Annular rotor wing and configuration parameter optimization method thereof

Abstract

The invention discloses an annular rotor wing and a configuration parameter optimization method thereof, and belongs to the technical field of aircrafts. An annular rotor wing is composed of m annular propeller discs and a propeller hub. A series of streamline two-dimensional airfoil sections, namely control sections, are distributed in the space along the annular track. The upper end points of the control sections are sequentially connected to form an upper sweeping track, the lower end points of the control sections are sequentially connected to form a lower sweeping track, sweeping is conducted through the control sections along the two sweeping tracks to construct the propeller disc, the propeller disc extends out of the propeller hub and finally returns to the propeller hub to form an annular structure, and all the sections of the propeller disc are streamline sections. A closed annular structure is defined by the propeller disc and the propeller hub, so that a cavity is formed in the propeller disc. According to the method, multi-objective optimization is carried out on the dynamic performance, the economical efficiency and the noise reduction performance of the annular rotor, the annular rotor which is high in lift coefficient and small in torque coefficient and gives consideration to the dynamic performance and the economical efficiency is obtained through optimization, and the obtained rotor is lower in power consumption and better in comprehensive performance.

Description

Technical Field

[0001] The present invention relates to an annular rotor and a method for optimizing its configuration parameters, belonging to the technical field of aircraft. Background Art

[0002] The rotor system provides necessary lift, propulsion force and operating force for the whole device, and at the same time faces challenges such as vibration and noise. The reasonable design of the rotor structure has become the key to improving the comprehensive performance of the rotor device.

[0003] As a component directly interacting with the medium (such as air), the performance of the rotor directly determines the efficiency and effect of energy conversion. The rotor can efficiently capture the kinetic energy of the medium and convert it into the force to push the device forward or upward. At the same time, by adjusting the angle and speed of the rotor, precise transformation of the position and attitude of the device can also be achieved. In this process, the design parameters of the rotor (such as shape, material, layout, etc.) are crucial for the rotor device to improve energy efficiency, reduce resistance, enhance stability and thus improve the comprehensive performance. The rotor structure design process needs to consider multiple factors comprehensively. First of all, it is necessary to ensure that the rotor has sufficient strength and stiffness to withstand various loads generated during flight, while reducing weight to improve flight efficiency. Secondly, it is necessary to optimize the aerodynamic performance of the rotor. Through calculation and experiment, the best rotor shape and layout scheme are found to maximize lift, propulsion force and operating force. In addition, the control problems of vibration and noise also need to be considered. Through reasonable structure design and material selection, the vibration level is reduced, the noise pollution is reduced, and the comfort and environmental protection of the device are improved.

[0004] There are two major methods for rotor noise reduction: active and passive. Among them, the active method suppresses noise by controlling the movement of the rotor. This method requires adding a relatively large volume of motion control device, so it is generally only applicable to the rotors of large helicopters and not applicable to small rotors. Passive noise reduction mainly relies on the traditional design or special structure design of the rotor to reduce rotor noise. This method has a certain effect and has been studied more in the field of helicopter rotor noise control. The traditional noise reduction method is to change the length or blade profile of the helical rotor, reduce its tip speed or change its interaction process with the air to reduce noise. The biggest drawback of this method is that it may affect the output power and thrust of the UAV power system, and its noise reduction effect is relatively limited.

[0005] In terms of improving the rotor structure, the adoption of bionic structures is a popular idea. For example, there are a large number of protruding feathers on the lower part of an owl's wings and a serrated trailing edge at the rear. Currently, the optimized structures for rotors include leading-edge serrated structures, leading-edge wavy structures, trailing-edge serrated structures, and combined structures of blade tip tapering and forward / rearward sweep. The actual effects of these methods of modifying and optimizing local structures in suppressing blade tip vortices and thus reducing aerodynamic noise are limited, and they will cause a decline in the aerodynamic performance of the rotor. Summary of the Invention

[0006] In view of the above technical problems existing in the design and comprehensive performance optimization of rotors in the prior art, the purpose of the present invention is to provide an annular rotor and a method for optimizing its configuration parameters. The technical problems to be solved are: (1) effectively improving the noise reduction performance of the rotor while ensuring the aerodynamic performance; (2) multi-objective optimization of the dynamic performance, economy, and noise reduction performance of the annular rotor, optimizing an annular rotor that takes into account both dynamic performance and economy with a higher lift coefficient and a smaller torque coefficient, so that the obtained rotor has lower power consumption and better comprehensive performance.

[0007] The purpose of the present invention is achieved through the following technical solutions.

[0008] An annular rotor disclosed by the present invention is composed of m annular blade disks and a hub. A series of streamlined two-dimensional airfoil sections, namely control sections, are distributed in space along an annular trajectory. The upper endpoints of the control sections are sequentially connected to form an upper sweep trajectory, and the lower endpoints of the control sections are sequentially connected to form a lower sweep trajectory. The blade disks are constructed by sweeping through the control sections along the two sweep trajectories. The blade disks extend from the hub and finally return to the hub, forming an annular structure. The cross-sections at various parts of the blade disks are all streamlined cross-sections. Denote the rotor radius as R, and the cross-section intercepted by the blade disk at the position of 1 / 2R is the dividing surface. Taking the dividing surface as the boundary, the part close to the hub is the blade root part, and the part far from the hub is the blade tip part. The blade disks and the hub enclose a closed annular configuration, so that a cavity is formed inside the blade disks.

[0009] A working method of a ring-shaped rotor: When the ring-shaped rotor is working, it has a relative motion with the air, generating eddy currents that flow into the cavity of the ring-shaped rotor blade disc from the upper part of the blade root. The closed configuration formed by the ring-shaped geometry and the hub confines the flow of the eddy currents within the cavity, causing the eddy currents to flow along the inner surface of the blade disc to form the eddy currents inside the ring-shaped rotor blade disc. The blade tip part is formed by sweeping along a ring-shaped trajectory, and its smooth configuration alleviates the interaction with the air, suppressing the tip vortices, thereby reducing the rotor aerodynamic noise generated by the vortices. Compared with a straight rotor, the ring-shaped rotor has a larger area of interaction with the medium under the same diameter. The streamlined airfoil and the smooth and fairing transition configuration of the ring shape alleviate the impact of the medium, making the pressure distribution on the rotor surface more uniform, the overall pressure pulsation more stable, the peak value of the surface velocity pulsation lower, the concentrated area of the velocity pulsation smaller, the energy loss during the rotor operation lower, the entropy production decreased, and the energy lost in the form of noise reduced, thereby reducing the aerodynamic noise and having better noise reduction performance.

[0010] A method for optimizing the configuration parameters of a ring-shaped rotor disclosed by the present invention includes the following steps:

[0011] Step 1: Use curve fitting to construct the mean camber line, construct a thickness distribution function in which the airfoil thickness changes along the mean camber line, and use the mean camber line to superimpose the thickness to complete the parametric expression of the two-dimensional streamlined airfoil section. After completing the expression of the mean camber line and the thickness distribution, a unit airfoil section with a chord length of 1 can be generated, and the unit airfoil section can be scaled into the required two-dimensional airfoil section by changing the section height and inclination angle.

[0012] The ring-shaped rotor airfoil section is represented by superimposing the thickness on the mean camber line. The mean camber line is the connection line of the incenter of the airfoil curve; the camber is the maximum vertical distance from the mean camber line to the airfoil chord line; the maximum thickness is the maximum distance between the upper and lower curves of the two-dimensional airfoil section in the normal direction of the mean camber line.

