Pneumatic-strength comprehensive optimization design method for centripetal turbine and related device

Abstract

The invention belongs to the field of turbomachinery design, and discloses a centripetal turbine pneumatic-strength comprehensive optimization design method and related device.The centripetal turbine pneumatic-strength comprehensive optimization design method comprises the steps that based on an initial design vector, pneumatic parameters and geometric parameters of preset parts in a to-be-designed centripetal turbine are determined; obtaining the evaluation efficiency of the to-be-designed centripetal turbine based on a pre-constructed neural network model; judging whether the evaluation efficiency of the to-be-designed centripetal turbine is converged or not; if yes, whether the evaluation efficiency of the to-be-designed centripetal turbine is optimal or not is judged; if the initial design vector is not optimal, updating the initial design vector based on a genetic algorithm to obtain an optimized design vector; judging whether a preset part in the centripetal turbine to be designed meets the material strength requirement or not; if yes, outputting an optimized design vector as a design optimization result of the centripetal turbine to be designed; the defect that traditional design based on experience needs multiple times of correction is effectively avoided, the research and development period is greatly shortened, the design cost is reduced, and the optimal solution in the design space can be deeply excavated.

Description

Technical Field

[0001] This invention belongs to the field of turbomachinery design technology, and specifically relates to a comprehensive aerodynamic-strength optimization design method and related device for centripetal turbines. Background Technology

[0002] In the field of power cycle, centripetal turbines are turbomachinery used to convert the thermal energy of the working fluid into the mechanical energy of the machine. Due to their compact structure, high heat-to-work conversion efficiency, and high single-stage enthalpy drop, they are widely used in aerospace, thermal power generation, and waste heat recovery. Due to the complex internal flow and high impeller linear velocity of centripetal turbines, their aerodynamic and structural optimization has always been the goal of academia and industry in design, manufacturing, and operation. Therefore, a more efficient and rational structure has always been the goal pursued by centripetal turbine designers.

[0003] Currently, the design of centripetal turbines generally includes design methods based on load coefficient and flow coefficient, as well as design methods based on reaction degree and speed ratio. However, the design parameters in these methods are usually selected based on the designer's accumulated experience, and the design process is constrained by many factors. Moreover, even with existing experience, multiple design and revisions are still required to obtain a reliable design method, which greatly increases the R&D cycle and design cost of centripetal turbines. Furthermore, the design results obtained based on the above methods can only serve as feasible solutions within the design space, making it difficult to deeply explore the optimal solution. Summary of the Invention

[0004] In view of the technical problems existing in the prior art, the present invention provides a comprehensive aerodynamic-strength optimization design method and related device for centripetal turbines, so as to solve the technical problems that the existing centripetal turbine design methods rely on human experience and require multiple design and correction, which greatly increases the R&D cycle and design cost of centripetal turbines, and makes it difficult to find the optimal solution.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] This invention provides a comprehensive aerodynamic-strength optimization design method for centripetal turbines, including:

[0007] Set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed to obtain the initial design vector;

[0008] Based on the initial design vector, the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed are determined;

[0009] Based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed, the evaluation efficiency of the centripetal turbine to be designed is obtained based on a pre-built neural network model; wherein, the pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters.

[0010] Determine whether the evaluation efficiency of the centripetal turbine to be designed has converged; if it has converged, determine whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level; if it has not reached the optimal level, update the initial design vector based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and obtain the optimized design vector.

[0011] Based on the optimized design vector, it is determined whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if they do, the optimized design vector is output as the design optimization result of the centripetal turbine to be designed.

[0012] Furthermore, the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed include the aerodynamic parameters of the nozzle, the geometric parameters of the nozzle, the inlet aerodynamic parameters of the impeller, the outlet aerodynamic parameters of the impeller, and the geometric parameters of the impeller.

[0013] Furthermore, the process of constructing a pre-built neural network model includes:

[0014] Determine the design optimization conditions for the centripetal turbine;

[0015] Based on the design optimization conditions of centripetal turbines, experimental data and loss models of different models of centripetal turbines were obtained;

[0016] Based on experimental data and loss models of different types of centripetal turbines, a centripetal turbine design sample library containing the total efficiency under different design parameter conditions is established.

[0017] Based on a centripetal turbine design sample library containing the total efficiency under different design parameter conditions, a pre-determined artificial neural network model is trained to obtain an artificial neural network model that characterizes the relationship between efficiency and aerodynamic and structural parameters, thus obtaining a pre-constructed neural network model.

[0018] Furthermore, the loss models include nozzle loss model, impeller angle of attack loss model, impeller channel loss model, impeller clearance leakage loss model, impeller trailing edge loss model, impeller back friction blow-through loss model, and impeller residual velocity loss model.

