An archimedes bidirectional spiral water turbine and its guide cover optimization method

Abstract

The application provides an Archimedes bidirectional spiral water turbine and an optimization method of a fairing thereof, and belongs to the technical field of water turbines.The bidirectional flow water turbine comprises a central hub and three spiral blades which are uniformly arranged on the central hub in a circumferential direction, the three spiral blades are arranged in an Archimedes spiral shape on the outer surface of the central hub, and the radial distance between the outer edges of the three spiral blades and the surface of the central hub gradually increases first and then gradually decreases along the axial length direction of the central hub, and the outer surface of the water turbine body is sleeved with a first symmetrical fairing or a second symmetrical fairing.The application has the following beneficial effects: the application focuses on bidirectional flow, is an application exploration of a bidirectional tidal current environment, and calculates the optimal structure parameters of the fairing, so that the application provides valuable insights for the optimal design of bidirectional flow turbines and fairings in a low flow environment.

Description

Technical Field

[0001] This invention relates to the field of Archimedes turbine technology, and in particular to an Archimedes bidirectional spiral turbine and its guide fairing optimization method. Background Technology

[0002] With the increasing global demand for renewable energy, ocean tidal energy has attracted much attention due to its high energy density and predictability. Ocean energy is typically manifested as low-speed ocean currents; for example, current velocities in most parts of the Bohai Sea are less than 0.77 m / s, while those in the Yellow Sea range from 0.5 m / s to 1.0 m / s. In recent years, Archimedes turbines have become a research hotspot due to their excellent self-starting capability and stable operation even at low current velocities. To improve energy conversion efficiency, researchers are optimizing turbine structures and combining tidal flow turbines with flow deflectors.

[0003] In summary, current research on Archimedes' turbines (ASHTs) primarily focuses on unidirectional flow, lacking exploration of applications in bidirectional tidal environments. Currently, both the Archimedes' turbine (ASHT) and its associated fairing are designed for unidirectional stable flow, exhibiting significant asymmetric characteristics. However, in semi-diurnal tidal areas, the tidal direction changes twice per cycle. Traditional unidirectional configurations experience a drastic performance drop or even fail to operate under reverse flow conditions, severely limiting the turbine's full-cycle energy capture efficiency in real tidal energy environments. Furthermore, the parameter space coupling between the rotor and the fairing is complex, lacking efficient systematic global optimization methods. Existing fairing optimization often remains at the stage of single-variable analysis or empirical trial and error, failing to fully consider the strong nonlinear coupling relationship between rotor geometric parameters and fairing profiles. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide an Archimedes bidirectional spiral turbine and its guide fairing optimization method, which solves the problems that current research on Archimedes spiral turbines mainly focuses on unidirectional flow and lacks application exploration in bidirectional tidal flow environments, as well as the complex parameter space coupling between the rotor and the guide fairing and the lack of an efficient systematic global optimization method.

[0005] To achieve the above and other related objectives, the present invention provides the following technical solution:

[0006] An Archimedes-style bidirectional spiral turbine includes a bidirectional flow turbine, wherein the bidirectional flow turbine includes a central hub and three spiral blades mounted on the central hub and evenly arranged along the circumference of the central hub. The three spiral blades are arranged in an Archimedes spiral shape around the outer surface of the central hub, and the radial distance between the outer edge of the three spiral blades and the surface of the central hub gradually increases and then gradually decreases along the axial length direction of the central hub. The outer surface of the turbine body is fitted with a first symmetrical guide shroud or a second symmetrical guide shroud.

[0007] In one embodiment of the present invention, the first symmetrical guide shield includes a first contraction section sleeved on the bidirectional flow turbine for accelerating fluid flow and a first expansion section installed on the first contraction section for reducing fluid flow. The connection between the first contraction section and the first expansion section is a throat. The throat of the first symmetrical guide shield is on the same radial plane as the midpoint of the central hub.

[0008] In one embodiment of the present invention, the end of the first contraction section away from the first expansion section is provided with a first inlet, and the end of the first expansion section away from the first contraction section is provided with a first outlet.

[0009] In one embodiment of the present invention, the second symmetrical guide shield includes a second contraction section and a second expansion section sleeved on the outside of the bidirectional flow turbine and used to accelerate the fluid flow rate and reduce the fluid flow rate, respectively. A parallel section is installed between the second contraction section and the second expansion section, and the center of the parallel section is on the same radial plane as the midpoint of the central hub.

[0010] In one embodiment of the present invention, the end of the second contraction section away from the second expansion section is provided with a second inlet, and the end of the second expansion section away from the second contraction section is provided with a second outlet.