[0013] The optimization parameters include the coordinates (x p , y f ) of the camber position point of the mean camber line, the leading edge angle α i , the trailing edge angle α o , the normalized maximum thickness t, the twist angle θ, and the chord length l. The mean camber line is the connection line of the incenter of the airfoil curve; the camber is the maximum vertical distance from the mean camber line to the airfoil chord line; the maximum thickness is the maximum distance between the upper and lower curves of the two-dimensional airfoil section in the normal direction of the mean camber line.

[0014] The camber position point of the mean camber line is (x p , y f ), which divides the mean camber line into two sections at the head and the tail. The two sections intersect at the camber position point, and the tangent slope at the intersection point is 0. Curve fitting is used to express the two sections of the mean camber line respectively.

[0015] After the parametric expression of the mean camber line is completed, the thickness superimposed on the mean camber line is described using the class function / shape function method. The CST method expression for a unit airfoil section with a chord length of 1 is as follows:

[0016]

[0017] ζ is the absolute value of the airfoil coordinate in the y direction, which is 1 / 2 of the thickness, that is, the half-thickness;

[0018] ζ T is the half-thickness of the airfoil trailing edge. By setting ζ T = 0, the trailing edge is constructed to form a completely closed section;

[0019] is the class function. The values of N1 and N2 will change the overall shape of the airfoil section. When N1 = 0.5 and N2 = 1, the resulting airfoil is a common airfoil with a round leading edge and a pointed trailing edge.

[0020] The shape function selects the Bernstein polynomial. The expression of the nth-order Bernstein polynomial is:

[0021]

[0022] In the formula, Ai is the shape function coefficient. The airfoil can be regarded as composed of the superposition of each term of the shape function. Therefore, a specific two-dimensional airfoil can be expressed by controlling this coefficient, but the shape function coefficient does not have a clear geometric meaning.

[0023] The thickness distribution function y t = ζ(x)·t, where t is the maximum thickness of the airfoil. In the unit airfoil, it is the normalized maximum thickness. The coordinates of the upper and lower surfaces of the unit airfoil are obtained from the following formula:

[0024]

[0025]

[0026] where x, y are the coordinates of the mean camber line, x u , y u , x l , y l are the coordinates of the upper and lower curves of the airfoil section, and θg = arctan(dy / dx). After the mean camber line expression and the thickness distribution expression are completed, a unit airfoil section with a chord length of 1 can be generated. By changing the section height and inclination angle, the unit airfoil section is scaled into the required two-dimensional airfoil section.

[0027] Step 2: Based on the parametric expression method of the two-dimensional streamlined airfoil section in Step 1, a series of two-dimensional airfoil sections are constructed along a circular trajectory in space as control sections, and the stretching and twisting of each control section are completed by changing the section parameters.

[0028] An airfoil control section is introduced to control the annular rotor profile.

[0029] Denote the control section as S i , i = 0, 1, 2, …, n. And let S 0 be the mother section, and the remaining S j , j = 1, 2, 3, …, n be the child sections.

[0030] All airfoil sections are constructed by using the two-dimensional airfoil section parameterization method described in Step 1.

[0031] The mother section is the source of the airfoil section of the entire annular rotor. It is a unit streamlined airfoil section located at the origin, with a chord length constantly being 1. The other controllable parameters are: the coordinates of the camber position point on the mean camber line (x p , y f ), the leading edge angle α i , the trailing edge angle α o , the normalized maximum thickness t, and the twist angle θ.

[0032] The child sections change the chord length l, the twist angle θ, and the normalized maximum thickness t, stretch and twist the mother section, and finally obtain a new section. In the child sections, the S 1 section of the annular rotor is the first section, and the S n section is the tail section, which are the airfoil sections at both ends of the annular rotor. The remaining sections are the control sections located between the head and tail sections, used to control the changes of the annular rotor at the blade root and blade tip. The stretching and twisting of each control section are completed by changing the section parameters.

[0033] Step 3: Based on the control sections constructed in Step 2, connect the upper endpoints of the control sections in sequence to form an upper sweep trajectory, and connect the lower endpoints of the control sections in sequence to form a lower sweep trajectory.

[0034] Use the upper and lower two trajectories to constrain the sweep process of the annular rotor, where the upper trajectory is denoted as L 1 , and the lower trajectory is denoted as L 2 .

[0035] The upper trajectory passes through the upper connection points of n control sections. Denote the n passing points as L 1i , with coordinates (x 1i , y 1i , z 1i ), i = 1, 2, 3, …, n. Similarly, the lower trajectory will also pass through the connection points at the lower ends of n control sections. Denote the passing points as L 2i , with coordinates (x 2i , y 2i , z 2i), i = 1, 2, 3, …, n. Since the upper and lower trajectories pass through the upper and lower endpoints of all control sections respectively, the upper and lower trajectories and all control sections have formed a closed structure. Based on the passing point coordinates, parametric expressions of the upper and lower trajectory curves are constructed by curve fitting.

[0036] Step 4: Based on the control sections constructed in Step 2 and the swept trajectories constructed in Step 3, sweep through the control sections along the two swept trajectories to construct a blade disk. The blade disk extends from the hub and finally returns to the hub, forming an annular closed structure, and a three-dimensional model of the annular rotor is obtained.

[0037] Since the upper and lower swept trajectories of the annular rotor pass through the upper and lower endpoints of all control sections respectively, the swept trajectories and the control sections have formed a closed structure at this time. The construction of the three-dimensional model of the rotor is completed by sweeping each control section along the swept trajectory.

[0038] Form a parameter file with the parametric expressions for constructing the two-dimensional airfoil section and the swept trajectory of the annular rotor, import it into the three-dimensional modeling software UG, set all the parameters affecting the shape of the annular rotor, use the sweep function, take each control section as the sweep section, select the swept trajectory, and perform the sweep to obtain the three-dimensional model of the annular rotor.

[0039] Step 5: Based on the three-dimensional parametric model of the annular rotor constructed in Steps 1 to 4, construct an optimization parameter matrix by Latin hypercube sampling, perform multi-objective optimization on the annular rotor, and obtain an annular rotor that takes into account high dynamic performance, low power consumption, and high noise reduction performance.

[0040] Use the optimal Latin hypercube method to construct an optimization parameter matrix. The optimal Latin hypercube method is a random sampling method optimized based on the Latin hypercube sampling (LHS). Latin hypercube sampling is a multi-dimensional stratified sampling method that avoids repeated sampling of sample points through a layering technique and ensures the calculation accuracy with fewer samples. However, since the Latin hypercube design (LHD) arranges factors randomly for each level, it cannot guarantee the uniformity of the design samples. The optimal Latin hypercube realizes the balance of the distribution of sampling points in the design space by optimizing the uniformity and orthogonality of the Latin square matrix, thereby further improving the calculation accuracy of the selection of sample points.

[0041] To evaluate the comprehensive aerodynamic performance of each rotor, select the lift coefficient C related to the dynamic performance TTaking the quality factor FM related to the energy efficiency and economy of the rotor as the optimization goal. The quality factor is the ratio of the rotor induced power to the actual power in the hover state of the rotor-ducted system, which is used to characterize the hover efficiency of the rotor-ducted system. For an open rotor, the hover efficiency is calculated from equations (5) to (8):

[0042]

[0043]

[0044]

[0045]

[0046] where R is the propeller radius, Ω is the rotational speed (rad / s), ρ is the air density under normal temperature and pressure, A is the disc area of the propeller rotation plane, T is the lift obtained during the CFD simulation calculation, Q is the torque obtained during the CFD simulation calculation, C T is the lift coefficient, C Q is the torque coefficient, which is the result of non-dimensionalizing the lift and torque.