[0019] Furthermore, the process of determining whether the evaluation efficiency of the centripetal turbine under design converges includes:

[0020] The evaluated efficiency of the centripetal turbine to be designed is compared with the initial efficiency of the centripetal turbine to be designed. If the relative deviation between the evaluated efficiency of the centripetal turbine to be designed and the initial efficiency of the centripetal turbine to be designed is less than or equal to the preset efficiency convergence threshold, then the evaluated efficiency of the centripetal turbine to be designed has converged; otherwise, the evaluated efficiency of the centripetal turbine to be designed has not converged.

[0021] The initial efficiency of the centripetal turbine to be designed is the total efficiency of the centripetal turbine to be designed.

[0022] Furthermore, based on the optimized design vectors, the process of determining whether the preset components in the centripetal turbine under design meet the material strength requirements includes:

[0023] Based on the optimized design vector, the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed are calculated.

[0024] Determine whether the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed meet the material strength requirements of the preset components.

[0025] This invention also provides a centripetal turbine aerodynamic-strength integrated optimization design system, comprising:

[0026] The design vector setting module is used to set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed, and to obtain the initial design vector.

[0027] The parameter calculation module is used to determine the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed based on the initial design vector.

[0028] The model evaluation module is used to obtain the evaluation efficiency of the centripetal turbine to be designed based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed and a pre-built neural network model. The pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters.

[0029] The design vector optimization module is used to determine whether the evaluation efficiency of the centripetal turbine to be designed has converged. If it has converged, it determines whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level. If it has not reached the optimal level, the initial design vector is updated based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and the optimized design vector is obtained.

[0030] The strength assessment module, based on the optimized design vector, determines whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if they do, it outputs the optimized design vector as the design optimization result of the centripetal turbine to be designed.

[0031] The present invention also provides an electronic device, comprising:

[0032] A processor is used to execute computer programs;

[0033] A computer-readable storage medium storing a computer program, which, when executed by the processor, performs the centripetal turbine aerodynamic-strength integrated optimization design method.

[0034] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned centripetal turbine aerodynamic-strength integrated optimization design method.

[0035] The present invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the centripetal turbine aerodynamic-strength integrated optimization design method.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] The aerodynamic-strength integrated optimization design method for centripetal turbines provided by this invention utilizes a neural network model to quickly evaluate efficiency and combines it with a genetic algorithm to efficiently search for optimal design parameters. This effectively avoids the drawbacks of traditional experience-based designs that require multiple corrections, significantly shortening the development cycle and reducing design costs. Furthermore, it can deeply explore the optimal solution within the design space, improving the design quality and performance of centripetal turbines. In addition, the optimization process comprehensively considers aerodynamic and strength factors, ensuring design reliability by verifying whether preset components meet material strength requirements. This guarantees that the centripetal turbine possesses both good aerodynamic performance and structural stability in practical applications, providing a significant improvement for centripetal turbine design. It provides a more efficient, reasonable, and reliable solution. Specifically, it updates the initial design vector based on a genetic algorithm, realizing an automated global search for the optimal design scheme under given constraints, thereby greatly reducing the reliance on human experience in the design parameter selection process. Secondly, it obtains the evaluation efficiency of the centripetal turbine to be designed by using a pre-built neural network model, getting rid of the dependence on loss models in traditional design methods and greatly shortening the one-dimensional design cycle of the centripetal turbine. In addition, it couples structural optimization design with centripetal turbine aerodynamic optimization, which can improve the aerodynamic performance of the centripetal turbine while ensuring the rationality of the structure.

[0038] The aerodynamic-strength integrated optimization design system, electronic device, computer-readable storage medium, and computer program product for centripetal turbines provided by this invention possess all the advantages of the aforementioned aerodynamic-strength integrated optimization design method for centripetal turbines. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A flowchart of the aerodynamic-strength integrated optimization design method for centripetal turbines provided in Example 1;

[0041] Figure 2 The meridian diagram of the centripetal turbine after optimization design in Example 1;

[0042] Figure 3 This is a three-dimensional model diagram of the impeller of the centripetal turbine after the optimized design in Example 1;

[0043] Figure 4 This is a schematic diagram of the pressure distribution on the blade surface of the example centripetal turbine after the optimized design in Example 1;

[0044] Figure 5 The graph shows the efficiency of the centripetal turbine after optimization design in Example 1 as a function of flow rate under various operating conditions.

[0045] Figure 6 The pressure ratio versus flow rate performance curve of the optimized radial turbine in Example 1 is shown.

[0046] Figure 7 This is a structural block diagram of the centripetal turbine aerodynamic-strength integrated optimization design system provided in Example 2;

[0047] Figure 8 This is a structural block diagram of the electronic device provided in Example 3. Detailed Implementation

[0048] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0049] This invention provides a comprehensive aerodynamic-strength optimization design method for centripetal turbines, comprising the following steps:

[0050] Step 100: Set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed to obtain the initial design vector.

[0051] Step 200: Based on the initial design vector, determine the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed.

[0052] Step 300: Based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed, obtain the evaluation efficiency of the centripetal turbine to be designed based on the pre-built neural network model; wherein, the pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters.