[0011] An optimization method for the fairing of an Archimedes bidirectional spiral turbine, based on the Archimedes bidirectional spiral turbine, includes the following steps: constructing a first bidirectional flow turbine three-dimensional model and a second bidirectional flow turbine three-dimensional model, and performing parametric modeling on the first and second bidirectional flow turbine three-dimensional models; determining the optimization objective of a single-objective optimization problem, and designing corresponding structural parameters and parameter ranges for the first and second symmetrical fairings respectively in the parametrically modeled first and second bidirectional flow turbine three-dimensional models, wherein the optimization objective is the power coefficient of the turbine;

[0012] Latin hypercube sampling is used to extract sample points within the parameter ranges of the first and second symmetrical fairings, and the objective function values ​​corresponding to the sample points are determined. A Kriging surrogate model is constructed based on the sample points and their corresponding power coefficients. The Kriging surrogate model is used to fit the sample points and objective function values ​​of the variables to an approximate model. A multi-island genetic algorithm is used to optimize the fitting results of the approximate model. The optimal structural parameters of the first and second symmetrical fairings are obtained based on the optimization results.

[0013] In one embodiment of the present invention, constructing the first and second bidirectional flow turbine three-dimensional models includes: modeling the bidirectional flow turbine, the first symmetrical guide vane, and the second symmetrical guide vane respectively using three-dimensional modeling software; obtaining three-dimensional models of the bidirectional flow turbine, the first symmetrical guide vane, and the second symmetrical guide vane based on the modeling results; and combining the three-dimensional model of the bidirectional flow turbine with the three-dimensional models of the first and second symmetrical guide vanes respectively, obtaining the first and second bidirectional flow turbine three-dimensional models based on the combination results.

[0014] In one embodiment of the present invention, in the step of using Latin hypercube sampling to extract sample points in the parameter ranges of the first symmetrical fairing and the second symmetrical fairing, and determining the objective function value corresponding to the sample points, the sample points are also design variables, and the Kriging surrogate model is used to map the design variables to the turbine power coefficient.

[0015] In one embodiment of the present invention, the step of constructing a Kriging surrogate model based on the sample points and their corresponding power coefficients, and fitting an approximate model to the sample points of the variables and the objective function value using the Kriging surrogate model, includes: training the initial Kriging surrogate model using the sample points and their corresponding power coefficients, obtaining the Kriging surrogate model based on the training results, fitting an approximate model to the sample points of the variables and the objective function using the Kriging surrogate model, and performing sensitivity analysis on the relationship between structural parameters.

[0016] In one embodiment of the present invention, after optimizing the fitting results of the approximate model using a multi-island genetic algorithm and obtaining the optimal structural parameters of the first symmetrical fairing and the second symmetrical fairing based on the optimization results, the method further includes: adding a flange structure to the optimal structural parameters of the fairing, and performing secondary optimization on the fairing structure with the added flange structure, and obtaining the optimal fairing scheme based on the secondary optimized structure.

[0017] As described above, the Archimedes bidirectional helical turbine and its fairing optimization method of the present invention have the following beneficial effects: The present invention compares the hydrodynamic characteristics of unidirectional and bidirectional turbines under the same structure using computational fluid dynamics (CFD) analysis. Representative sample points are generated in the design space using Latin hypercube sampling. A Kriging surrogate model is established based on these samples, and a multi-island genetic algorithm is used to optimize the approximate model, obtaining the optimal structural parameters of the fairing. Based on this, a flange is added to perform secondary optimization of the fairing structure. Simulation results verify that when the flange angle of the second symmetrical fairing is 70°, the combined operating efficiency with the turbine is increased by 27.2%, and the impeller's energy harvesting capacity is significantly improved. The results of this invention provide valuable insights for the optimized design of bidirectional flow turbines and fairings under low flow conditions. Attached Figure Description

[0018] Figure 1 The image shown is a three-dimensional model of a conventional unidirectional flow turbine according to the first embodiment of the present invention.

[0019] Figure 2 The diagram shows the structural parameters of a conventional unidirectional flow turbine as shown in the first embodiment of the present invention.

[0020] Figure 3 The image shown is a three-dimensional model of the Archimedes bidirectional spiral water turbine according to the first embodiment of the present invention.

[0021] Figure 4 This diagram illustrates the modeling process of the Archimedes bidirectional spiral water turbine according to the first embodiment of the present invention.

[0022] Figure 5 The diagram shown is a flow field analysis diagram of the Archimedes bidirectional spiral turbine in the first embodiment of the present invention.

[0023] Figure 6 This is a comparison diagram of the power coefficients of unidirectional and bidirectional flow turbines in the Archimedes bidirectional spiral turbine of the first embodiment of the present invention.

[0024] Figure 7 The image shown is a three-dimensional model of the assembly of the bidirectional flow turbine and the guide fairing in the Archimedes bidirectional spiral turbine according to the first embodiment of the present invention.

[0025] Figure 8 The diagram shown is a structural diagram of the base-type bidirectional flow turbine in the Archimedes bidirectional spiral turbine of the first embodiment of the present invention.

[0026] Figure 9 The diagram shows the structural parameters of the first symmetrical guide vane in the Archimedes bidirectional spiral turbine according to the first embodiment of the present invention.

