also causes strong sediment abrasion and cavitation erosion, and may even cause changes in the dynamic stress characteristics, natural frequency and fatigue life of the runner, leading to turbine structural failure. In particular, the hydraulic units responsible for regulating the load fluctuations of the power grid need to change the unit operation status by a large value in a short period of time, and the units will frequently cross the unstable operation area, which puts forward new requirements for the safe and stable operation of the turbine. Therefore, when optimizing the structure of the runner bucket of the impulse turbine, it is necessary to fully consider the wide load operation conditions of the turbine in the sediment-rich river, and comprehensively consider the hydraulic efficiency, sediment abrasion and cavitation characteristics of the turbine under different working conditions.

[0004] In the past, the structural optimization of impulse turbines was mainly based on empirical data obtained during the operation of hydropower stations. However, the quantitative prediction of the deformation of the impeller of impulse turbines after sediment erosion is insufficient, and the transient characteristics of the multiphase flow inside the turbine are significant. Therefore, the research on the impact of sediment erosion on the hydraulic characteristics and structural reliability of turbines is still insufficient, resulting in a lack of reliable basis for the fatigue life and bucket crack prediction of impulse turbine units, which cannot meet the current situation of maintaining wide load operation for a long time in sediment-rich rivers. The development of intelligent optimization algorithms has overcome the limitations of empirical data to a certain extent, which can effectively reduce the experience requirements for designers and improve the design level, thereby completing the design of turbine bucket structures with high quality in a short time. Therefore, how to optimize the design of turbine bucket structures with the help of intelligent optimization algorithms and improve the design level of impulse turbines is a technical problem that must be solved at present. Summary of the invention

[0005] In view of the fact that the current existing design methods are backward and cannot meet the long-term safe operation of impulse turbines in a wide load range in sediment-rich rivers, resulting in low turbine operation efficiency, severe abrasion of flow-through components, and significant cavitation, the present invention proposes a multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine to improve the abrasion resistance of the bucket surface, inhibit the occurrence of cavitation, and enable the impulse turbine to remain safe and stable under wide load operating conditions in sediment-rich rivers.

[0006] To achieve the above object, the present invention provides a multi-condition and multi-objective optimization design method for a bucket structure of an impulse turbine, comprising:

[0007] Constructing a three-dimensional numerical model of the original bucket structure of the impulse turbine, dividing the three-dimensional numerical model into the front and back geometric structures of the bucket, obtaining parameterized optimization design variables, and modeling through the parameterized optimization design variables to obtain constrained optimization design variables;

[0008] According to the parameterized optimization design variables and the constrained optimization design variables, a sample space of optimization design is constructed, hydraulic performance numerical calculation is performed on each impulse turbine runner bucket model in the sample space of optimization design, and an optimization objective function for each impulse turbine operating under different operating conditions is established;

[0009] Constructing a response surface model between the post-constraint optimization design variables and the optimization objective function, and judging whether the response surface model meets the design requirements through the optimization objective function;

[0010] A global optimization is performed on the response surface model that meets the design requirements to obtain an optimal solution set, and an optimal solution is obtained from the optimal solution set. The optimal solution corresponds to a multi-condition and multi-objective optimized impulse turbine runner bucket model under wide load operating conditions in a sediment-rich river.

[0011] Preferably, obtaining the parameterized optimized design variables comprises:

[0012] The front and back geometric structures of the original bucket of the impulse turbine runner are divided by two clusters of equally spaced and orthogonally distributed spline curves, forming (n+1)×(m+1) intersection points P in the design space. ij , with the spline curve intersection point P ij As the control points of the front and back geometric structures of the impulse turbine runner bucket, the geometric distribution characteristics of the bucket are characterized to obtain the parameterized optimization design variables.