[0053] Step 400: Determine whether the evaluation efficiency of the centripetal turbine to be designed has converged; if it has converged, determine whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level; if it has not reached the optimal level, update the initial design vector based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and obtain the optimized design vector.

[0054] Step 500: Based on the optimized design vector, determine whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if they do, output the optimized design vector as the design optimization result of the centripetal turbine to be designed.

[0055] The centripetal turbine aerodynamic-strength integrated optimization design method described in this invention utilizes a neural network model to quickly evaluate efficiency and combines a genetic algorithm for global search optimization of the design vector. This significantly shortens the development cycle, reduces design costs, and deeply explores the optimal solution, thereby improving design quality. Simultaneously, the optimization process considers both aerodynamic performance and structural strength, and strength verification ensures the reliability and practicality of the design results. This guarantees the stability, safety, and performance of the centripetal turbine in practical applications, providing a more efficient, reasonable, and reliable approach for centripetal turbine design.

[0056] The following specific embodiments further explain the aerodynamic-strength integrated optimization design method for centripetal turbines provided by the present invention:

[0057] Example 1

[0058] As attached Figure 1 As shown, this embodiment 1 provides a comprehensive aerodynamic-strength optimization design method for centripetal turbines, including the following steps:

[0059] Step 1: Establish an artificial neural network model to characterize the relationship between efficiency and aerodynamic and structural parameters, and obtain a pre-constructed neural network model.

[0060] Specifically, the process is as follows:

[0061] Step 11: Determine the optimal design conditions for the centripetal turbine; the optimal design conditions for the centripetal turbine include flow rate m, inlet temperature T0, inlet pressure P0, and outlet pressure P6.

[0062] Step 12: Based on the optimized design conditions of the centripetal turbine, obtain experimental data and loss models for different types of centripetal turbines; among which, the loss models include nozzle loss model, impeller angle of attack loss model, impeller channel loss model, impeller clearance leakage loss model, impeller trailing edge loss model, impeller back friction blow-through loss model, and impeller residual velocity loss model.

[0063] The spray-following loss model is as follows:

[0064]

[0065] Where, Δh n α is the jet follow-through loss; C4 is the absolute velocity at the nozzle exit, m / s; Re is the Reynolds number; α4 is the absolute airflow angle at the nozzle exit, rad; s is the pitch, m; c is the chord length, m; b is the blade height, m.

[0066] The impeller angle of attack loss model is as follows:

[0067]

[0068] Where, Δh i W4 is the impeller angle of attack loss, used to characterize the energy consumed in correcting the fluid flow direction to the direction of the flow channel; W4 is the relative velocity at the impeller outlet, m / s; β4 is the relative airflow angle at the impeller outlet, rad; β 4,opt The optimized relative airflow angle at the blade outlet is expressed in rad.

[0069] The impeller channel loss model is as follows:

[0070]

[0071] Where, Δh p For impeller channel losses, it is used to characterize various flow losses that occur within the channel. In the case of a centripetal turbine, it mainly refers to lateral flow, secondary flow, and kinetic energy loss caused by boundary layer growth. r4 is the average radius of the impeller throat, in meters; r4 is the impeller inlet radius, in meters; β b5 b5 is the blade throat geometry angle, rad; b4 is the blade throat height, m; W5 is the impeller inlet relative velocity, m / s; W6 is the impeller throat inlet relative velocity, m / s; L H L is the hydraulic length, in meters (m). D Z is the hydraulic diameter, in meters. r b4 is the impeller axial width, in meters; b4 is the impeller inlet blade height, in meters; r h5 r is the hub radius at the blade throat, in meters. s5 denoted as the radius of the leaf throat and apex, in meters (m).

[0072] The impeller clearance leakage loss model is as follows:

[0073]

[0074] Where, Δh c The impeller clearance leakage loss takes into account the interaction between axial and radial clearance leakage; U4 is the impeller circumferential velocity, m / s; K a K r K a,r For empirical coefficients, when K a=0.4, K r =0.75, K a,r When ε = -0.3, the agreement with experimental data is best; a ε is the axial clearance. r For radial clearance; C a and C r r is the empirical coefficient; s6 C is the impeller outlet blade tip radius, in meters. m4 C is the impeller inlet meridional velocity, m / s; m6 r6 is the average velocity at the impeller exit, in m / s; r6 is the impeller exit radius, in m; b6 is the impeller exit blade height, in m.

[0075] The impeller trailing edge loss model is as follows:

[0076]

[0077] Where, Δh te For impeller trailing edge losses; R is the gas constant, J / (kg·K); T6 is the impeller outlet temperature, K; P t6,rel ΔP is the total pressure at the impeller outlet in the relative coordinate system, in Pa; t,te ρ6 is the pressure loss at the impeller trailing edge; ρ6 is the density of the working fluid at the impeller outlet, kg / m³. 3 W 6,rms The mean square value of the relative velocity at the impeller outlet, in m / s; N r t represents the number of impeller blades. 6,rms r is the mean square value of the impeller outlet blade thickness, in meters (m). h6 β is the impeller outlet hub radius, in meters; 6,rms denoted as the mean square value of the relative airflow angle at the impeller outlet, in rad.