[0047] Set the value ranges of each parameter, design n ring-shaped rotor sample points with different parameters, analyze the aerodynamic performance of the ring-shaped rotor through the flow field aerodynamic performance simulation method, extract the lift and torque values of the ring-shaped rotor under specific rotational speed conditions, calculate the lift coefficient and the quality factor, and take the lift coefficient and the quality factor as the optimization goals to select the ring-shaped rotor with excellent dynamic performance and energy efficiency. Further perform transient simulation and acoustic simulation on the optimized ring-shaped rotor to verify and select the ring-shaped rotor that meets the noise reduction performance, that is, optimize to obtain a ring-shaped rotor that takes into account high dynamic performance, low power consumption, and high noise reduction performance.

[0048] In step one, to comprehensively consider the calculation accuracy and speed, preferably, the third-order Bezier curve is used to represent the two middle arcs respectively. A third-order Bezier curve is represented by the coordinates of four control points. Therefore, the coordinates of all control points at the head and tail of the middle arc need to be represented, which are P Li (x Li , y Li ), P Ti (x Ti , y Ti ), where i = 0, 1, 2, 3.

[0049] The equations of the Bezier curves at the head and tail are as follows:

[0050]

[0051]

[0052] After the parametric expression of the mean camber line is completed, the CST method is used to describe the thickness superimposed on the mean camber line. The expression of the CST method for a unit airfoil section with a chord length of 1 is as follows:

[0053]

[0054] ζ is the absolute value of the airfoil coordinate in the y direction, which is exactly 1 / 2 of the thickness, that is, the half-thickness;

[0055] ζ T is the half-thickness of the airfoil trailing edge. If a section with a completely closed trailing edge is to be constructed, then ζ T can be set to 0;

[0056] is a class function. The values of N1 and N2 will change the overall shape of the airfoil section. When N1 = 0.5 and N2 = 1, the resulting airfoil is a common airfoil with a round leading edge and a pointed trailing edge, and this type of airfoil is selected.

[0057] The shape function is selected as the Bernstein polynomial. Using the fifth-order Bernstein polynomial to obtain a higher fitting accuracy, the expression is:

[0058]

[0059] In the formula, Ai is the shape function coefficient, i = 1, 2, 3, 4, 5. The airfoil is equivalently composed of the superposition of each term of the shape function. Therefore, a specific two-dimensional airfoil is described by controlling this coefficient.

[0060] is the thickness distribution function y t = ζ(x)·t, where t is the maximum thickness of the airfoil. In a unit airfoil with a chord length of 1, it is the normalized maximum thickness. The coordinates of the upper and lower surfaces of the unit airfoil are obtained from the following formula:

[0061]

[0062]

[0063] where x, y are the coordinates of the mean camber line, x u , y u , x l , y l are the coordinates of the upper and lower curves of the airfoil section, and θg = arctan(dy / dx). After the mean camber line description and the thickness distribution description are completed, a unit airfoil section with a chord length of 1 is generated, and the unit airfoil section is scaled into the required two-dimensional airfoil section by changing the section height and inclination angle.

[0064] In step three, preferably, both the upper and lower trajectories are fitted using non-uniform rational B-spline (NURBS) curves.

[0065] The common form of parameterizing curves and surfaces using NURBS is as follows: Using a set of shape points located on the curve or surface, the coordinates of the control points of the curve or surface are solved inversely, and then the obtained control point coordinates are used to complete the fitting and representation of the entire curve or surface.

[0066] The NURBS curve is obtained by performing a weighted operation on the control points of the B-spline curve. The formula for the B-spline curve is as follows:

[0067]

[0068] where P i , i ∈ [0, n] is denoted as the B-spline control point. N i,p (u) is expressed as the basis function of the B-spline curve, i represents the sequence number, and p represents the order of the basis function of the B-spline curve.

[0069] Define P i as the following 4×1 vector:

[0070] P i = [x i y i z i 1] T (16)

[0071] where (x i , y i , z i ) represents the three-dimensional coordinates of P i in the Cartesian coordinate system, and the fourth component represents the weight of the control point, which is always 1 in the B-spline curve.

[0072] Therefore, in the NURBS curve spline, a non-zero weight value ω i is defined for each control point. Since multiplying the coordinates of the B-spline curve control point P i by a non-zero value (weight value) does not change the three-dimensional coordinates of the point in the Cartesian coordinate system, the coordinates of P i are multiplied by the corresponding non-zero weight value ω i to obtain the new coordinate vector after weighting in NURBS

[0073]

[0074] Since after the weighting process still represents the same point in the original three-dimensional coordinate system as the unweighted P i , so substitute into the formula:

[0075]

[0076] The obtained This is the B-spline curve obtained in the weighted coordinate system. Divide it by the value of the fourth coordinate component, i.e., the weight value ω i , and the NURBS spline curve in the original coordinate system is obtained. After arrangement, the final NURBS curve formula applicable to the outer contour of the annular rotor is:

[0077]

[0078] To control the curve shape of NURBS at the head and tail, it is required that the weight values ω 0 >0, ω n >0, and the remaining ω j ≥0, j ∈ (0, n). To ensure that the denominator is not zero, it is required that p consecutive weight values cannot be zero simultaneously.

[0079] The basis function N i,p (u) of the NURBS spline curve is determined by the knot vector U. U is a vector of order p + n + 2, expressed as:

[0080] U = [u 0 ,..., u i ,..., u p+n+1 T (20)

[0081] According to the deBoor - Cox recurrence formula, the recurrence formula for the n - th degree B - spline basis function is obtained as:

[0082]

[0083] For an open - loop curve with non - coincident head and tail endpoints, let p + 1 nodes adjacent to the two end nodes have the same value, and the multiplicity r = p + 1:

[0084]

[0085] If a series of passing - point coordinates are given, the control - point coordinates of the NURBS spline curve can be obtained by back - solving, and then the curve can be fitted. According to this method, with the help of the NURBS curve, while parameterizing the passing - point coordinates, the parametric expression of the three - dimensional curve of the annular rotor's spatial trajectory can be completed.

[0086] Select the upper and lower two trajectories to constrain the sweeping process of the annular rotor. The upper trajectory is denoted as L1, and the lower trajectory is denoted as L2. The upper trajectory will pass through the upper connection points of five control sections. Denote the five passing points as L1i, with coordinates (X1i, Y1i, Z1i), i = 1, 2, 3, 4, 5. Similarly, the lower trajectory will also pass through the lower connection points of five control sections. Denote the five passing points as L2i, with coordinates (X2i, Y2i, Z2i), i = 1, 2, 3, 4, 5. ​

[0087] Both trajectories are fitted using NURBS spline curves (Non-Uniform Rational B-Splines). To facilitate more intuitive observation and modification of the shape of the annular rotor trajectory, the inverse solution of the NURBS curve is used here. In the CAD software, the parametric spline is directly parameterized for the L1 and L2 trajectories through the passing point coordinates (X1i, Y1i, Z1i) and (X2i, Y2i, Z2i).