[0027] Figure 10 The diagram shows the structural parameters of the second symmetrical guide vane in the Archimedes bidirectional spiral turbine according to the first embodiment of the present invention.

[0028] Figure 11 The diagram shows the overall flow chart of the guide fairing optimization method for the Archimedes bidirectional spiral turbine in the second embodiment of the present invention.

[0029] Figure 12 The diagram shows the overall watershed in the Archimedes bidirectional spiral turbine guide fairing optimization method according to the second embodiment of the present invention.

[0030] Figure 13 The diagram shown is a three-dimensional model of the fairing and turbine assembly in the fairing optimization method of the Archimedes bidirectional spiral turbine in the second embodiment of the present invention.

[0031] Figure 14 The optimization method for the guide vane of the Archimedes bidirectional spiral turbine in the second embodiment of the present invention is shown, which considers the influence of various factors on the overall performance. Bar chart showing the percentage distribution of impact;

[0032] Figure 15 The image shown is a Kriging model fitting diagram in the fairing optimization method of the Archimedes bidirectional spiral turbine in the second embodiment of the present invention.

[0033] Figure 16 The diagram shows a 3D scatter plot matrix of the genetic algorithm in the optimization method of the Archimedes bidirectional spiral turbine fairing in the second embodiment of the present invention, in which the optimal parameter points of the Type-A fairing are found by the genetic algorithm.

[0034] Figure 17 The image shown is a 3D scatter plot matrix of the genetic algorithm used in the optimization method for the Archimedes bidirectional spiral turbine fairing in the second embodiment of the present invention to find the optimal parameter points of the Type-B fairing.

[0035] 1. Unidirectional flow turbine; 11. Central hub of unidirectional flow turbine; 12. Three helical blades of unidirectional flow turbine; 2. Bidirectional flow turbine; 21. Central hub of bidirectional flow turbine; 22. Three helical blades of bidirectional flow turbine; 221. First helical blade; 222. Second helical blade; 223. Third helical blade; 3. First symmetrical guide fairing; 31. First contraction section; 32. First expansion section; 33. Throat; 34. First inlet; 35. First outlet; 4. Second symmetrical guide fairing; 41. Second contraction section; 42. Second expansion section; 43. Parallel section; 44. Second inlet; 45. Second outlet; 5. Permanent magnet generator; 6. Turbine support; 7. Support column; 8. Base. Detailed Implementation

[0036] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. It should be noted that, unless otherwise specified, the following embodiments and features described herein can be combined with each other.

[0037] The first embodiment of the present invention relates to an Archimedes bidirectional spiral water turbine, for details please refer to Figure 3 , Figure 7 and Figure 8 The system includes a bidirectional flow turbine 2, which includes a central hub 21 and three spiral blades 22 mounted on the central hub 21 and evenly arranged around the circumference of the central hub 21. In this embodiment, the three spiral blades 22 are a first spiral blade 221, a second spiral blade 222, and a third spiral blade 223. The three spiral blades 22 are arranged in an Archimedean spiral shape around the outer surface of the central hub 21, and the radial distance between the outer edge of the three spiral blades 22 and the surface of the central hub 21 gradually increases and then gradually decreases along the axial length direction of the central hub 21. The outer surface of the turbine body 1 is fitted with a first symmetrical guide shroud 3 or a second symmetrical guide shroud 4.

[0038] The first symmetrical flow guide shroud 3 includes a first contraction section 31 fitted outside the bidirectional flow turbine 2 for accelerating fluid flow rate and a first expansion section 32 installed on the first contraction section 31 for reducing fluid flow rate. The connection between the first contraction section 31 and the first expansion section 32 is a throat 33. The throat 33 of the first symmetrical flow guide shroud 3 is on the same radial plane as the midpoint of the central hub 21. The end of the first contraction section 31 away from the first expansion section 32 is provided with a first inlet 34, and the end of the first expansion section 32 away from the first contraction section 31 is provided with a first outlet 35.

[0039] The second symmetrical flow guide shroud 4 includes a second contraction section 41 and a second expansion section 42, which are fitted outside the bidirectional flow turbine 2 and are used to accelerate the fluid flow rate and reduce the fluid flow rate, respectively. A parallel section 43 is installed between the second contraction section 41 and the second expansion section 42. The center of the parallel section 43 is on the same radial plane as the midpoint of the central hub 21. The end of the second contraction section 41 away from the second expansion section 42 is provided with a second inlet 44, and the end of the second expansion section 42 away from the second contraction section 41 is provided with a second outlet 45.