[0078] It should be noted that because the impeller has thickness, when the airflow passes over the trailing edge of the blade, it will cause instantaneous expansion, which will lead to trailing edge loss.

[0079] The impeller back friction blower loss model is as follows:

[0080]

[0081] Where, Δh w The blower loss is due to friction between the impeller back and the airflow. The average density of the gas is kg / m³. 3 m is the mass flow rate, kg; W6 is the relative velocity at the impeller outlet, m / s; k f ε is the coefficient of friction; b The blade gap is in meters (m).

[0082] The impeller residual velocity loss model is as follows:

[0083]

[0084] Where, Δh e C1 is the impeller residual velocity loss; C6 is the impeller outlet absolute velocity, m / s.

[0085] Step 13: Based on the experimental data and loss models of different types of centripetal turbines, establish a centripetal turbine design sample library that includes the total efficiency under different design parameter operating conditions.

[0086] Step 14: Based on the centripetal turbine design sample library containing the total efficiency under different design parameter conditions, train the pre-determined artificial neural network model to obtain an artificial neural network model that characterizes the relationship between efficiency and aerodynamic and structural parameters, i.e., obtain the pre-constructed neural network model.

[0087] Specifically, based on a centripetal turbine design sample library containing the overall efficiency under different design parameter conditions, the number of hidden layer nodes of a pre-determined Artificial Neural Network (ANN) model is adjusted. By changing the number of hidden layer nodes, the average error and root mean square error of the pre-determined ANN model are calculated to obtain the optimal number of hidden layer nodes; an optimized ANN model is obtained; at least 40 samples are randomly selected to verify the accuracy of the optimized ANN model; if the verification accuracy does not meet the requirements, the process returns to continue adjusting the number of hidden layer nodes of the pre-determined Artificial Neural Network (ANN) model and retraining until the accuracy meets the preset verification accuracy, thus obtaining an ANN model used to characterize the relationship between efficiency and aerodynamic and structural parameters.

[0088] The expressions for the mean error and the root mean square deviation are as follows:

[0089]

[0090] Where, Δ k,m σ is the average error; k y is the mean squared error; * For predicted values; y 0 This is a sample library of centripetal turbine designs containing the overall efficiency under different design parameters; k represents the number of nodes in different hidden layers; m represents different output variables; i represents different training iterations; and j represents different samples.

[0091] Preferably, when training the predetermined artificial neural network model, the number of training samples is 50, and the number of nodes in each hidden layer is repeatedly trained 50 times; the average error is required to be less than 10. -4 .

[0092] Step 2: Set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed to obtain the initial design vector.

[0093] Step 3: Initialize the total efficiency of the centripetal turbine to be designed, and obtain the initial efficiency η0 of the centripetal turbine to be designed.

[0094] Step 4: Based on the initial design vector, determine the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed. These parameters include the aerodynamic and geometric parameters of the nozzle, the inlet aerodynamic parameters of the impeller, the outlet aerodynamic parameters of the impeller, and the geometric parameters of the impeller.

[0095] Preferably, the aerodynamic parameters of the nozzle include nozzle inlet temperature, nozzle inlet pressure, and absolute airflow angle at the nozzle outlet; the geometric parameters of the nozzle include nozzle inlet radius, nozzle outlet radius, nozzle blade height, and number of nozzle blades; the aerodynamic parameters of the impeller inlet include impeller inlet temperature, impeller inlet pressure, impeller inlet density, impeller inlet Mach number, impeller inlet absolute velocity, impeller inlet relative velocity, impeller inlet absolute airflow angle, impeller inlet relative airflow angle, and impeller inlet circumferential velocity; the aerodynamic parameters of the impeller outlet include impeller outlet temperature, impeller outlet pressure, impeller outlet density, impeller outlet relative Mach number, impeller outlet absolute velocity, impeller outlet relative velocity, impeller outlet absolute airflow angle, impeller outlet relative airflow angle, and impeller outlet circumferential velocity; the geometric parameters of the impeller include impeller inlet radius, impeller inlet blade height, impeller outlet root radius, impeller outlet top radius, impeller outlet blade height, impeller axial width, and number of impeller blades.

[0096] Specifically, the process of determining the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed is as follows:

[0097] The definitions of the load factor Ψ and flow factor φ for the centripetal turbine to be designed are as follows:

[0098]

[0099] Where, ΔH u To reduce the effective specific enthalpy, kJ / kg.

