Centrifugal fan design method based on curve collaborative optimization and centrifugal fan

Abstract

The invention relates to the field of fan pneumatic optimization, and provides a centrifugal fan design method based on curve collaborative optimization and a centrifugal fan in order to effectively solve the problems of fan energy consumption and ventilation quantity, a Bezier curve is adopted to construct transition contours of an air inlet cylinder, an air outlet cylinder and a volute, and a divergent curve is adopted to construct a volute flow channel molded line; then, a Bezier curve control point and a gradual expansion curve control parameter are used as optimization variables, and total pressure efficiency maximization and air volume improvement are used as objective functions; and finally, collaborative optimization is carried out through a genetic algorithm and an automatic process of CFD simulation. Through collaborative optimization of the divergent curve and the Bezier curve, loss mechanisms such as flow separation and vortex generation in the volute are effectively inhibited, so that the internal flow loss is reduced, finally, the air volume and the efficiency are improved under the same power consumption, and the purposes of saving energy and improving efficiency are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fan aerodynamic optimization, and particularly relates to a centrifugal fan design method based on curve collaborative optimization. BACKGROUND

[0002] As the core equipment of the grain warehouse ventilation system, the aerodynamic performance of the centrifugal fan directly affects the quality of the stored grain and the operating cost of the system. In the field of grain storage, the ventilation system needs to be continuously operated for a long time, and the energy consumption of the fan usually accounts for more than 40% of the total energy consumption of the grain warehouse. At the same time, the high-resistance characteristics of the grain pile and the dust prevention requirements pose special challenges to the aerodynamic performance of the fan: sufficient air volume needs to be provided to overcome high resistance. The traditional optimization method of centrifugal fan is mainly aimed at general industrial scenarios, and lacks targeted optimization for special conditions of grain warehouse. SUMMARY

[0003] In order to effectively solve the problems of fan energy consumption and ventilation volume, the present application provides a centrifugal fan design method based on curve collaborative optimization and a centrifugal fan.

[0004] The technical scheme adopted by the present application to solve the above problems is:

[0005] The centrifugal fan design method based on curve collaborative optimization comprises:

[0006] Step 1: Establish a parameterized three-dimensional model of the fan, wherein the transition profile of the inlet duct, the outlet duct and the volute adopts a Bezier curve configuration, and the volute flow passage profile adopts a gradual expansion curve configuration;

[0007] Step 2: Establish an optimization objective function;

[0008] Step 3: Optimize and solve the optimization objective function based on the parameterized three-dimensional model and CFD simulation calculation to obtain the optimal curve parameter combination.

[0009] Further, the Bezier curve is a third-order Bezier curve, which is expressed as: wherein, determining the starting position of the connection between the inlet duct and the volute, , used to adjust the curvature change of the transition profile, determining the connection end point of the transition profile and the volute body, and t is an auxiliary parameter.

[0010] Further, the gradual expansion curve is expressed as: , is the starting radius, is the unwinding angle of the volute, , , respectively control the flatness of the starting segment, the expansion rate of the middle segment and the convergence trend of the end segment.

[0011] Further, the optimization objective function is a multi-objective function.

[0012] Further, the fan total pressure efficiency and outlet air volume are multi-objective optimized.

[0013] Further, the total pressure efficiency calculation method is: η is the total pressure efficiency, P is the total pressure, Q is the outlet air volume, T is the torque, and ω is the angular velocity.

[0014] Further, a multi-objective genetic algorithm with an elite reservation strategy is used for optimization and solution.

[0015] Further, the optimization process is specifically:

[0016] Generate an optimization variable combination based on a genetic algorithm;

[0017] Update the fan geometry based on the optimization variable combination driving parameterization model;

[0018] Carry out CFD grid division and flow field calculation;

[0019] The calculated total pressure efficiency and outlet air volume are fed back to the genetic algorithm for cyclic optimization.

[0020] Further, the population size is 80, the maximum iteration number is 150, the crossover probability is 0.8, and the mutation probability is 0.05.

[0021] The centrifugal fan comprises an air inlet cylinder, an air outlet cylinder and a symmetrical guide volute connecting the air inlet cylinder and the air outlet cylinder, the transition profile of the air inlet cylinder, the air outlet cylinder and the volute adopts a Bezier curve configuration, the volute flow passage profile adopts a gradual expansion curve configuration, and the Bezier curve and the gradual expansion curve are determined by a centrifugal fan design method based on curve collaborative optimization.