[0088] Since the change amplitude of the lower trajectory L2 in the Z direction of the annular rotor configuration is small, in the actual parametric expression process, to reduce the number of parameters and simplify the calculation amount in the parametric expression process, the Z coordinate in the lower trajectory control point coordinates (X2i, Y2i, Z2i) is set to the same parameter, that is, each Z2i is equal and is 0, that is, Z2i = Z2 = 0. Thus, a two-dimensional spline curve is used to fit L2, while L1 is still fitted using a three-dimensional spline curve. Therefore, the parametric expression of the swept trajectory of the annular rotor requires (X1i, Y1i, Z1i), (X2i, Y2i), i = 1, 2, 3, 4, 5, and Z2, a total of 26 parameters. At the same time, it should be noted that the chord length l and the tilt angle θ in the airfoil section have the following relationship with the above point coordinates, where l i is the length of the L1iL2i line segment, and θ i is the acute angle between L1iL2i and the Z coordinate axis direction.

[0089]

[0090]

[0091] R a 、R b1 、R b2 The three parameters respectively correspond to the following passing point coordinate values of the lower trajectory.

[0092] Y23 = R a (25)

[0093] X22 = R b1 (26)

[0094] X24 = R b2 (27)

[0095] Combined with equations (23) - (27), all passing point coordinates are solved, and the upper and lower swept trajectories are drawn based on all the solved passing point coordinates.

[0096] Beneficial effects:

[0097] 1. A ring-shaped rotor and its configuration parameter optimization method disclosed by the present invention. Since the rotor disk of the ring-shaped rotor extends from the hub and extends along a ring-shaped trajectory and finally returns to the hub, compared with a straight two-blade propeller of the same diameter, it has a larger acting area with the air medium, a more compact structure, a larger pressure surface and suction surface area, a greater lift, a lift coefficient greater than that of a straight two-blade propeller of the same diameter, and better aerodynamic performance.

[0098] 2. A ring-shaped rotor and its configuration parameter optimization method disclosed by the present invention. When the ring-shaped rotor is working, it has a relative motion with the air, generating eddy currents that flow into the cavity of the ring-shaped rotor disk from the upper part of the blade root. The closed configuration formed by the ring-shaped geometry and the hub confines the flow of the eddy currents in the cavity, causing the eddy currents to flow along the inner surface of the rotor disk to form an eddy current inside the ring-shaped rotor disk. The blade tip part is formed by sweeping along a ring-shaped trajectory, and its smooth configuration eases the interaction with the air, suppressing the tip vortex, thereby reducing the rotor aerodynamic noise generated by the vortex. Entropy production is the energy loss caused by the blades during the operation of the rotor. The higher the entropy production, the greater the energy loss caused by the velocity pulsation during the blade movement. A part of this energy loss is dissipated in the form of noise during the operation of the rotor. Therefore, the greater the entropy production value, the greater the generated noise. The ring-shaped rotor has a compact configuration, a larger acting area with the medium under the condition of the same rotor diameter, and the streamlined airfoil and the smoothly transitioning configuration of the ring-shaped rotor ease the impact of the medium, making the pressure distribution on the rotor surface more uniform, the overall pressure pulsation more stable, the peak value of the surface velocity pulsation lower, and the concentrated area of the velocity pulsation smaller. The energy loss during the operation of the rotor is lower, the entropy production decreases, and the energy lost in the form of noise is reduced, thereby reducing the aerodynamic noise and having better noise reduction performance.

[0099] 3. A ring-shaped rotor and its configuration parameter optimization method disclosed by the present invention. Based on the parametric expression method of the swept ring-shaped rotor, by constructing a control section and a sweeping trajectory, a three-dimensional model of the ring-shaped rotor is constructed by sweeping along the trajectory, avoiding the problem of complex parameter relationships brought about by the complex geometric configuration of the ring-shaped rotor, simplifying the parametric expression process, and improving the efficiency of optimizing the configuration parameters of the ring-shaped rotor.

[0100] 4. A ring-shaped rotor and its configuration parameter optimization method disclosed by the present invention. Through the DOE multi-objective optimization platform, the optimization objectives are set as the lift coefficient related to power performance and the quality factor related to economy. For rotors of the same diameter, by changing their airfoil section parameters and sweeping trajectory parameters, a ring-shaped rotor that takes into account both power performance and economy with a higher lift coefficient and a smaller torque coefficient is optimized, making the obtained rotor have lower power consumption and better comprehensive performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 Schematic diagram of the ring-shaped rotor of the present invention;

[0102] Figure 2 Flow chart of the configuration parameter optimization method for a ring-shaped rotor disclosed by the present invention;

[0103] Figure 3 Two-dimensional airfoil section;

[0104] Figure 4 Airfoil thickness distribution;

[0105] Figure 5 Schematic diagram of the control section;

[0106] Figure 6 Schematic diagram of the upper trajectory;

[0107] Figure 7 Schematic diagram of the lower trajectory;

[0108] Figure 8 Influence of the shape function coefficients on the airfoil when using a fifth-order Bernstein polynomial;

[0109] Figure 9 Distribution of the positions of 5 airfoil control sections;

[0110] Figure 10 Schematic diagram of the upper swept trajectory of the ring-shaped rotor;

[0111] Figure 11 Schematic diagram of the lower swept trajectory of the ring-shaped rotor;

[0112] Figure 12 Schematic diagram of the lower swept trajectory of the ring-shaped rotor;

[0113] Wherein: 1 - hub, 2 - blade disk, 3 - array blade disk. Specific implementation mode

[0114] To better illustrate the purpose and advantages of the present invention, the following further describes the content of the invention with reference to the drawings and examples.

[0115] Example 1:

[0116] As Figure 1As shown in the figure, a ring-shaped rotor disclosed in this embodiment is composed of m ring-shaped blade disks 2 and a hub 1. A series of streamline two-dimensional airfoil sections, that is, control sections, are distributed in space along a ring-shaped trajectory. The upper endpoints of the control sections are sequentially connected to form an upper sweep trajectory, and the lower endpoints of the control sections are sequentially connected to form a lower sweep trajectory. The blade disk 2 is constructed by sweeping along the two sweep trajectories through the control sections. The blade disk 2 extends from the hub 1 and finally returns to the hub 1 to form a ring structure. The cross-sections of all parts of the blade disk 2 are streamline cross-sections. Denote the radius of the rotor as R, and the cross-section intercepted by the blade disk 2 at the position of 1 / 2R is the dividing surface. Taking the dividing surface as the boundary, the part close to the hub 1 is the blade root part, and the part far from the hub 1 is the blade tip part. The blade disk 2 and the hub 1 enclose a closed ring-shaped configuration, so that a cavity is formed inside the blade disk 2.

[0117] As Figure 2 shown, a method for optimizing the configuration parameters of a ring-shaped rotor disclosed in this embodiment is specifically implemented as follows:

[0118] Step 1: Use curve fitting to construct the mean camber line, construct a thickness distribution function in which the airfoil thickness changes along the mean camber line, and use the mean camber line to superimpose the thickness to complete the parametric expression of the two-dimensional streamline airfoil section.

[0119] The airfoil section of the ring-shaped rotor is represented by superimposing the mean camber line and the thickness as Figure 3 shown.

[0120] To comprehensively consider the calculation accuracy and speed, the two segments of the mean camber line are expressed separately. To completely express a third-order Bezier curve, the coordinates of four control points are required. Therefore, to express the mean camber line, the coordinates of all the control points of the first and last segments are required, which are P Li (x Li , y Li ), P Ti (x Ti , y Ti ), where i = 0, 1, 2, 3.