[0040] Specifically, 1. Design of a bidirectional flow turbine: To adapt to the periodic reversal of ocean currents, this invention designs a novel horizontal-axis bidirectional flow turbine based on a unidirectional flow impeller; such as... Figure 1 and Figure 2As shown, the unidirectional flow turbine 1 is designed based on the Archimedes' spiral principle, mainly consisting of a central hub 11 and three spiral blades 12 (Blade-A, B, C) arranged in a circular array. The included angle at the top of the shaft is... The blade inclination angles are respectively , , and The blade spacing is respectively ;

[0041] Based on the above unidirectional model, the Archimedean spirals are arranged in a back-to-back symmetrical manner, and the tails of the unidirectional blades extend smoothly with the same curvature to achieve bidirectional symmetry of the impeller, ultimately forming a bidirectional Archimedean spiral impeller, also known as the bidirectional flow turbine 2. Let the turbine diameter be D, the hub diameter be d, and the length of the bidirectional flow turbine be L. The three-dimensional model of the bidirectional Archimedean spiral turbine is as follows: Figure 3 As shown, and for the formation process of the bidirectional Archimedes' screw turbine, please refer to [link / reference needed]. Figure 4 ; Figure 5 This diagram illustrates the hydrodynamics of the ASHT blade. The axial force (Fx) exerted by the water flow parallel to the axis of rotation can be decomposed into two core components: drag perpendicular to the blade surface (F...). D ) and lift (F) parallel to the blade surface L These two forces work together to drive the blades to rotate around the turbine's central axis;

[0042] Further analysis reveals that the lift force (F) experienced by ASHT L ) and resistance (F) D It can be decomposed into directional components parallel and perpendicular to the water flow, i.e., F LX F Ly F DX and F Dy These components, when superimposed, form the resultant force acting on the turbine blades, which can adapt to the energy extraction requirements under different wind conditions. Their interaction satisfies the following equation: , The rotational motion of ASHT is mainly caused by the vertical force component (F). Y Driven by the relative velocity (W) between the wind and the blades, this force is the result of the combined action of the absolute velocity (C), the blade linear velocity (U), and the angle (θ) between them. This hydrodynamic interaction repeats every 120° between the three helical blades, ensuring the continuity and smoothness of the turbine rotation. From the front view, the axial force (F)... X This can be further broken down into tangential forces (F). T ) and normal force (FN), tangential force (F TThe blade is responsible for converting the kinetic energy of the wind into mechanical energy, and is the core power source for power generation; the normal force (FN) acts on the blade structure, and the stress generated directly affects the long-term durability of the blade.

[0043] Based on the forces acting on the turbine in the flow field Figure 5 It can be seen that, due to the unique helical blade structure of this turbine, the rear blades cannot completely capture the optimal incident angle of the water flow; at the same time, the tips of the front blades will form vortices under the impact of the water flow, causing instability to the rear flow field. The main parameters of the bidirectional Archimedes helical turbine are shown in Table 1:

[0044] Table 1. Relevant Parameters of the Water Turbine

[0045]

[0046] The turbine designed in this invention inherits from the research object in the previous paper. To verify whether it conforms to the conclusions of the previous paper, CFD method was used to numerically simulate the turbine with different blade spacings γ. The boundary conditions and turbulence model used in the calculation process are consistent, such as... Figure 6 As shown, the water turbine designed in this invention is highly consistent with the relevant parameter conclusions obtained in the previous study, further confirming the applicability of the previous conclusions in this design scenario.

[0047] 2. Flow Deflector Design: The hydrodynamic performance of the energy harvesting device significantly depends on the precise control of key geometric parameters. Optimizing various parameters of the flow deflector can effectively improve the energy harvesting efficiency of the bidirectional turbine. Because the turbine designed in this invention exhibits high drag characteristics, significant flow channel blockage effects, and is highly sensitive to the directionality and uniformity of the incoming flow, the flow deflector design must strike a balance between flow acceleration and blockage suppression and loss reduction to prevent the interaction between the high-solidity blades and the flow deflector from deteriorating the flow field. Symmetrical flow deflectors are key components adapted to bidirectional tidal flow environments and are widely used in high-solidity drag turbines. Their core advantage lies in utilizing the symmetry of geometry and fluid dynamics to ensure efficient flow acceleration, drag reduction, and flow field stability in both forward and reverse flow conditions, effectively solving the problem of abrupt efficiency drops in reverse operation for asymmetric flow deflectors. The integrated device of the flow deflector and turbine is as follows: Figure 7 As shown, a fairing is also installed on base 8. Please refer to [link / reference]. Figure 8 In actual use, permanent magnet generators 5 are installed at both ends of the central hub 21 of the bidirectional flow turbine 2. The permanent magnet generators 5 are installed on the inner wall of the guide shroud through the turbine bracket 6, and the guide shroud is installed on the base 8 through the support column 7.

[0048] The geometric parameters of the symmetrical guide fairing must satisfy a strict mirror relationship. Based on the flow-converging characteristics of the guide fairing, this invention designs two flow-converging devices with different structures: a first symmetrical guide fairing 3 (contraction section-expansion section); and a second symmetrical guide fairing 4 (contraction section-parallel section-expansion section). CFD calculations are used to analyze the impact of each structure on the performance of the horizontal axis drag-type turbine. The overall length of the Type A fairing is... The overall length of the Type B fairing is... For the parallel section dimensions of the type B fairing, D A D B It is the inlet radius of the fairing, TC A TC B d1 and d2 are the clearance dimensions between the blade tip and the throat, and d1 and d2 are the radii of the deflector throat. Specific parameters are as follows: Figure 9 and Figure 10 As shown.