[0100] Based on the optimized design conditions of the centripetal turbine, and in conjunction with setting the flow coefficient, load coefficient, and corresponding total static efficiency of the centripetal turbine to be designed, the isentropic specific enthalpy drop ΔH of the centripetal turbine to be designed is calculated. s And effective specific enthalpy decrease ΔH u Furthermore, based on the isentropic enthalpy drop ΔH of the centripetal turbine to be designed... s And effective specific enthalpy decrease ΔH uThe absolute velocity U4 at the nozzle exit is calculated; the calculation process for the absolute velocity U4 at the nozzle exit is as follows:

[0101]

[0102] Based on the Euler equations for a turbine, the expression for the load factor is rewritten to obtain the rewritten expression for the load factor; the Euler equations for a turbine and the rewritten expression for the load factor are as follows:

[0103] ΔH u =C θ4 U4-C θ6 U6

[0104]

[0105] ε=r6·r4 -1

[0106] Among them, C θ4 C is the tangential velocity at the impeller inlet, in m / s; θ6 U1 is the tangential velocity at the impeller outlet, in m / s; U2 is the circumferential velocity at the impeller outlet, in m / s.

[0107] Due to the circumferential component C of the impeller outlet airflow velocity in actual operating conditions θ6 Since it is very small, it is often assumed to be 0 in the initial design stage of a radial turbine. Therefore, the load factor can be written in the following form:

[0108]

[0109] The ratio of the meridional components of the impeller inlet and outlet velocities is defined as ξ, which is a value very close to 1 but less than 1; therefore, it is often assumed to be 1 in the initial design stage of a centripetal turbine; thus, the flow coefficient expression can be written in the following form:

[0110]

[0111] ξ=C m4 ·C m6 -1

[0112] Based on the above derivation process, the impeller inlet velocity triangle is calculated; the calculation process of the impeller inlet velocity triangle is as follows:

[0113] C θ4 =ΨU4

[0114] C m4 =ξφU4

[0115]

[0116] W θ4 =C θ4 -U4

[0117] W m4 =C m4

[0118]

[0119] Among them, W θ4 W represents the relative velocity at the impeller inlet, in m / s. m4 The impeller inlet meridional velocity is given in m / s.

[0120] The calculation process for the impeller inlet angle of attack is as follows:

[0121] i4=β4-β b4

[0122] Where i4 is the impeller inlet angle of attack; β b4 The impeller inlet blade geometry angle is usually designed to be 0° to avoid generating large bending stress, considering the high speed characteristics of the radial turbine.

[0123] Initialize the nozzle loss coefficient and calculate the isentropic static temperature at the impeller inlet; the calculation process for the isentropic static temperature at the impeller inlet is as follows:

[0124]

[0125] Among them, K n To initialize the nozzle loss coefficient; T s4 The impeller inlet isentropic static temperature; C p is the specific heat capacity of the working fluid at constant pressure, J / (kg·K).

[0126] The calculation process for the state parameters of the impeller inlet is as follows:

[0127]

[0128] Among them, T t4 Total temperature at the impeller inlet, K; P t0 Total pressure at the nozzle inlet, Pa; T t0 ρ is the total temperature at the nozzle inlet, K; k is the specific heat ratio; ρ4 is the impeller inlet density, kg / m³ 3 P4 is the impeller inlet pressure, Pa; T4 is the impeller inlet temperature, K.

[0129] The calculation process for the geometric parameters of the impeller inlet is as follows:

[0130]

[0131]

[0132] Where n is the impeller speed, r / min; B4 is the impeller inlet blockage coefficient; t θ4 Thickness of the impeller inlet blades, in meters (m); N r This represents the number of impeller blades.

[0133] The calculation process for the geometric parameters of the impeller outlet is as follows:

[0134] C θ6 =0

[0135] C m6 =φU4

[0136]

[0137] r h6 =J r r4

[0138]

[0139] Z r =J Z (r s6 -r h6 )

[0140] W θ6 =C θ6 -U6

[0141] W m6 =C m6

[0142]

[0143] Where α6 is the absolute airflow angle at the impeller outlet, in rad; T6 is the impeller outlet temperature, in K; P6 is the impeller outlet pressure, in Pa; and ρ6 is the impeller outlet density, in kg / m³. 3 J r r6 is the ratio of the impeller inlet and outlet radii; r6 is the average impeller outlet radius, in meters. The relative tangential velocity at the impeller outlet is m / s; The impeller outlet relative meridional velocity is given in m / s.

[0144] Step 5: Based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed, and using a pre-built neural network model, obtain the evaluation efficiency of the centripetal turbine to be designed. Specifically, the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed are used as input data and input into the pre-built neural network model, and the evaluation efficiency η of the centripetal turbine to be designed is output.

[0145] Step 6: Determine whether the evaluation efficiency of the centripetal turbine to be designed has converged; if it has converged, determine whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level; if it has not reached the optimal level, update the initial design vector based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and obtain the optimized design vector.

[0146] Specifically, the evaluated efficiency of the centripetal turbine to be designed is compared with its initial efficiency. If the relative deviation between the evaluated efficiency and the initial efficiency is less than or equal to a preset efficiency convergence threshold, the evaluated efficiency of the centripetal turbine to be designed has converged. Otherwise, the evaluated efficiency of the centripetal turbine to be designed has not converged. The initial efficiency of the centripetal turbine to be designed is the total efficiency of the centripetal turbine to be designed.