[0121] The equations of the first and last segments of the Bezier curve are as follows:

[0122]

[0123]

[0124] After completing the parametric expression of the mean camber line, the CST method needs to be used to describe the thickness superimposed on the mean camber line. The CST method expression of a unit airfoil section with a chord length of 1 is as follows:

[0125]

[0126] ζ is the absolute value of the airfoil coordinate in the y direction, which is exactly half of the thickness, i.e., the half-thickness;

[0127] ζ T is the half-thickness of the airfoil trailing edge. If a cross-section with a completely closed trailing edge is to be constructed, ζ T can be set to 0;

[0128] is a class function. The values of N1 and N2 will change the overall shape of the airfoil cross-section. When N1 = 0.5 and N2 = 1, the resulting airfoil is a common airfoil with a round head and a pointed tail, and this type of airfoil is selected.

[0129] The shape function is selected as the Bernstein polynomial. A fifth-order Bernstein polynomial can be used to obtain a high fitting accuracy. As Figure 8 shown, the expression is:

[0130]

[0131] In the formula, Ai is the shape function coefficient, and i = 1, 2, 3, 4, 5. The airfoil can be regarded as composed of the superposition of each term of the shape function. Therefore, a specific two-dimensional airfoil can be expressed by controlling this coefficient.

[0132] is the thickness distribution function y t = ζ(x)·t, where t is the maximum thickness of the airfoil. In a unit airfoil with a chord length of 1, it is the normalized maximum thickness. The coordinates of the upper and lower surfaces of the unit airfoil are obtained from the following formula:

[0133]

[0134]

[0135] where x, y are the coordinates of the mean camber line, x u , y u , x l , y l are the coordinates of the upper and lower curves of the airfoil cross-section, and θg = arctan(dy / dx). After completing the description of the mean camber line and the thickness distribution, a unit airfoil cross-section with a chord length of 1 is generated. By changing parameters such as the cross-section height and inclination angle, the unit airfoil cross-section can be scaled into the required two-dimensional airfoil cross-section.

[0136] Step 2: Based on the parametric expression method of the two-dimensional streamlined airfoil cross-section in Step 1, a series of two-dimensional airfoil cross-sections are constructed along a circular trajectory in space as control cross-sections, and the stretching and twisting of each control cross-section are completed by changing the cross-section parameters.

[0137] To ensure the smoothness of the surface of the annular rotor three-dimensional model and ensure that its cross-section is always an airfoil cross-section to provide better aerodynamic performance, considering the model accuracy and calculation speed comprehensively, the number of control cross-sections is set to 5, and the control cross-sections are denoted as Si, where i = 0, 1, 2, 3, 4, 5. Among them, S0 is called the mother cross-section, and the remaining Si, where i = 1, 2, 3, 4, 5 are sub-cross-sections, as Figure 9 shown. All airfoil cross-sections are constructed by using the parameterized expression method of the streamlined airfoil cross-section described above.

[0138] Among the sub-cross-sections, the S1 cross-section of the annular rotor is the first cross-section, and the S5 cross-section is the last cross-section, which are the airfoil cross-sections at both ends of the annular rotor. The S2 and S4 cross-sections are the cross-sections of the annular rotor at a position with a radial distance of 1 / 2 radius. They control the lateral width of the annular rotor at the same time. In some annular rotor configurations, its sweep trajectory has a certain symmetry. Therefore, in the parameterized expressions of these annular rotors, the parameters of the S2 and S4 cross-sections are exactly the same. The S3 cross-section is the cross-section of the annular rotor at the blade tip. It is used to control the shape of the annular rotor at the position with the most severe twist. At the same time, the configuration of the blade tip will also have a greater impact on the aerodynamic performance and noise performance of the entire blade.

[0139] Through the above five control cross-sections, the blade of the annular rotor can be divided into three parts. Among them, the blade between the S1 and S2 cross-sections and the blade between the S4 and S5 cross-sections are the two blade root parts of the annular rotor. These two parts have a certain symmetry, and the part from S2 through S3 to S4 is the blade tip part, which is the twisted part of the annular rotor and has a greater impact on the aerodynamic performance of the blade. At the same time, its structural strength is relatively low. By parametrically controlling the tip cross-section, the shape of this part can be adjusted, thereby optimizing the relevant performance.

[0140] Step 3: Based on the control cross-sections constructed in Step 2, connect the upper endpoints of the control cross-sections in sequence to form the upper sweep trajectory, and connect the lower endpoints of the control cross-sections in sequence to form the lower sweep trajectory.

[0141] Both the upper and lower trajectories are fitted with non-uniform rational B-spline (NURBS) curves.

[0142] The common form of using NURBS to parametrically express curves and surfaces is: using a set of control points located on the curves and surfaces, inversely solving the coordinates of the control points of the curves and surfaces, and then using the obtained control point coordinates to complete the fitting and expression of the entire curves and surfaces.

[0143] The NURBS curve is obtained by performing a weighted operation on the control points of the B-spline curve. The formula of the B-spline curve is as follows:

[0144]

[0145] Among them, Pi , where \(i\in[0,n]\) is denoted as the B-spline control point. \(N\) i,p \((u)\) is expressed as the basis function of the B-spline curve, \(i\) represents the sequence number, and \(p\) represents the order of the B-spline curve basis function.

[0146] Let \(P\) i be defined as the following 4×1 vector:

[0147] \(P\) i \(=\left[x\right.\) i \(y\) i \(z\) i \(1\right]\) T (35)

[0148] where \((x\) i , \(y\) i , \(z\) i ) represents the three-dimensional coordinates of \(P\) i in the Cartesian coordinate system, and the fourth component represents the weight of this control point, which is always 1 in the B-spline curve.

[0149] Therefore, in the NURBS curve spline, a non-zero weight value \(\omega\) is defined for each control point i . Since multiplying the coordinates of the B-spline curve control point \(P\) i by a non-zero value (weight value) does not change the three-dimensional coordinates of this point in the Cartesian coordinate system, the coordinates of \(P\) i are multiplied by the corresponding non-zero weight value \(\omega\) i to obtain the new coordinate vector after weighting in NURBS

[0150]

[0151] Since after the weighting process and the unweighted \(P\) i still represent the same point in the original three-dimensional coordinate system, so is substituted into the formula:

[0152]

[0153] The obtained is the B-spline curve obtained in the weighted coordinate system. Dividing it by the value of the fourth coordinate component, that is, the weight value \(\omega\) i , the NURBS spline curve in the original coordinate system is obtained. After arrangement, the final NURBS curve formula applicable to the outer contour of the annular rotor is:

[0154]

[0155] To control the curve shape of the NURBS at the head and tail, it is required that the weight values \(\omega\) at the head and tail 0 \(>0,\omega\)n > 0, with the remainder ω j ≥ 0, j ∈ (0, n). To ensure that the denominator is not zero, it is required that the sequential p weight values cannot be zero simultaneously.

[0156] The basis function N of the NURBS spline curve i,p (u) is determined by the knot vector U, where U is a vector of order p + n + 2, expressed as:

[0157] U = [u 0 ,..., u i ,..., u p+n+1 T (39)

[0158] According to the deBoor - Cox recurrence formula, the recurrence formula for the n - th B - spline basis function is:

[0159]

[0160] For an open - loop curve with non - coincident start and end points, let the p + 1 nodes adjacent to the two end nodes have the same value, and the multiplicity r = p + 1:

[0161]

[0162] If a series of passing - point coordinates are given, the control - point coordinates of the NURBS spline curve can be obtained by inverse solution, and then the curve can be fitted. According to this method, with the help of the NURBS curve, while parameterizing the passing - point coordinates, the parametric expression of the three - dimensional curve of the annular rotor spatial trajectory can be completed.