[0013] Preferably, optimizing the design variables after obtaining the constraints includes:

[0014] Modeling is performed based on the parameterized optimized design variables to obtain a three-dimensional model of a runner bucket of a impulse turbine after preliminary optimized design, and a control point coordinate change direction and range of the three-dimensional model of the runner bucket of the impulse turbine is constrained to obtain the constrained optimized design variables;

[0015] The direction and range of the coordinate change of the control points of the three-dimensional model of the impulse turbine runner bucket are achieved by constraining the coordinates of the (n+1)×(m+1) intersection points.

[0016] Preferably, constructing the sample space of the optimal design includes:

[0017] According to the parameterized optimized design variables and the constrained optimized design variables, a plurality of groups of sample points are generated through a Latin hypercube sampling experimental design method to form the sample space, and a plurality of impulse turbine runner bucket models are obtained through the sample space.

[0018] Preferably, performing numerical calculation of hydraulic performance of each impulse turbine runner bucket model in the sample space of the optimal design includes:

[0019] Each impulse turbine runner structure in the sample space is meshed in batches, and the remaining components of each impulse turbine are geometrically modeled and meshed. The full-flow channel numerical simulation discrete models of all impulse turbines under different operating conditions are obtained, and the discrete models including all flow-through components of the impulse turbine are numerically calculated to obtain the turbine operating characteristics of each impulse turbine under no-load conditions, partial load conditions, and rated conditions.

[0020] Preferably, establishing the optimization objective function for each impulse turbine to operate under different operating conditions includes:

[0021] Based on the geometric mean super transfer approximation method, the weight coefficients of hydraulic efficiency, abrasive characteristics and cavitation characteristics in the optimization objective function under different working conditions are determined respectively.

[0022] Preferably, the response surface model is a load Kriging model constructed using polynomials.

[0023] Preferably, judging whether the response surface model meets the design requirements by the optimization objective function includes:

[0024] Generate several groups of test samples according to the Latin hypercube sampling method, use the test samples to verify the accuracy of the response surface model, compare the optimization objective function value predicted by the response surface model with the numerical calculation results of the test samples, if the comparison result deviation is greater than the preset value, the accuracy of the response surface model does not meet the requirements, increase the number of samples to the sample space, re-calculate the numerical value and establish a new response surface model until the accuracy meets the optimization design requirements; if the comparison result is less than or equal to the preset value, output the response surface model.

[0025] Preferably, a global optimization is performed on the response surface model that meets the design requirements to obtain the optimal solution set including:

[0026] A multi-objective intelligent optimization algorithm is used to perform global optimization on the response surface model to obtain a global optimal solution set, which is in the form of a Pareto optimal solution set.

[0027] Preferably, obtaining the optimal solution from the optimal solution set includes:

[0028] The MOGA genetic algorithm is selected for optimization design, and the Pareto solution percentage range is set as the standard for determining whether it is the optimal solution, specifically:

[0029] Step a, set the population size and randomly generate the initial population, that is, the initial sample number;

[0030] Step b, set the number of iterative samples and calculate the objective function value of each sample;

[0031] Step c: Select, crossover and mutate according to the fitness of each sample point to generate a new population;

[0032] Step d: Check whether it converges. If not, go to step b. If it converges, go to step e.

[0033] Step e, output the Pareto optimal solution set in the candidate solution set and the related optimization objective function value;

[0034] Step f: Select the optimal solution from the Pareto solution set according to the actual situation.

[0035] Compared with the prior art, the present invention has the following advantages and technical effects:

[0036] (1) The present invention provides a multi-condition and multi-objective optimization design method for a bucket structure of an impulse turbine, which obtains (n+1)×(m+1) control points P representing the distribution of the runner bucket surface through two clusters of equally spaced and orthogonally distributed spline curves. ij(i=0,1,2,,n;j=0,1,2,,m) are used as optimization design variables; the range of control point coordinates is constrained to obtain the sample space of optimization design. Combined with the water, gas and sand three-phase numerical simulation, the modified runner bucket model is numerically calculated to analyze its performance indicators such as hydraulic efficiency, abrasion distribution and cavitation characteristics. According to the actual situation of the power station, the weight coefficient of each performance indicator is determined, and the optimization objective function is constructed to determine whether the response surface model meets the design requirements; the response surface model is globally optimized to obtain a solution set with improved performance indicators, that is, a multi-condition, multi-objective optimization design model for the runner bucket when operating at wide load in a sediment-rich river.