[0147] The specific process is as follows:

[0148] Step 61: Compare the evaluated efficiency of the centripetal turbine to be designed with the initial efficiency of the centripetal turbine to be designed; if the relative deviation between the evaluated efficiency of the centripetal turbine to be designed and the initial efficiency of the centripetal turbine to be designed is less than or equal to the preset efficiency convergence threshold, then the evaluated efficiency of the centripetal turbine to be designed has converged, and the process jumps to step 62; otherwise, the evaluated efficiency of the centripetal turbine to be designed has not converged, and the process jumps to step 4 until the evaluated efficiency of the centripetal turbine to be designed converges.

[0149] Step 62: Determine whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level; if it has not reached the optimal level, proceed to step 63; if it has reached the optimal level, proceed to step 7.

[0150] Step 63: Update the initial design vector based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and obtain the optimized design vector; wherein, the flow coefficient, load coefficient and speed of the centripetal turbine to be designed in the initial design vector are used as the population, and the combination of flow coefficient, load coefficient and speed corresponding to the optimal evaluation efficiency of the centripetal turbine to be designed is adopted to output the optimized design vector; preferably, the recommended value of the genetic generation is not less than 30.

[0151] Step 7: Based on the optimized design vector, determine whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if so, output the optimized design vector as the design optimization result of the centripetal turbine to be designed. Specifically, based on the optimized design vector, calculate the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed; determine whether the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed meet the material strength requirements of the preset components.

[0152] The specific process is as follows:

[0153] The maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed are compared with the material strength index of the preset components. If the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed are less than or equal to the material strength index of the preset components, then the preset components in the centripetal turbine to be designed meet the material strength requirements, and the optimized design vector is output as the design optimization result of the centripetal turbine to be designed. The design optimization result of the centripetal turbine to be designed also includes the impeller profile data generated based on the optimized design vector. Otherwise, return to step 2 until the preset components in the centripetal turbine to be designed meet the material strength requirements.

[0154] The calculation process for the maximum stress in the impeller of the radial turbine to be designed is as follows:

[0155]

[0156] Where S is the maximum stress of the impeller in the centripetal turbine to be designed; K′ is an empirical coefficient, with a recommended value of 0.3.

[0157] Example explanation:

[0158] Take the optimization design process of a 400kW steam centrifugal turbine as an example.

[0159] 1) Design parameters

[0160] The thermodynamic design parameters of the 400kW steam centripetal turbine are shown in Table 1. The design output power is 400MW, the rotational speed is 9000r / min, the inlet total temperature is 105℃, the total pressure is 0.1208MPa, the outlet pressure is 0.008MPa, and the flow rate is 8.5t / h.

[0161] 2) Design Scheme

[0162] Using the aerodynamic-strength integrated optimization design method for centripetal turbines described in Example 1, the example centripetal turbine was optimized, and the design results are shown in Table 1.

[0163] Table 1. Thermal design results for an example centripetal turbine.

[0164]

[0165]

[0166] As attached Figure 2 As shown, attached Figure 2 The diagram shows the meridional plot of the example centripetal turbine after the optimized design in Example 1; from the appendix... Figure 2As can be seen, the impeller diameter is 930mm, the impeller outlet hub diameter is 260mm, the impeller outlet cover diameter is 474mm, and the impeller inlet blade height is 26.2mm. The turbine's meridian transition is smooth, and the meridian diagram structure is reasonable; as shown in the attached diagram. Figure 3 As shown, attached Figure 3 The figure shows a three-dimensional schematic diagram of the impeller of the example centripetal turbine after the optimized design in Example 1.

[0167] As attached Figure 4 As shown, attached Figure 4 The attached diagram shows the pressure distribution on the blade surface of the example centripetal turbine after the optimized design in Example 1; Figure 4 As can be seen, the pressure in the turbine gradually decreases from the inlet to the outlet, and the surface pressure is evenly distributed.

[0168] As attached Figure 5 As shown, attached Figure 5 The figure shows the efficiency performance curves of the example centripetal turbine after optimization design in Example 1, under varying operating conditions with different flow rates; from the attached figure... Figure 5 As can be seen, the centripetal turbine has the highest efficiency when both the relative flow rate and the relative speed are 1, indicating that the design conditions represent the optimal values ​​for turbine operation.

[0169] As attached Figure 6 As shown, attached Figure 6 The figure shows the pressure ratio versus flow rate performance curves of the optimized radial turbine in Example 1; from the appendix... Figure 6 As can be seen, the pressure ratio increases with the increase of relative flow rate and relative rotational speed, which is consistent with the theory of turbine machinery.

[0170] The centripetal turbine aerodynamic-strength integrated optimization design method described in Embodiment 1, compared with the traditional optimization design method based on genetic algorithms, introduces a neural network model. Based on the experimental data of the centripetal turbine and the database of loss models, the neural network model is trained using the database. At the same time, the turbine efficiency is used as the target variable to complete the design of the centripetal turbine. This avoids the design scheme relying too much on traditional models and improves the accuracy of the optimization design results. In Embodiment 1, through coupled structural optimization design, the aerodynamic performance of the centripetal turbine can be improved while ensuring the rationality of the structure, realizing the integrated aerodynamic-structural design.