[0163] Select the upper and lower two trajectories to constrain the sweeping process of the annular rotor. Among them, the upper trajectory is denoted as L1, and the lower trajectory is denoted as L2. The upper trajectory will pass through the upper connection points of five control sections. Denote the five passing points as L1i, with coordinates (X1i, Y1i, Z1i), i = 1, 2, 3, 4, 5 as Figure 10 shown. Similarly, the lower trajectory will also pass through the lower connection points of five control sections. Denote the five passing points as L2i, with coordinates (X2i, Y2i, Z2i), i = 1, 2, 3, 4, 5, as Figure 11 shown.

[0164] Both trajectories are fitted using NURBS spline curves (non - uniform rational splines). To facilitate more intuitive observation and change the shape of the annular rotor trajectory, here the inverse solution method of the NURBS curve described above is used to directly parameterize the L1 and L2 trajectories in the CAD software by the passing - point coordinates (X1i, Y1i, Z1i) and (X2i, Y2i, Z2i) of the parametric spline.

[0165] ​Since the variation amplitude of the lower trajectory L2 in the Z direction of the annular rotor configuration is relatively small, in the actual parametric expression process, to reduce the number of parameters and simplify the calculation amount in the parametric expression process, the Z coordinate in the lower trajectory control point coordinates (X2i, Y2i, Z2i) is set to the same parameter, that is, each Z2i is equal and is 0, namely Z2i = Z2 = 0. Thus, a two-dimensional spline curve is used to fit L2, while L1 still uses a three-dimensional spline curve for fitting. Therefore, the parametric expression of the swept trajectory of the annular rotor requires (X1i, Y1i, Z1i), (X2i, Y2i), i = 1, 2, 3, 4, 5, and Z2, a total of 26 parameters. At the same time, it should be noted that there is the following relationship between the parameter chord length l and the inclination angle θ in the airfoil section and the above point coordinates, where l i is the length of the line segment L1iL2i, and θ i is the acute angle between L1iL2i and the Z coordinate axis direction.

[0166]

[0167]

[0168] R a 、R b1 、R b2 The three parameters respectively correspond to the following lower trajectory passing point coordinate values.

[0169] Y23 = R a (44)

[0170] X22 = R b1 (45)

[0171] X24 = R b2 (46)

[0172] Combining the above formula, all passing point coordinates can be solved, and the upper and lower swept trajectories can be drawn.

[0173] Step 4: Based on the control section and the swept trajectory constructed in Step 2 and Step 3, sweep through the control section along the two swept trajectories to construct the paddle disk 2. The paddle disk 2 extends from the hub 1 and finally returns to the hub 1, forming an annular closed structure, and a three-dimensional model of the annular rotor is obtained.

[0174] After completing the parametric expression of the control section and the swept trajectory of the annular rotor, the three-dimensional model of the paddle disk 2 of the annular rotor can be obtained by sweeping along the swept trajectory. The hub 11 is constructed by stretching a cylinder. Taking the axis of the hub 11 as the rotation axis and the paddle disk 22 as the array feature, a symmetric array of paddle disks 32 is generated around the axis, and the hub 11, the paddle disk 22, and the array of paddle disks 323 are merged to obtain a complete three-dimensional model of the annular rotor.

[0175] Step 5: Based on the three-dimensional parametric model of the annular rotor constructed in Steps 1 to 4, an optimization parameter matrix is constructed through Latin hypercube sampling, and multi-objective optimization is performed on the annular rotor to obtain an annular rotor that takes into account high dynamic performance, low power consumption, and high noise reduction performance.

[0176] Taking the lift coefficient and quality factor as the optimization objectives, the optimization simulation calculation condition is 5000 r / min, the parameter variation range is 15%, and optimization simulation calculations are performed on 100 optimization parameter points generated by the optimal Latin hypercube method. Tables 1, 2, and 3 show the parameter values of 10 rotors selected by the optimization simulation calculation.

[0177] Table 1 Common parameters of the mother section;

[0178]

[0179] Table 2 Geometric parameters of the annular rotor;

[0180]

[0181] Table 3 Continued table of geometric parameters of the annular rotor;

[0182]

[0183] Figure 12 For the distribution diagrams of the lift coefficient and quality factor of 100 sample points in the optimization simulation calculation, the red coordinate points are the origin of the optimization parameters. It can be seen that the parametric design, modeling, and optimization method can effectively select an annular rotor configuration with better lift coefficient and quality factor, which can be applied to the field of multi-rotor aircraft, effectively improve the aerodynamic performance and energy efficiency of the rotor, and lay a foundation for the subsequent application of the annular rotor.

[0184] A working method of an annular rotor disclosed in this embodiment: When the annular rotor works, it has a relative motion with the air, generating eddy currents that flow into the cavity of the annular rotor blade disc 2 from the upper part of the blade root. The closed configuration formed by the annular geometric structure and the hub 1 confines the flow of the eddy currents in the cavity, causing the eddy currents to flow along the inner surface of the blade disc 2 to form an inner eddy current in the annular rotor blade disc 2. The blade tip part is swept along an annular trajectory, and its smooth configuration eases the interaction with the air, suppressing the tip vortex, thereby reducing the rotor aerodynamic noise generated by the vortex. Compared with a straight rotor, the annular rotor has a larger interaction area with the medium under the same diameter. The streamlined airfoil and the annular smooth and fairing transition configuration ease the impact of the medium, making the surface pressure distribution of the rotor more uniform, the overall pressure pulsation more stable, the peak value of the surface velocity pulsation lower, the area where the velocity pulsation is concentrated less, the energy loss lower during the working process of the rotor, the entropy production decreased, and the energy lost in the form of noise reduced, thereby reducing the aerodynamic noise and having better noise reduction performance.

[0185] The specific description above further elaborates on the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is only a specific embodiment of the present invention and is not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A ring-shaped rotor, characterized in that: It is composed of m annular impeller disks and a hub; a series of streamlined two-dimensional airfoil sections, namely control sections, are distributed in space along the annular trajectory; the upper end points of the control sections are connected in sequence to form an upper swept trajectory, the lower end points of the control sections are connected in sequence to form a lower swept trajectory, and the impeller disk is constructed by sweeping the control sections along the two swept trajectories; the impeller disk extends from the hub and finally returns to the hub to form an annular structure, and all sections of the impeller disk are streamlined sections; the rotor radius is R, the section cut by the impeller disk at the position of 1 / 2R is the interface, the part close to the hub is the root part, and the part away from the hub is the tip part, and the impeller disk and the hub surround each other to form a closed annular configuration, so that a cavity is formed inside the impeller disk.

2. The annular rotor according to claim 1, characterized in that: The annular rotor moves relative to the air when working, generating vortices that flow from the upper part of the blade root into the cavity of the annular rotor disc. The closed configuration formed by the annular geometric structure and the hub constrains the flow of the vortex within the cavity, causing the vortex to flow along the inner surface of the disc to form an internal vortex in the annular rotor disc; the blade tip part sweeps along an annular trajectory to form a smooth configuration. The smooth configuration mitigates the action of the air, suppresses the wingtip vortex, and thus reduces the rotor aerodynamic noise generated by the vortex; compared with straight rotors, the annular rotor has a larger area of ​​action with the medium under the condition of the same diameter. The streamlined airfoil and the annular smooth and smooth transition configuration mitigate the impact of the medium, making the rotor surface pressure distribution more uniform, the overall pressure pulsation more stable, the surface velocity pulsation peak lower, the velocity pulsation concentrated area less, the energy loss during the operation of the rotor lower, the entropy generation reduced, and the energy lost in the form of noise reduced, thereby reducing aerodynamic noise.