[0037] (2) The present invention takes into account the actual needs of the project when setting the optimization parameter range, effectively avoiding the bucket blades with excessive deformation during the multi-operating condition optimization design process. In the optimization design process, a method combining a response surface model and a multi-objective genetic algorithm is used to perform global optimization on the runner blades of the impulse turbine, forming a Pareto optimum, avoiding falling into a local optimum, shortening the design cycle, and reducing the design cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0039] Figure 1 A schematic diagram of the structure of an impulse turbine according to an embodiment of the present invention;

[0040] Figure 2 This is a flow chart of optimizing the runner bucket structure according to an embodiment of the present invention;

[0041] Figure 3 Schematic diagram of the position of the control points on the bucket surface according to an embodiment of the present invention;

[0042] Figure 4 It is a schematic cross-sectional diagram of the middle part of a water bucket according to an embodiment of the present invention;

[0043] Figure 5 Schematic diagram of the bucket structure before and after optimization of an embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0045] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] This embodiment uses the hydraulic efficiency, abrasion distribution and cavitation characteristics of the turbine under no-load conditions, partial load conditions and rated conditions as the objective function. The front and back curved surface structures of the runner bucket are the geometric parameters to be optimized. Under the condition that no excessive geometric deformation occurs, the design optimization of the runner bucket structure is carried out in combination with the actual project. Figure 1 The specific optimization process is as shown in Figure 2 shown.

[0047] A multi-condition and multi-objective optimization design method for a bucket structure of an impulse turbine, comprising:

[0048] Constructing a three-dimensional numerical model of the original bucket structure of the impulse turbine, dividing the three-dimensional numerical model into the front and back geometric structures of the bucket, obtaining parameterized optimization design variables, and modeling through the parameterized optimization design variables to obtain constrained optimization design variables;

[0049] According to the parameterized optimization design variables and the constrained optimization design variables, a sample space of optimization design is constructed, hydraulic performance numerical calculation is performed on each impulse turbine runner bucket model in the sample space of optimization design, and an optimization objective function for each impulse turbine operating under different operating conditions is established;

[0050] Constructing a response surface model between the post-constraint optimization design variables and the optimization objective function, and judging whether the response surface model meets the design requirements through the optimization objective function;

[0051] A global optimization is performed on the response surface model that meets the design requirements to obtain an optimal solution set, and an optimal solution is obtained from the optimal solution set. The optimal solution corresponds to a multi-condition and multi-objective optimized impulse turbine runner bucket model under wide load operating conditions in a sediment-rich river.

[0052] This embodiment proposes a new multi-objective optimization design method for the bucket structure of an impulse turbine, which can quickly and accurately improve the operating efficiency of the turbine according to the specific conditions of different power stations, improve the abrasion resistance of the bucket surface, inhibit cavitation, and enable the impulse turbine to remain safe and stable under wide load operating conditions in sediment-rich rivers.

[0053] Furthermore, the parameterized optimization design variables include:

[0054] The front and back geometric structures of the original bucket of the impulse turbine runner are divided by two clusters of equally spaced and orthogonally distributed spline curves, as shown in Figure 3 As shown, (n+1)×(m+1) intersection points P are formed in the design space. ij , with the spline curve intersection point P ij As the control point of the front and back geometric structures of the runner bucket of the impulse turbine, the front and back geometric structures of the bucket are characterized to obtain the parameterized optimization design variables.