[0171] In this embodiment 1, during the optimization design of the centripetal turbine, the artificial neural network model is combined with the genetic algorithm, which can quickly screen out the poor design schemes during the optimization process, overcome the disadvantage of the slow global search speed of the genetic algorithm, and effectively improve the calculation speed of the optimization design.

[0172] Example 2

[0173] As attached Figure 7 As shown, this embodiment 2 provides a centripetal turbine aerodynamic-strength integrated optimization design system, including a design vector setting module, a parameter calculation module, a model evaluation module, a design vector optimization module, and a strength evaluation module.

[0174] The design vector setting module is used to set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed, obtaining the initial design vector. The parameter calculation module is used to determine the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed based on the initial design vector. The model evaluation module is used to obtain the evaluation efficiency of the centripetal turbine to be designed based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed, using a pre-built neural network model. The pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters. The design vector optimization module is used to determine whether the evaluation efficiency of the centripetal turbine to be designed has converged. If converged, it determines whether the evaluation efficiency of the centripetal turbine to be designed has reached its optimum. If not, it updates the initial design vector based on a genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches its optimum, obtaining the optimized design vector. The strength evaluation module, based on the optimized design vector, determines whether the preset components in the centripetal turbine to be designed meet the material strength requirements. If they do, it outputs the optimized design vector as the design optimization result of the centripetal turbine to be designed.

[0175] Optionally, the centripetal turbine aerodynamic-strength integrated optimization design system described in Embodiment 2 further includes a model building module and an efficiency initialization module; the model building module is used to build an artificial neural network model to characterize the relationship between efficiency and aerodynamic parameters and structural parameters, and obtain a pre-built neural network model; the efficiency initialization module is used to initialize the total efficiency of the centripetal turbine to be designed, and obtain the initial efficiency of the centripetal turbine to be designed.

[0176] Example 3

[0177] As attached Figure 8 As shown, this embodiment 3 provides an electronic device, including: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the centripetal turbine aerodynamic-strength integrated optimization design method; or, the processor executing the computer program to implement the functions of each module in the above-mentioned centripetal turbine aerodynamic-strength integrated optimization design system.

[0178] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a preset function, the instruction segments describing the execution process of the computer program in the electronic device.

[0179] The electronic device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are examples of electronic devices and do not constitute a limitation on the electronic device. It may include more components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0180] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor, or any conventional processor. The processor is the control center of the electronic device, connecting various parts of the electronic device through various interfaces and lines.

[0181] The memory can be used to store the computer program and / or module. The processor implements various functions of the electronic device by running or executing the computer program and / or module stored in the memory and by calling the data stored in the memory.

[0182] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function (such as sound playback or image playback). The data storage area may store data created based on the use of the phone (such as audio data or a phonebook). Furthermore, the memory may include high-speed random access memory (RAM) and non-volatile memory, such as hard disks, RAM, plug-in hard disks, SmartMediaCards (SMC), Secure Digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0183] Example 4

[0184] This embodiment 4 also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the centripetal turbine aerodynamic-strength integrated optimization design method.

[0185] If the modules / units integrated in the centripetal turbine aerodynamic-strength integrated optimization design system are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0186] Based on this understanding, the present invention can implement all or part of the processes in the above-mentioned centripetal turbine aerodynamic-strength integrated optimization design method, or it can be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-mentioned centripetal turbine aerodynamic-strength integrated optimization design method. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form, etc.

[0187] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0188] Example 5

[0189] This embodiment 5 provides a computer product, which includes a computer program stored in a computer-readable storage medium. The processor of the electronic device reads the computer program from the computer-readable storage medium and executes the computer program, so that the electronic device can execute the centripetal turbine aerodynamic-strength integrated optimization design method described in embodiment 1, which will not be repeated here.

[0190] It should be noted that those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods.

[0191] The centripetal turbine aerodynamic-strength integrated optimization design described in this invention, by setting an initial design vector, quickly obtains the evaluation efficiency of the centripetal turbine under design using a pre-built neural network model, avoiding the repeated design and correction processes in traditional methods. Utilizing the high computational power of the neural network model, a large number of design parameter combinations can be evaluated in a short time, significantly shortening the development cycle and reducing design costs. After obtaining the evaluation efficiency, if it is not optimal, the initial design vector is updated using a genetic algorithm. The genetic algorithm is a global optimization algorithm that can perform extensive searches within the design space, continuously seeking better combinations of design parameters until the evaluation efficiency of the centripetal turbine under design reaches its optimum. It can deeply explore the optimal solution within the design space, improve the design quality of radial turbines, and enable them to achieve better performance. After obtaining the optimized design vector, it further determines whether the preset components in the radial turbine under design meet the material strength requirements. When the preset components meet the material strength requirements, it outputs the optimized design vector as the design optimization result, fully considering the aerodynamic performance and structural strength requirements of the radial turbine in actual operation, ensuring the reliability and practicality of the design results, and improving the stability and safety of the radial turbine in practical applications. It can quickly and accurately complete the design of radial turbines and predict their performance under varying operating conditions, with fast calculation speed, and at the same time has the best aerodynamic performance and reasonable structure, meeting the engineering design requirements.