3. A method for optimizing the configuration parameters of an annular rotor, used to optimize an annular rotor as claimed in claim 1 or 2, characterized in that: The following steps are included: Step 1: Use curve fitting to construct the mid-arc line, construct the thickness distribution function of the airfoil thickness along the mid-arc line, and use the mid-arc line to superimpose the thickness to complete the parametric expression of the two-dimensional streamlined airfoil section; after completing the mid-arc line expression and thickness distribution expression, a unit airfoil section with a chord length of 1 can be generated, and the unit airfoil section can be scaled to the required two-dimensional airfoil section by changing the section height and inclination; Step 2: Based on the parametric expression method of the two-dimensional streamlined airfoil section in step 1, a series of two-dimensional airfoil sections are constructed along a circular trajectory in space as control sections, and the stretching and twisting of each control section are completed by changing the section parameters; Step 3: Based on the control section constructed in step 2, the upper end points of the control section are sequentially connected to form an upper sweep trajectory, and the lower end points of the control section are sequentially connected to form a lower sweep trajectory; Step 4: Based on the control section constructed in step 2 and the sweep trajectory constructed in step 3, the propeller disc is constructed by sweeping the control section along the two sweep trajectories. The propeller disc extends from the propeller hub and finally returns to the propeller hub to form an annular closed structure, thereby obtaining an annular rotor three-dimensional model. Step 5: Based on the three-dimensional parameterized model of the annular rotor constructed in steps 1 to 4, an optimization parameter matrix is ​​constructed through Latin hypercube sampling, and multi-objective optimization is performed on the annular rotor to obtain an annular rotor with high power performance, low power consumption, and high noise reduction performance.

4. The method for optimizing the configuration parameters of an annular rotor according to claim 3, characterized in that: The implementation method of step one is: The cross section of the annular rotor airfoil is represented by the superposition of the median arc line and thickness. The median arc line is the line connecting the centers of the inscribed circles of the airfoil curve. The camber is the maximum vertical distance from the median arc line to the airfoil chord line. The maximum thickness is the maximum distance between the upper and lower curves of the two-dimensional airfoil cross section in the normal direction of the median arc line. The optimization parameters include the coordinates of the mid-arc curvature position point (x p ,y f ), leading edge angle α i , trailing edge angle α o , normalized maximum thickness t, twist angle θ, chord length l; the mid-arc line is the line connecting the centers of the inscribed circles of the airfoil curve; the camber is the maximum vertical distance from the mid-arc line to the airfoil chord line; the maximum thickness is the maximum distance between the upper and lower curves of the two-dimensional airfoil section normal to the mid-arc line; The mid-arc curvature position point is (x p ,y f ), which divides the middle arc into two segments, the first and the last, and the two segments intersect at the curvature position point, and the tangent slope at the intersection point is 0; curve fitting is used to describe the two segments of the middle arc respectively; After completing the parametric expression of the median camber line, the class function / shape function method is used to describe the thickness superimposed on the median camber line; the CST method expression of the unit airfoil section with a chord length of 1 is as follows: ζ is the absolute value of the airfoil coordinate in the y direction, which is 1 / 2 of the thickness, i.e., half thickness; ζ T is the half thickness of the airfoil trailing edge, by setting ζ T =0 makes the rear edge of the construction form a completely closed section; is a class function, where the values ​​of N1 and N2 will change the overall shape of the airfoil section. When N1 = 0.5 and N2 = 1, the resulting airfoil is a common round-headed and pointed-tailed airfoil. The shape function uses Bernstein polynomials, and the expression of n-order Bernstein polynomials is: Where Ai is the shape function coefficient; the airfoil is equivalent to the superposition of the shape functions; Introducing thickness distribution function y t =ζ(x)·t, where t is the maximum thickness of the airfoil. In the unit airfoil, it is the normalized maximum thickness. The coordinates of the upper and lower surfaces of the unit airfoil are obtained by the following formula: Where x, y are the coordinates of the mid-arc, x u ,y u ,x l ,y l are the upper and lower curve coordinates of the airfoil section, θg=arctan(dy / dx); after completing the mid-arc line expression and thickness distribution expression, a unit airfoil section with a chord length of 1 can be generated, and the unit airfoil section can be scaled to the required two-dimensional airfoil section by changing the section height and inclination.

5. The method for optimizing the configuration parameters of an annular rotor according to claim 4, characterized in that: The implementation method of step 2 is: Introducing airfoil control sections to control the annular rotor shape; The control section is denoted as S i , i = 0, 1, 2, ..., n; and let S0 be the parent section, and the remaining S j ,j=1,2,3,…,n is the sub-section; All airfoil sections are constructed using the two-dimensional airfoil section parameterization method described in step 1; The parent section is the source of the airfoil section of the entire annular rotor. It is a unit streamlined airfoil section located at the origin, and its chord length is always 1. The other controllable parameters are: the coordinates of the mid-arc curvature position point (x p ,y f ), leading edge angle α i , trailing edge angle α o , normalized maximum thickness t, torsion angle θ; The sub-section changes the chord length l, torsion angle θ and normalized maximum thickness t, stretches and torsions the parent section, and finally obtains a new section; in the sub-section, the S1 section of the annular rotor is the first section, S n The cross section is the tail cross section, which is the airfoil cross section at the head and tail ends of the annular rotor; the remaining cross sections are control cross sections located between the head and tail cross sections, which are used to control the changes of the annular rotor at the blade root and blade tip; the stretching and torsion of each control cross section is completed by changing the cross-sectional parameters.

6. The method for optimizing the configuration parameters of an annular rotor according to claim 5, characterized in that: In step three, The sweeping process of the annular rotor is constrained by using two upper and lower trajectories, where the upper trajectory is recorded as L1 and the lower trajectory is recorded as L2; The upper trajectory passes through the upper connection points of n control sections, and the n passing points are recorded as L 1i , the coordinates are (x 1i ,y 1i ,z 1i ), i = 1, 2, 3, ..., n; similarly, the lower trajectory will also pass through the connection point at the lower end of the n control sections, and the passing point is recorded as L 2i , the coordinates are (x 2i ,y 2i ,z 2i ),i=1,2,3,…,n; since the upper and lower trajectories pass through the upper and lower endpoints of all control sections respectively, the upper and lower trajectories and all control sections have formed a closed structure; according to the coordinates of the passing points, the parameterized expressions of the upper and lower trajectory curves are constructed by curve fitting.

7. The method for optimizing the configuration parameters of an annular rotor according to claim 6, characterized in that: In step four, The parametric expressions of the two-dimensional airfoil section and the swept trajectory of the annular rotor are constructed to form a parameter file, which is imported into the three-dimensional modeling software UG. All parameters that affect the shape of the annular rotor are set. The sweep function is used, and each control section is used as the swept section. The sweep trajectory is selected and swept to obtain the three-dimensional model of the annular rotor.