[0022] The present application has the beneficial effects compared with the prior art: the Bezier curve is used to construct the transition profile of the air inlet cylinder, the air outlet cylinder and the volute, and the gradual expansion curve is used to construct the volute flow passage profile; then the Bezier curve control points and the gradual expansion curve control parameters are used as optimization variables, and the maximum total pressure efficiency and air volume improvement are used as objective functions; finally, the genetic algorithm and the CFD simulation automatic process are used for collaborative optimization. Through the collaborative optimization of the gradual expansion curve and the Bezier curve, the flow separation, vortex generation and other loss mechanisms in the volute are effectively suppressed, thereby reducing the internal flow loss, ultimately improving the air volume and efficiency under the same power consumption, and achieving the purpose of energy saving and efficiency improvement. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 It is a flow chart of the centrifugal fan design method based on curve collaborative optimization.

[0024] Figure 2 This is a schematic diagram of the fan structure;

[0025] Figure 3 A schematic diagram of the wind turbine optimization curve;

[0026] Figure 4 A schematic diagram of the Bézier curve before optimization;

[0027] Figure 5 A schematic diagram of the optimized Bézier curve;

[0028] Reference numerals: 1 is the air inlet, 2 is the air outlet, 3 is the volute, 4 is the transition profile, and 5 is the flow channel profile. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0030] Existing research on fan optimization design mainly focuses on the impeller structure and the Archimedean spiral shape of the volute, neglecting the flow field coupling effect between the inlet / outlet duct and the volute. Secondly, the curve design methods are simplistic, often employing fixed-curvature circular arcs or straight-line splices, failing to achieve continuous and smooth control of airflow parameters, leading to severe flow separation and vortex phenomena. Based on this, this invention proposes a centrifugal fan design method based on curve co-optimization, such as... Figure 1 As shown, it includes:

[0031] Step 1: Establish a parametric 3D model of the fan, in which the transition contours of the inlet duct, outlet duct and volute adopt the Bézier curve configuration, and the flow channel profile of the volute adopts the gradually expanding curve configuration.

[0032] The Bézier curve is a third-order Bézier curve, represented as: ,in, Determine the starting position for the connection between the air inlet duct and the volute. , Used to adjust the curvature variation of the transition profile. Determine the connection endpoint between the transition profile and the volute body, where t is an auxiliary parameter with a range of [0,1].

[0033] The gradual expansion curve is represented as: , The starting radius, The angle at which the volute unfolds. , , The smoothness of the initial section of the flow channel, the expansion rate of the middle section, and the convergence trend of the final section are controlled respectively.

[0034] Step 2: Establish the optimization objective function. Based on improving energy consumption using Bézier curves and asymptotic curves, the objective function can focus solely on ventilation volume. In this embodiment, to further optimize energy consumption, multi-objective collaborative optimization is performed on the fan's total pressure efficiency and outlet air volume.

[0035] Step 3: Based on the parametric 3D model and CFD simulation calculation, optimize the objective function to obtain the optimal combination of curve parameters.

[0036] In this embodiment, a multi-objective genetic algorithm with an elite retention strategy is used for optimization. The specific process is as follows:

[0037] Genetic algorithms are used to generate optimal variable combinations, specifically including: , , , , , and ;

[0038] Update the wind turbine geometry based on a parameterized model driven by optimized variable combinations;

[0039] Perform CFD mesh generation and flow field calculation;

[0040] The calculated total pressure efficiency and outlet air volume are fed back into the genetic algorithm for iterative optimization.

[0041] The third-order Bézier curve exhibits unique advantages in the design of the transition profile between the inlet and outlet ducts and the volute of a wind turbine: its four control points ( , , , The flexible configuration of the third-order Bézier curve allows for highly controllable curve shape, precisely matching the natural path of airflow. Compared to traditional circular transitions, the third-order Bézier curve enables continuous curvature variation, effectively eliminating flow separation and ensuring a smooth airflow path. When airflow passes over a curved wall, a pressure gradient is generated due to centrifugal force. Sudden changes in curvature can lead to pressure abrupt changes, causing airflow separation and eddies, increasing energy loss. The third-order Bézier curve ensures that the curvature is a continuous function, meaning that the change in centrifugal force on the airflow is smooth, effectively "adhering" to the wall, eliminating flow separation, and reducing local drag loss. In practical applications, this smooth transition reduces local drag loss and improves the uniformity of airflow distribution within the volute.