[0049] The second embodiment of the present invention relates to a method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine, the process of which is as follows: Figure 11 As shown, the details are as follows:

[0050] Step 101: Construct a three-dimensional model of the first bidirectional flow turbine and a three-dimensional model of the second bidirectional flow turbine, and perform parametric modeling on the first bidirectional flow turbine three-dimensional model and the second bidirectional flow turbine three-dimensional model.

[0051] Specifically, the bidirectional flow turbine and the fairing were modeled separately using the 3D modeling software Solidworks, and the two parts were then combined. The resulting 3D model was parametrically modeled using Design Modeler, allowing the structural parameters to be automatically adjusted during CFD calculations, and creating rotational and stationary domains; for example... Figure 12 and Figure 13 As shown, both the rotating and stationary domains are cylindrical flow domains; the distance from the inlet of the stationary domain to the turbine is 10D, the distance from the outlet to the turbine is 20D, the diameter of the cylindrical flow domain is 10D, the diameter of the rotating domain is 1.1D, and the longitudinal dimension parallel to the turbine is 1.27D.

[0052] Step 102: Determine the optimization objective of the single-objective optimization problem, and design the corresponding structural parameters and parameter ranges for the first symmetrical guide shield 3 and the second symmetrical guide shield 4 in the parametrically modeled first bidirectional flow turbine 3D model and second bidirectional flow turbine 3D model respectively.

[0053] Specifically, a key indicator of a turbine's hydrodynamic performance is its power coefficient. The decision is made, and for drag turbines where resistance is the primary driving force, the drag coefficient... It is also one of the key parameter indicators; power factor As a key evaluation criterion for water turbines, the power coefficient directly determines whether the device's performance meets the standards. This invention defines the optimization problem as a single-objective optimization, with the optimization objective being the power coefficient of the water turbine. For the two types of fairings, the corresponding parameter variables and parameter ranges are designed respectively. The relevant variable parameters of the fairings are shown in Table 2:

[0054] Table 2. Parameters of the fairing

[0055]

[0056] Step 103: Latin hypercube sampling is used to extract sample points within the parameter range of the first symmetrical fairing 3 and the second symmetrical fairing 4, and the objective function value corresponding to the sample points is determined.

[0057] Specifically, Latin hypercube sampling (LHS) is a commonly used Monte Carlo sampling method that effectively improves the coverage of the sample space and reduces the estimation variance by performing stratified sampling on each input dimension; for a given... 3D input vector In each marginal distribution Above, its domain is divided into n sub-intervals with equal probability:

[0058] ;

[0059] Unlike regular Monte Carlo methods, each dimension is randomly selected independently. Each point is different; LHS guarantees that exactly one sample point is drawn in each interval of each dimension. Random pairing is performed within the interval. In the j-th dimension, a random permutation is first generated for the interval index {1,2,…,n}. For the i-th sample in the j-th dimension, its interval index is: Uniform random sampling points are formed by uniform sampling within the interval. And mapped to real space:

[0060] ;

[0061] The structural parameters of the fairing need to be set within a reasonable range. To avoid extreme values ​​that would increase computational costs; and finally generate the output sample set. Regarding the selection of sample size, Arshad Afzal et al. proposed an initial LHS sample size of 15×k for simulation testing of fluid dynamics and optimization problems, to balance global exploration and local optimization performance. The power coefficient was calculated by performing CFD calculations on each of the above sample points. The values ​​are used to construct the training set for the subsequent proxy model; therefore, the present invention selects 45 and 60 sampling points for the first symmetrical fairing 3 and the second symmetrical fairing 4, respectively.

[0062] Step 104: Construct a Kriging surrogate model based on the sample points and their corresponding power coefficients, and fit the sample points of the variables and the objective function values ​​to the approximate model using the Kriging surrogate model.

[0063] Specifically, this invention uses Kriging as a surrogate model to map design variables to turbine power coefficients. First, the initial Kriging surrogate model is trained using the sample points obtained in step 103 and their corresponding power coefficients. Based on the training results, the Kriging surrogate model is obtained. Then, the parameter ranges of the corresponding parameters involved in step 102 are input one by one into the trained Kriging surrogate model, thereby obtaining the power coefficients of all corresponding water turbines. Kriging features accurate interpolation, global fitting, and uncertainty estimation. Furthermore, thanks to Gaussian process priors and structured information from the covariance matrix, Kriging often achieves lower prediction errors than methods such as multinomial regression and radial basis functions, making it very suitable for expensive simulations or scenarios where sample acquisition costs are high.