[0055] Furthermore, the optimized design variables after obtaining the constraints include:

[0056] Modeling is performed based on the parameterized optimized design variables to obtain a three-dimensional model of the runner bucket of the impulse turbine after preliminary optimized design, and the direction and range of the change of the coordinates of the control points of the three-dimensional model of the runner bucket of the impulse turbine are constrained to obtain the constrained optimized design variables;

[0057] The change direction and range of the coordinates of the control points of the three-dimensional model of the impulsive turbine runner bucket are realized by constraining the coordinates of the (n+1)×(m+1) intersection points. In this embodiment, the change range is constrained within Z ij ±5mm, Z ij is the ordinate of the control point in row i and column j.

[0058] Specifically, the results of the runner bucket parameterization are verified through reverse modeling. In order to better control the deformation of the runner bucket geometry, ensure the smoothness of the bucket surface, and avoid excessive deformation of the bucket structure, it is necessary to constrain the variation range of the optimization design variables. Figure 4 Shown is a schematic diagram of the structure of a section in the middle of the water bucket.

[0059] The optimized design variables after parameterization were remodeled, and the fitting degree of the initial blade and the remodeled bucket blade was compared. After verification, the parameterization did not distort the shape of the runner bucket. The variation range of the runner bucket structure is realized by constraining the coordinates of (n+1)×(m+1) control points. The variation range is constrained in Z ij ±5mm.

[0060] Furthermore, the weighted average efficiency characterized by the hydraulic efficiency, abrasion distribution and cavitation characteristics of the impulse turbine unit during operation is used as the optimization target to evaluate the performance of each impulse turbine runner in the sample space. The most important step is to reasonably determine the weight factor.

[0061] In this embodiment, the weighted average efficiency of the impulse turbine with different operating characteristics is taken as the optimization target, that is, the optimization objective function is: η max =ω w η w +ωe η e +ω v η v Among them, η max is the maximum value of the weighted average efficiency of the unit operation characteristics, ω w is the weight coefficient representing the hydraulic efficiency, ω e is the weight factor characterizing the abrasion distribution, ω v is the weight factor characterizing the cavitation characteristics, η w is the functional expression for hydraulic efficiency, η e is the functional expression that characterizes the abrasion distribution, η v is the function expression that characterizes the cavitation characteristics.

[0062] Furthermore, constructing the sample space of the optimal design includes:

[0063] According to the parameterized optimized design variables and the constrained optimized design variables, a plurality of groups of sample points are generated through a Latin hypercube sampling experimental design method to form the sample space, and a plurality of impulse turbine runner bucket models are obtained through the sample space.

[0064] Furthermore, the hydraulic performance numerical calculation of each impulse turbine runner bucket model in the optimally designed sample space includes:

[0065] Each impulse turbine runner structure in the sample space is meshed in batches, and the remaining components of each impulse turbine are geometrically modeled and meshed. The full flow channel numerical simulation discrete models of all impulse turbines under different operating conditions are obtained, and the discrete models including all flow-through components of the impulse turbine are numerically calculated to obtain the turbine operating characteristics of each impulse turbine under no-load conditions, partial load conditions, and rated conditions.

[0066] Specifically, the numerical analysis of the Pelton turbine runner bucket performance evaluation includes two basic processes:

[0067] 1. Discretize the computational domain, that is, divide the computational domain into grids:

[0068] 2. Numerical calculation of the flow field inside the turbine runner, that is, using computational fluid dynamics theory to calculate the flow field of the entire turbine runner.

[0069] The optimization objective function for each impulse turbine operating under different operating conditions includes:

[0070] Based on the geometric mean super transfer approximation method, the weight coefficients of hydraulic efficiency, abrasive characteristics and cavitation characteristics in the optimization objective function under different working conditions are determined respectively.