[0042] The expansion curve is a parametric curve with continuous expansion characteristics. Its core principle is to control the expansion law of the flow channel cross-section through a polynomial function. Its three control parameters ( , , These correspond to the smoothness of the initial section, the expansion rate of the middle section, and the convergence trend of the final section of the flow channel, respectively. This design allows the flow channel cross-section to expand smoothly along the airflow direction, possessing the following geometric characteristics:

[0043] Continuous expansion: The flow channel area increases monotonically with increasing angle;

[0044] Curvature controllability: Different expansion rates can be achieved by adjusting parameters;

[0045] Airflow adaptability: The curve shape conforms to the natural diffusion law of airflow.

[0046] By employing a gradually expanding curve, the airflow achieves a smooth conversion of velocity energy to pressure energy within the volute, avoiding the energy loss caused by sudden expansion in traditional constant-width volutes. Optimizing control parameters maintains a stable pressure gradient throughout the flow channel, significantly improving energy conversion efficiency.

[0047] To accurately evaluate the aerodynamic performance after co-optimization of the expansion curve and the Bezier curve, a reliable numerical simulation model is needed to simulate the complex flow characteristics inside the fan. This invention employs computational fluid dynamics (CFD) to achieve accurate simulation of the three-dimensional flow field by solving the Reynolds-averaged Navier-Stokes (RANS) equations. The selection of the turbulence model is crucial to the simulation accuracy; the standard k-ε model is chosen due to its maturity and stability in simulating flows inside rotating machinery. This model closes the governing equations by solving the transport equations for the turbulent kinetic energy k and the turbulent dissipation rate ε. Its basic equations are as follows:

[0048] ,

[0049] ,

[0050] In the formula, The density of the fluid, in units k represents turbulent kinetic energy, characterizing the intensity of turbulent fluctuations, in units of... 3 t represents time, in seconds (s). Dynamic viscosity of fluid molecules, in units of 1000 kJ / m³. ; This represents the component of the velocity vector in the i-direction, with units of m / s; and In terms of spatial coordinate direction (i, j=1,2,3); Turbulent viscosity, in units of ,Depend on The calculation shows that, among which , is an empirical constant; The values ​​are Prandtl numbers for turbulence, taken as 1.0 and 1.3 respectively; The turbulent kinetic energy generation term is calculated from the average velocity gradient, and its formula is: S is the mean strain rate tensor, in units of ; The turbulent dissipation rate is expressed in m² / s³, representing the rate of turbulent kinetic energy dissipation. These are empirical constants, with standard values ​​of 1.44 and 1.92, respectively.

[0051] The simulation principle lies in constructing the key parameter, turbulent viscosity, by solving the transport equations for the two variables: turbulent kinetic energy k and turbulent dissipation rate ε. This allows for the closure of the Reynolds-averaged Navier-Stokes equations in a relatively simplified manner, enabling effective simulation of complex turbulent phenomena. A key advantage of this method is that it introduces empirical constants (such as...) Furthermore, a universal relationship between k and ε is established, which can still provide sufficiently reliable and stable turbulence prediction results for engineering applications without the huge computational cost of directly analyzing all turbulence scales. This makes it possible to conduct rapid and efficient parameterization research and performance optimization of the three-dimensional flow field inside the wind turbine.

[0052] Seven optimization variables (coordinates of four Bezier control points + three asymptotic parameters) together constitute the optimization space, and global optimization is achieved through a genetic algorithm. This collaborative mechanism ensures a perfect match between the transition profile of the inlet and outlet air ducts and the volute flow channel profile, enabling seamless airflow throughout the entire flow channel.

[0053] Correspondingly, such as Figure 2 As shown, the present invention also provides a centrifugal fan, including an inlet duct 1, an outlet duct 2, and a symmetrical guide volute 3 connecting the inlet duct 1 and the outlet duct 2. The transition profile 4 of the inlet duct 1, the outlet duct 2, and the volute 3 adopts a Bezier curve configuration, and the flow channel profile 5 of the volute 3 adopts a gradually expanding curve configuration, as shown. Figure 3 As shown, the Bézier curve and the incremental curve were determined using a centrifugal fan design method based on curve co-optimization.