[0064] In the Kriging agent model, let the response function in the design space be... The response function to be fitted is considered as a Gaussian process with a mean: In the formula, Given a set of known regression basis functions, The regression coefficients to be estimated are: It is a zero-mean stationary Gaussian process. The covariance function is: In the formula, For process variance, It is determined by parameters The control function; in parameter estimation, given n training samples ;

[0065] Through maximum likelihood estimation : ; ;and Generally, it is obtained by maximizing the logarithmic fit through numerical optimization; at the same time, the kriging model has good predictive ability and can improve the accuracy and efficiency of the genetic algorithm in solving for the optimal value; based on the prediction formula, new points are... The prediction is: ; ;

[0066] The corresponding uncertain prediction equation is: In the formula, R is the correlation matrix, corresponding to the training sample points. The relevant vector represents the new point. Correlation between each training point For a set of regression basis functions, The design matrix is ​​formed by stacking the regression vectors of all training points;

[0067] After obtaining sample data points through experimental design in Isight, the impact of different parameter changes on the objective function value can be analyzed based on the changes in the sample points and objective function value of the DOE experimental design. Experimental points constructed using the DOE method can be combined with the corresponding objective function response value change patterns to quantitatively analyze the degree and direction of influence of each design variable on the objective function value; for example... Figure 14 As shown, the influence factor analysis diagram illustrates the percentage contribution of each structural parameter to the energy capture efficiency of the first symmetrical fairing 3 and the second symmetrical fairing 4; for the first symmetrical fairing 3, the inlet diameter D A The most important and dominant factor for C P The influence of the throat diameter (d1) accounts for nearly 76%, followed by the throat diameter (d1), which accounts for approximately 20%, while the total length (L) A It exhibited the lowest sensitivity, indicating that, for a two-stage fairing, maximizing the inflow collection area and moderately narrowing the throat channel are crucial for improving C. P Crucial; as for the second symmetrical fairing 4, the inlet diameter D B It also plays a decisive role, with its influence exceeding 75%. It is worth noting that the horizontal segment length L... B2 It became the second most significant factor (contributing approximately 20%), indicating that its importance was significantly higher than that of the throat diameter d2 and the total length L. B1 By changing the corresponding parameter values, the energy harvesting efficiency of the water turbine can be improved.

[0068] By establishing a Kriging model and utilizing the spatial correlation of data, accurate prediction and uncertainty quantification for unknown regions can be achieved; as follows: Figure 15 As shown, the kriging model for The error between the predicted and actual values ​​is small, and the red sample points have a high degree of overlap with the fitted function model, indicating that the fitting algorithm of the model is quite suitable.

[0069] Step 105: The fitting results of the approximate model are optimized using a multi-island genetic algorithm. Based on the optimization results, the optimal structural parameters of the first symmetrical fairing 3 and the second symmetrical fairing 4 are obtained.

[0070] Specifically, after establishing an approximate model based on sample data of design variables and objective function, a global optimization algorithm is needed to optimize the objective function and find the optimal design variables. This is because in step 104, the parameter ranges of all the fairing structure parameters are input into the Kriging surrogate model, and the corresponding turbine power coefficient is output through the Kriging surrogate model. The overall computational load is too large, so optimization methods are needed. Multi-island genetic algorithm is a distributed optimization technique based on genetic algorithms (GA). It simulates population migration and exchange through the concept of immigration, allowing each island's population to continuously receive new genetic information, thus achieving better search performance. This process can be executed in both continuous and discrete spaces and can be coupled with domain-specific constraint handling, local search, or surrogate model techniques to accelerate convergence and improve solution accuracy. In the field of hydraulic turbines, genetic algorithms are widely used for automated optimization of flow channel shapes and blade parameters. Puji Widiyanto et al. conducted multi-objective optimization of axial flow turbine blade cascades and impeller profiles to improve efficiency and reduce hydraulic losses. In fairing design, Tianxin Wu et al. used a hybrid optimization of multinomial neural networks and GA to adjust five key geometric variables, thereby significantly reducing energy loss in the impeller's rear diffuser section. These works demonstrate that GA, with its global search capabilities and flexibility in collaboration with surrogate models, has become an important tool for CFD optimization.

[0071] Figure 16 This is a 3D scatter plot matrix of the genetic algorithm used to find the optimal parameter points of the first symmetrical fairing 3. The first symmetrical fairing 3 has three input parameters and one output parameter. Each row corresponds to one of the parameters, and they are compared with... The relationship is shown, with each optimal parameter point represented by a red square in the graph. Based on the information from the scatter plot, the optimal parameter values ​​can be derived as D. A =360.54mm, L A1 =1157.53mm, d1=170.1mm, at this time Reaching 36.1%; such as Figure 17 As shown, the second symmetrical fairing 4 has four input variables and one output variable, expressed in the same form as... Figure 16 Consistent, the optimal parameter points are represented by the red squares in the diagram. According to the 3D scatter plot matrix, the optimal parameters for each item of the second symmetrical fairing 4 correspond to D... B =449.74mm, L B1 =706mm, L B2 =206.36mm, d2=183.47mm, at this time It reached 37%.