8. The method for optimizing the configuration parameters of an annular rotor according to claim 7, characterized in that: The implementation method of step five is: The optimal Latin hypercube method is used to construct the optimization parameter matrix; the optimal Latin hypercube achieves the balanced distribution of sampling points in the design space by optimizing the uniformity and orthogonality of the Latin square matrix; In order to evaluate the comprehensive aerodynamic performance of each rotor, the lift coefficient C related to the power performance is selected T The quality factor FM related to the rotor energy efficiency and economy is the optimization target; the quality factor is the ratio of the rotor induced power to the actual power in the hovering state of the rotor-ducted system, which is used to characterize the hovering efficiency of the rotor-ducted system. For the open rotor, the hovering efficiency is calculated by equations (5) to (8): Where R is the propeller radius, Ω is the rotation speed (rad / s), ρ is the air density at normal temperature and pressure, A is the disk area of ​​the propeller rotation plane, T is the lift obtained during the CFD simulation calculation, Q is the torque obtained during the CFD simulation calculation, and C T is the lift coefficient, C Q is the torque coefficient, which is the result of dimensionless treatment of lift and torque; The value range of each parameter is set, and n sample points of annular rotors with different parameters are designed. The aerodynamic performance of the annular rotors is analyzed by flow field aerodynamic performance simulation method, and the lift and torque values ​​of the annular rotors under specific speed conditions are extracted. The lift coefficient and quality factor are calculated, and the lift coefficient and quality factor are used as optimization targets to select annular rotors with excellent power performance and energy efficiency. The optimized annular rotor is further subjected to transient simulation and acoustic simulation to verify and screen out annular rotors that meet the noise reduction performance, that is, annular rotors that have high power performance, low power consumption, and high noise reduction performance are optimized.

9. The method for optimizing the configuration parameters of an annular rotor according to claim 8, characterized in that: In step 1, a third-order Bezier curve is used to describe the two mid-arcs respectively. The coordinates of four control points are used to describe a third-order Bezier curve. Therefore, the coordinates of all the control points of the first and last segments are required to describe the mid-arc, which are P Li (x Li ,y Li ), P Ti (x Ti ,y Ti ), where i = 0, 1, 2, 3; The equations of the first and last Bezier curves are as follows: After completing the parametric expression of the median camber line, the CST method is used to describe the thickness superimposed on the median camber line. The CST method expression of the unit airfoil section with a chord length of 1 is as follows: ζ is the absolute value of the airfoil coordinate in the y direction, which is exactly 1 / 2 of the thickness, i.e., half thickness; ζ T is the half thickness of the airfoil trailing edge. If we want to construct a completely closed section at the trailing edge, we can let ζ T =0; is a class function, where the values ​​of N1 and N2 will change the overall shape of the airfoil section. When N1 = 0.5 and N2 = 1, the resulting airfoil is a round-headed and pointed-tailed airfoil. The shape function uses Bernstein polynomials. The fifth-order Bernstein polynomials are used to obtain a higher fitting accuracy expression: Where Ai is the shape function coefficient, i = 1, 2, 3, 4, 5; the airfoil is equivalent to the superposition of the shape functions, so by controlling this coefficient, a specific two-dimensional airfoil can be expressed; is the thickness distribution function y t =ζ(x)·t, where t is the maximum thickness of the airfoil. In a unit airfoil with a chord length of 1, it is the normalized maximum thickness. The coordinates of the upper and lower surfaces of the unit airfoil are obtained by the following formula: Where x and y are the coordinates of the mid-arc. u ,y u , x l ,y l are the upper and lower curve coordinates of the airfoil section, θg=arctan(dy / dx); after completing the mid-arc line expression and thickness distribution expression, a unit airfoil section with a chord length of 1 is generated, and the unit airfoil section is scaled to the required two-dimensional airfoil section by changing the section height and inclination.

10. The method for optimizing configuration parameters of an annular rotor according to claim 9, characterized in that: In step 3, both the upper and lower trajectories are fitted using non-uniform rational spline NURBS curves; Use a set of type value points on the curve surface to inversely solve the coordinates of the control points of the curve surface, and then use the obtained control point coordinates to complete the fitting and expression of the entire curve surface; The NURBS curve is obtained by weighting the control points of the B-spline curve. The formula of the B-spline curve is as follows: Among them, P i , i∈[0,n] is recorded as the B-spline control point; N i,p (u) represents the basis function of the B-spline curve, i represents the sequence number, and p represents the order of the basis function of the B-spline curve; P i Defined as the following 4×1 vector: P i =[x i y i z i 1] T (16) Where (x i ,y i ,z i ) represents P i The 3D coordinates in the Cartesian coordinate system, the 4th component represents the weight of the control point, and its value is always 1 in the B-spline curve; In a NURBS curve spline, a non-zero weight value ω is defined for each control point i , due to the control point P of the B-spline curve i Multiplying the coordinates by a non-zero value does not change the three-dimensional coordinates of the point in the Cartesian coordinate system, so for P i The coordinate is multiplied by the corresponding non-zero weight value ω i Get the weighted new coordinate vector P in NURBS i ω : P i ω =ω i P i (17) Since after weighted processing P i ω With unweighted P i Still represents the same point in the original three-dimensional coordinate system, so P i ω Bring it into the formula: Income This is the B-spline curve obtained in the weighted coordinate system, divided by the value of the fourth coordinate component, which is the weight value ω i , and the NURBS spline curve in the original coordinate system is obtained. After sorting, the final NURBS curve formula suitable for the outer contour of the annular rotor is: To control the curve shape of NURBS at the beginning and end, the weight value at the beginning and end is ω0>0,ω n >0, remaining ω j ≥0,j∈(0,n); To ensure that the denominator is not 0, the p weight values ​​in sequence cannot be 0 at the same time; Basis function N of NURBS spline curve i,p (u) is determined by the node vector U, which is a p+n+2 order vector and is expressed as: U=[u0,...,u i ,...,u p+n+1 ] T (20) According to the deBoor Cox recursion formula, the n-order B-spline basis function recursion formula is obtained as follows: For an open-loop curve whose end points do not coincide, let the p+1 nodes adjacent to the nodes at both ends have the same value, and the repetition degree r = p+1: If a series of passing point coordinates are given, the control point coordinates of the NURBS spline curve can be obtained by inverse solution, and then the curve can be fitted; according to this method, with the help of the NURBS curve, the passing point coordinates are parameterized and the parametric expression of the three-dimensional curve of the annular rotor space trajectory is completed; The upper and lower trajectories are selected to constrain the sweeping process of the annular rotor, where the upper trajectory is recorded as L1 and the lower trajectory is recorded as L2; ​​the upper trajectory will pass through the upper connection points of the five control sections, and the five passing points are recorded as L1i, with coordinates (X1i, Y1i, Z1i), i = 1, 2, 3, 4, 5; similarly, the lower trajectory will also pass through the lower connection points of the five control sections, and the five passing points are recorded as L2i, with coordinates (X2i, Y2i, Z2i), i = 1, 2, 3, 4, 5; Using the inverse solution of the NURBS curve, the L1 and L2 trajectories are parameterized directly in the CAD software through the passing point coordinates (X1i, Y1i, Z1i) and (X2i, Y2i, Z2i) of the parametric spline; The Z coordinate in the coordinates of the lower trajectory control point (X2i, Y2i, Z2i) is set to the same parameter, that is, each Z2i is equal and 0, that is, Z2i = Z2 = 0, so that L2 is fitted using a two-dimensional spline curve, while L1 is still fitted using a three-dimensional spline curve; therefore, the parameterized expression of the swept trajectory of the annular rotor requires 26 parameters (X1i, Y1i, Z1i), (X2i, Y2i), i = 1, 2, 3, 4, 5, and Z2; the parameters of the chord length l and the inclination angle θ in the airfoil section have the following relationship with the coordinates of the above points, where l i is the length of the L1iL2i line segment, θ i It is the acute angle between L1iL2i and the Z coordinate axis direction; R a , R b1 , R b2 The three parameters correspond to the coordinate values ​​of the points through which the trajectory passes as follows; Y23=R a (25) X22=R b1 (26) X24=R b2 (27) Combine equations (23) to (27) to solve the coordinates of all passing points, and draw the upper and lower sweeping trajectories based on the solved coordinates of all passing points.

Images