[0192] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

Claims

1. A comprehensive aerodynamic-strength optimization design method for centripetal turbines, characterized in that, include: Set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed to obtain the initial design vector; Based on the initial design vector, the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed are determined; Based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed, the evaluation efficiency of the centripetal turbine to be designed is obtained based on a pre-built neural network model; wherein, the pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters. Determine whether the evaluation efficiency of the centripetal turbine to be designed has converged; if it has converged, determine whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level; if it has not reached the optimal level, update the initial design vector based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and obtain the optimized design vector. Based on the optimized design vector, it is determined whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if they do, the optimized design vector is output as the design optimization result of the centripetal turbine to be designed.

2. The aerodynamic-strength integrated optimization design method for a centripetal turbine according to claim 1, characterized in that, The aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed include the aerodynamic parameters of the nozzle, the geometric parameters of the nozzle, the inlet aerodynamic parameters of the impeller, the outlet aerodynamic parameters of the impeller, and the geometric parameters of the impeller.

3. The aerodynamic-strength integrated optimization design method for a centripetal turbine according to claim 1, characterized in that, The process of building a pre-built neural network model includes: Determine the design optimization conditions for the centripetal turbine; Based on the design optimization conditions of centripetal turbines, experimental data and loss models of different models of centripetal turbines were obtained; Based on experimental data and loss models of different types of centripetal turbines, a centripetal turbine design sample library containing the total efficiency under different design parameter conditions is established. Based on a centripetal turbine design sample library containing the total efficiency under different design parameter conditions, a pre-determined artificial neural network model is trained to obtain an artificial neural network model that characterizes the relationship between efficiency and aerodynamic and structural parameters, thus obtaining a pre-constructed neural network model.

4. The aerodynamic-strength integrated optimization design method for a centripetal turbine according to claim 3, characterized in that, The loss models include nozzle loss model, impeller angle of attack loss model, impeller channel loss model, impeller clearance leakage loss model, impeller trailing edge loss model, impeller back friction blow-through loss model, and impeller residual velocity loss model.

5. The aerodynamic-strength integrated optimization design method for a centripetal turbine according to claim 1, characterized in that, The process of determining whether the evaluation efficiency of the centripetal turbine to be designed has converged includes: The evaluated efficiency of the centripetal turbine to be designed is compared with the initial efficiency of the centripetal turbine to be designed. If the relative deviation between the evaluated efficiency of the centripetal turbine to be designed and the initial efficiency of the centripetal turbine to be designed is less than or equal to the preset efficiency convergence threshold, then the evaluated efficiency of the centripetal turbine to be designed has converged; otherwise, the evaluated efficiency of the centripetal turbine to be designed has not converged. The initial efficiency of the centripetal turbine to be designed is the total efficiency of the centripetal turbine to be designed.

6. The aerodynamic-strength integrated optimization design method for a centripetal turbine according to claim 1, characterized in that, The process of determining whether preset components in a centripetal turbine under design meet material strength requirements based on optimized design vectors includes: Based on the optimized design vector, the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed are calculated. Determine whether the maximum stress and maximum deformation of the preset components in the centripetal turbine to be designed meet the material strength requirements of the preset components.

7. A centripetal turbine aerodynamic-strength integrated optimization design system, characterized in that, include: The design vector setting module is used to set the flow coefficient, load coefficient, and rotational speed of the centripetal turbine to be designed, and to obtain the initial design vector. The parameter calculation module is used to determine the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed based on the initial design vector. The model evaluation module is used to obtain the evaluation efficiency of the centripetal turbine to be designed based on the aerodynamic and geometric parameters of the preset components in the centripetal turbine to be designed and a pre-built neural network model. The pre-built neural network model is an artificial neural network model used to characterize the relationship between efficiency and aerodynamic and structural parameters. The design vector optimization module is used to determine whether the evaluation efficiency of the centripetal turbine to be designed has converged. If it has converged, it determines whether the evaluation efficiency of the centripetal turbine to be designed has reached the optimal level. If it has not reached the optimal level, the initial design vector is updated based on the genetic algorithm until the evaluation efficiency of the centripetal turbine to be designed reaches the optimal level, and the optimized design vector is obtained. The strength assessment module, based on the optimized design vector, determines whether the preset components in the centripetal turbine to be designed meet the material strength requirements; if they do, it outputs the optimized design vector as the design optimization result of the centripetal turbine to be designed.

8. An electronic device, characterized in that, include: A processor is used to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, performs the centripetal turbine aerodynamic-strength integrated optimization design method as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the centripetal turbine aerodynamic-strength integrated optimization design method as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the centripetal turbine aerodynamic-strength integrated optimization design method as described in any one of claims 1-6.

Images