[0072] Step 106: Add a flange structure to the optimal structural parameters of the fairing, and perform secondary optimization on the fairing structure with the added flange structure. Obtain the optimal fairing scheme based on the secondary optimized structure.

[0073] Specifically, although the global optimization described above has significantly improved the geometric performance of the first symmetrical guide shield 3 and the second symmetrical guide shield 4, there is still great potential for improvement by introducing a flange structure at the trailing edge of the guide shield; the logic of unidirectional optimization cannot be directly translated because, under bidirectional constraints, the gain structure at the inlet may become a resistance source at the outlet when the water flow is reversed; the addition of flanges to both types of guide shields... Depending on the angle The variation (0°~90°) shows significant differences; the energy trapping capability of the first symmetrical fairing 3 first increases and then decreases with the increase of the flange angle. When the flange angle is 60°, The maximum value of 38.5% was achieved; the flange angle has a more significant impact on the second symmetrical fairing 4, when the flange angle is 70°. The maximum value of 45.9% was obtained. In summary, the "second symmetrical fairing + 70° flange" was finally determined to be the optimal fairing scheme of the present invention.

[0074] In summary, this invention addresses the problems of turbulent flow, high energy loss, and weak low-speed tidal energy capture in traditional bidirectional horizontal-axis turbines. Optimization of the impeller and guide vane structure is necessary, but existing optimization methods do not fully utilize the blade characteristics. Therefore, this invention proposes a method to maximize... For a single objective, this study integrates the LHS, Kriging surrogate model, and IGA collaborative optimization framework to conduct a systematic research on the structural design, parameter optimization, and flow field control mechanism of an Archimedes spiral bidirectional flow impeller and its adaptive fairing. The main conclusions are as follows:

[0075] 1. To adapt to the marine environment characterized by two daily reversals of tidal currents, this invention, based on the Archimedes' spiral principle, smoothly extends the tail of the unidirectional flow blades with the same curvature to form a back-symmetrical blade structure. This enables efficient work to be performed in both forward and reverse flow. Numerical simulations verify that the unidirectional flow impeller has the maximum... The bidirectional impeller designed in this invention has a value of 0.13. 1. Up to 0.18; 2. Considering that the designed bidirectional flow impeller has high solidity characteristics, its flow field blocking effect is significant and sensitive to the uniformity of the incoming flow, traditional guide shields are difficult to meet the requirements of flow acceleration and suppression separation. Therefore, this invention proposes two types of symmetrical guide shield structures. The first symmetrical guide shield 3 is a two-section structure of "contraction section-expansion section", and the second symmetrical guide shield 4 is a three-section structure of "contraction section-parallel section-expansion section". Both types of guide shields are designed with geometric symmetry to ensure the consistency of flow field control under bidirectional flow conditions.

[0076] 3. Flow field analysis shows that the parallel section of the second symmetrical guide shield 4, through the effects of "increased flow, stabilized flow, and suppressed separation," allows the high-speed flow to continuously impact the turbine's working surface. The pressure difference and vortex concentration are significantly better than those of the first symmetrical guide shield 3, providing core support for subsequent performance improvements. 4. To further enhance the flow-converging effect of the guide shield, flanges were added to both types of optimized guide shields for secondary optimization. The flange angle was systematically analyzed. (0°~90°) The influence mechanism; the results show that the optimal flange angle of the first symmetrical fairing 3 is 60°, at which point... Increased to 38.5%; the optimal flange angle of the second symmetrical fairing 4 is 70°. Reaching 45.9%, and when the flange angle exceeds the optimal value, the second symmetrical fairing 4... The drop is much smaller than that of the first symmetrical fairing 3, demonstrating stronger resistance to angle overload;

[0077] Therefore, after the impeller-guide synergistic optimization and flange secondary optimization, the operating efficiency of the entire turbine system is improved by 27.2% compared with the traditional structure, which fully verifies the scientificity and effectiveness of the "Archimedes spiral impeller + three-section guide shroud" design scheme. At the same time, the performance data under bidirectional flow conditions is highly consistent with the conclusions of the previous study, ensuring the inheritance and reliability of the design. The high-solidity bidirectional flow impeller design, three-section guide shroud structure and LHS-Kriging-IGA optimization framework proposed in this invention provide a new technical path and theoretical support for improving the performance of bidirectional flow turbines under low flow velocity environments.

[0078] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. All equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this invention should still be covered by the claims of this invention.

Claims

1. An Archimedes bidirectional spiral water turbine, comprising a bidirectional flow turbine (2), characterized in that: The bidirectional flow turbine (2) includes a central hub (21) and three helical blades (22) mounted on the central hub (21) and evenly arranged around the circumference of the central hub (21). The three helical blades (22) are arranged in an Archimedean spiral shape around the outer surface of the central hub (21). The radial distance between the outer edge of the three helical blades (22) and the surface of the central hub (21) gradually increases and then gradually decreases along the axial length of the central hub (21). The outer surface of the turbine body (1) is fitted with a first symmetrical guide shield (3) or a second symmetrical guide shield (4).