[0071] Specifically, based on the geometric mean super transfer approximation method, the weight coefficient of the impulse turbine under different working conditions is determined, which is specifically achieved by the following method:

[0072] Firstly, the relative importance of different working conditions is evaluated by pairwise comparison to generate a binary complementary matrix; the supertransfer approximate matrix is ​​constructed using the binary complementary matrix; finally, the maximum eigenvalue of the supertransfer approximate matrix is ​​calculated using the eigenvector method, and the eigenvector corresponding to the maximum eigenvalue is the weight coefficient of each working condition.

[0073] Specifically, in this embodiment, the response surface model is a load Kriging model constructed using polynomials.

[0074] Further, judging whether the response surface model meets the design requirements by the optimization objective function includes:

[0075] Generate several groups of test samples according to the Latin hypercube sampling method, use the test samples to verify the accuracy of the response surface model, compare the optimization objective function value predicted by the response surface model with the numerical calculation results of the test samples, if the comparison result deviation is greater than the preset value, the accuracy of the response surface model does not meet the requirements, increase the number of samples to the sample space, re-calculate the numerical value and establish a new response surface model until the accuracy meets the optimization design requirements; if the comparison result is less than or equal to the preset value, output the response surface model.

[0076] Specifically, 10 groups of test samples are generated according to the Latin hypercube sampling method, and the test samples are used to test the accuracy of the established response surface model. The optimization objective function value predicted by the response surface model is compared with the numerical calculation results of the test samples. If the deviation between the two results is greater than 5%, the accuracy of the response surface model does not meet the requirements, and it is necessary to increase the number of samples to the sample space, re-calculate the numerical calculation and establish a new response surface model until its accuracy meets the optimization design requirements; if the accuracy meets the requirements, the response surface model is output.

[0077] Furthermore, a global optimization is performed on the response surface model that meets the design requirements, and the optimal solution set obtained includes:

[0078] A multi-objective intelligent optimization algorithm is used to perform global optimization on the response surface model to obtain a global optimal solution set, which is in the form of a Pareto optimal solution set.

[0079] Specifically, obtaining the optimal solution from the optimal solution set includes:

[0080] The MOGA genetic algorithm is selected for optimization design, and the Pareto solution percentage range is set as the standard for determining whether it is the optimal solution, specifically:

[0081] Step a, set the population size and randomly generate the initial population, that is, the initial sample number;

[0082] Step b, set the number of iterative samples and calculate the objective function value of each sample;

[0083] Step c: Select, crossover and mutate according to the fitness of each sample point to generate a new population;

[0084] Step d: Check whether it converges. If not, go to step b. If it converges, go to step e.

[0085] Step e, output the Pareto optimal solution set in the candidate solution set and the related optimization objective function value;

[0086] Step f: Select the optimal solution from the Pareto solution set according to the actual situation.

[0087] According to the solution of the optimal bucket structure, a group of bucket structures with improved hydraulic efficiency, improved abrasion distribution and suppressed cavitation are selected as the final optimization results. The runner bucket structures before and after optimization are shown in the following figure: Figure 5 shown.

[0088] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-condition and multi-objective optimization design method for a bucket structure of an impulse turbine, characterized in that: include: Constructing a three-dimensional numerical model of the original bucket structure of the impulse turbine, dividing the three-dimensional numerical model into the front and back geometric structures of the bucket, obtaining parameterized optimization design variables, and modeling through the parameterized optimization design variables to obtain constrained optimization design variables; According to the parameterized optimization design variables and the constrained optimization design variables, a sample space of optimization design is constructed, hydraulic performance numerical calculation is performed on each impulse turbine runner bucket model in the sample space of optimization design, and an optimization objective function for each impulse turbine operating under different operating conditions is established; Constructing a response surface model between the post-constraint optimization design variables and the optimization objective function, and judging whether the response surface model meets the design requirements through the optimization objective function; A global optimization is performed on the response surface model that meets the design requirements to obtain an optimal solution set, and an optimal solution is obtained from the optimal solution set. The optimal solution corresponds to a multi-condition and multi-objective optimized impulse turbine runner bucket model under wide load operating conditions in a sediment-rich river.

2. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 1, characterized in that: The parameterized optimization design variables include: The front and back geometric structures of the original bucket of the impulse turbine runner are divided by two clusters of equally spaced and orthogonally distributed spline curves, forming (n+1)×(m+1) intersection points P in the design space. ij , with the spline curve intersection point P ij As the control points of the front and back geometric structures of the impulse turbine runner bucket, the geometric distribution characteristics of the bucket are characterized to obtain the parameterized optimization design variables.

3. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 2 is characterized in that: After obtaining the constraints, the optimized design variables include: Modeling is performed based on the parameterized optimized design variables to obtain a three-dimensional model of a runner bucket of a impulse turbine after preliminary optimized design, and a control point coordinate change direction and range of the three-dimensional model of the runner bucket of the impulse turbine is constrained to obtain the constrained optimized design variables; The direction and range of the coordinate change of the control points of the three-dimensional model of the impulse turbine runner bucket are achieved by constraining the coordinates of the (n+1)×(m+1) intersection points.

4. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 1, characterized in that: The sample space for constructing the optimization design includes: According to the parameterized optimized design variables and the constrained optimized design variables, a plurality of groups of sample points are generated through a Latin hypercube sampling experimental design method to form the sample space, and a plurality of impulse turbine runner bucket models are obtained through the sample space.

5. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 4, characterized in that: The hydraulic performance numerical calculation of each impulse turbine runner bucket model in the sample space of the optimal design includes: Each impulse turbine runner structure in the sample space is meshed in batches, and the remaining components of each impulse turbine are geometrically modeled and meshed. The full-flow channel numerical simulation discrete models of all impulse turbines under different operating conditions are obtained, and the discrete models including all flow-through components of the impulse turbine are numerically calculated to obtain the turbine operating characteristics of each impulse turbine under no-load conditions, partial load conditions, and rated conditions.

6. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 5, characterized in that: The optimization objective function for each impulse turbine operating under different operating conditions includes: Based on the geometric mean super transfer approximation method, the weight coefficients of hydraulic efficiency, abrasive characteristics and cavitation characteristics in the optimization objective function under different working conditions are determined respectively.

7. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 1, characterized in that: The response surface model is a load Kriging model constructed using polynomials.

8. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 7, characterized in that: Judging whether the response surface model meets the design requirements by the optimization objective function includes: Generate several groups of test samples according to the Latin hypercube sampling method, use the test samples to verify the accuracy of the response surface model, compare the optimization objective function value predicted by the response surface model with the numerical calculation results of the test samples, if the comparison result deviation is greater than the preset value, the accuracy of the response surface model does not meet the requirements, increase the number of samples to the sample space, re-calculate the numerical value and establish a new response surface model until the accuracy meets the optimization design requirements; if the comparison result is less than or equal to the preset value, output the response surface model.

9. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 8, characterized in that: The response surface model that meets the design requirements is globally optimized, and the optimal solution set obtained includes: A multi-objective intelligent optimization algorithm is used to perform global optimization on the response surface model to obtain a global optimal solution set, which is in the form of a Pareto optimal solution set.

10. The multi-condition and multi-objective optimization design method for the bucket structure of an impulse turbine according to claim 9, characterized in that: Obtaining the optimal solution from the optimal solution set includes: The MOGA genetic algorithm is selected for optimization design, and the Pareto solution percentage range is set as the standard for determining whether it is the optimal solution, specifically: Step a, set the population size and randomly generate the initial population, that is, the initial sample number; Step b, set the number of iterative samples and calculate the objective function value of each sample; Step c: Select, crossover and mutate according to the fitness of each sample point to generate a new population; Step d: Check whether it converges. If not, go to step b. If it converges, go to step e. Step e, output the Pareto optimal solution set in the candidate solution set and the related optimization objective function value; Step f: Select the optimal solution from the Pareto solution set according to the actual situation.

Images