2. The Archimedes bidirectional spiral turbine according to claim 1, characterized in that: The first symmetrical guide shield (3) includes a first contraction section (31) sleeved outside the bidirectional flow turbine (2) for accelerating the fluid flow rate and a first expansion section (32) installed on the first contraction section (31) for reducing the fluid flow rate. The connection between the first contraction section (31) and the first expansion section (32) is a throat (33). The throat (33) of the first symmetrical guide shield (3) and the midpoint of the central hub (21) are on the same radial plane.

3. An Archimedes bidirectional spiral turbine according to claim 2, characterized in that: The first contraction section (31) is provided with a first inlet (34) at the end away from the first expansion section (32), and the first expansion section (32) is provided with a first outlet (35) at the end away from the first contraction section (31).

4. An Archimedes bidirectional spiral turbine according to claim 1, characterized in that: The second symmetrical flow guide (4) includes a second contraction section (41) and a second expansion section (42) fitted outside the bidirectional flow turbine (2) and used to accelerate the fluid flow rate and reduce the fluid flow rate, respectively. A parallel section (43) is installed between the second contraction section (41) and the second expansion section (42). The center of the parallel section (43) is on the same radial plane as the midpoint of the central hub (21).

5. An Archimedes bidirectional spiral turbine according to claim 4, characterized in that: The second contraction section (41) is provided with a second inlet (44) at the end away from the second expansion section (42), and the second expansion section (42) is provided with a second outlet (45) at the end away from the second contraction section (41).

6. A method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine, characterized in that: The Archimedes bidirectional spiral turbine based on any one of claims 1-5 comprises the following steps: Construct a three-dimensional model of the first bidirectional flow turbine and a three-dimensional model of the second bidirectional flow turbine, and perform parametric modeling on the first bidirectional flow turbine three-dimensional model and the second bidirectional flow turbine three-dimensional model; The optimization objective of the single-objective optimization problem is determined, and the corresponding structural parameters and parameter ranges are designed for the first symmetrical guide shield (3) and the second symmetrical guide shield (4) in the first and second bidirectional flow turbine three-dimensional models after parameterized modeling. The optimization objective is the power coefficient of the turbine. Latin hypercube sampling is used to extract sample points in the parameter ranges of the first symmetrical fairing (3) and the second symmetrical fairing (4), and the objective function value corresponding to the sample points is determined. A Kriging surrogate model is constructed based on the sample points and their corresponding power coefficients. The sample points of the variables and the objective function values ​​are then fitted to the approximate model using the Kriging surrogate model. The fitting results of the approximate model are optimized using a multi-island genetic algorithm, and the optimal structural parameters of the first symmetrical fairing (3) and the second symmetrical fairing (4) are obtained based on the optimization results.

7. The method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine according to claim 6, characterized in that: The construction of the first and second bidirectional flow turbine three-dimensional models includes: The bidirectional flow turbine (2), the first symmetrical guide shield (3) and the second symmetrical guide shield (4) were modeled using 3D modeling software. Based on the modeling results, the 3D models of the bidirectional flow turbine (2), the first symmetrical guide shield (3) and the second symmetrical guide shield (4) were obtained. The three-dimensional model of the bidirectional flow turbine (2) is combined with the three-dimensional models of the first symmetrical guide shield (3) and the second symmetrical guide shield (4) respectively, and the three-dimensional models of the first bidirectional flow turbine and the second bidirectional flow turbine are obtained according to the combination results.

8. The method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine according to claim 6, characterized in that: In the process of using Latin hypercube sampling to extract sample points in the parameter ranges of the first symmetrical guide vane (3) and the second symmetrical guide vane (4) and determining the objective function value corresponding to the sample points, the sample points are also design variables. The Kriging surrogate model is used to map the design variables to the turbine power coefficient.

9. The method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine according to claim 8, characterized in that: The step of constructing a Kriging surrogate model based on the sample points and their corresponding power coefficients, and then fitting an approximate model to the sample points of the variables and the objective function values ​​using the Kriging surrogate model, includes: The initial Kriging surrogate model is trained using the sample points and their corresponding power coefficients. Based on the training results, a Kriging surrogate model is obtained. The sample points of the variables and the objective function are fitted to an approximate model using the Kriging surrogate model, and a sensitivity analysis is performed on the relationship between the structural parameters.

10. The method for optimizing the guide fairing of an Archimedes bidirectional spiral turbine according to claim 6, characterized in that: After optimizing the fitting results of the approximate model using a multi-island genetic algorithm and obtaining the optimal structural parameters of the first symmetrical fairing (3) and the second symmetrical fairing (4) based on the optimization results, the method further includes: A flange structure is added to the optimal structural parameters of the fairing, and the fairing structure with the added flange structure is further optimized. The optimal fairing scheme is obtained based on the secondary optimized structure.

Images