[0054] Taking the optimization of this wind turbine as an example, each step will be explained in detail:

[0055] First, based on the target operating conditions and preliminary design, the core parameters of the fan are determined: a mixed-flow centrifugal fan structure is adopted, with an impeller outer diameter D=650mm, a rotational speed n=1450rpm, an inlet cylinder diameter D1=500mm, and an outlet cylinder diameter D2=500mm.

[0056] Using the parametric function of 3D modeling software, a fully parametric model of the fan was established. The transition profile between the inlet duct, outlet duct, and symmetrical guide volute was defined as a third-order Bézier curve, which is composed of four control points ( , , , The two-dimensional or three-dimensional coordinates of the four control points are uniquely determined; in the parametric model, the coordinates of these four control points are set as variable. The flow profile of the symmetrical guide volute is defined as a continuously expanding profile, and its expansion law is described by a polynomial function, the three key coefficients of which ( , , As a control parameter, it determines the smoothness of the initial section of the flow channel, the expansion rate of the middle section, and the convergence trend of the final section. These coefficients are also set as variables.

[0057] Secondly, a synergistic optimization objective based on maximizing the total pressure efficiency of the fan and maximizing the outlet air volume is constructed, denoted as: Max(η), Max(Q), , The total pressure efficiency is given by P, where P is the total pressure in Pa, and Q is the outlet air volume in m³ / s. 3 / s, T is torque, in N·m; ω is angular velocity, in rad / s.

[0058] Then, an automated simulation optimization process is built:

[0059] 1. Integration and encoding of optimization variables: The seven optimization variables (coordinates of four Bézier curve control points and three inflection curve control parameters) are uniformly encoded to form an "individual" (i.e. a potential design scheme) in the genetic algorithm.

[0060] 2. Automated Simulation Workflow Construction: Integrating parametric modeling software, CFD simulation software, and a genetic algorithm optimization platform, the following automated workflow is constructed: a. The genetic algorithm generates a set of variable values ​​(one individual). b. This set of variables is automatically transferred to the parametric model, driving the software to update the wind turbine's 3D geometry. c. The updated 3D model is automatically imported into the CFD software and meshed (using an unstructured mesh, with boundary layer refinement in the near-wall region). d. CFD boundary conditions are automatically set: the inlet is a mass flow rate inlet; the outlet is a pressure outlet, the values ​​of which are calculated using the empirical formula for the system resistance of the grain pile, the formula being: Where v is the wind speed (m / min), P is the grain pile resistance (Pa), and h is the grain pile height (m); the Standard k-ε model is used for the turbulence model. e. The CFD solver automatically runs and calculates the flow field, and automatically extracts the data required for the objective function: fan efficiency (η) and actual outlet air volume (Q). f. The η and Q values ​​are returned to the genetic algorithm platform to calculate the fitness of the individual.

[0061] 3. Genetic algorithm parameters: population size is 80, maximum number of iterations is 150, crossover probability is 0.8, and mutation probability is 0.05. The selection mechanism adopts an elite retention strategy and a tournament selection method.

[0062] 4. Result Convergence and Solution Selection: After approximately 120 iterations, the Pareto front stabilizes, indicating that the algorithm has converged. From the converged Pareto solution set, an equilibrium solution is selected as the final design scheme based on practical engineering preferences.

[0063] The Bézier curves before and after optimization are as follows: Figure 4 , 5 As shown in the table below, the specific numerical comparisons are as follows:

[0064] Scheme Air volume (m3 / h) Efficiency (%) Model before optimization 11768 75 Model after optimization 12351 83

[0065] After optimization using this method, the fan's air volume increased from 11768 m³ / h to 12351 m³ / h, an increase of 4.9%, significantly improving the fan's aerodynamic performance. The fan's total pressure efficiency increased from 75% to 83%, an increase of 8%, greatly reducing energy consumption during grain silo ventilation.

Claims

1. A centrifugal fan design method based on curve collaborative optimization, characterized in that, include: Step 1: Establish a parametric 3D model of the fan, in which the transition contours of the inlet duct, outlet duct and volute adopt the Bézier curve configuration, and the flow channel profile of the volute adopts the gradually expanding curve configuration. Step 2: Establish the objective function for optimization; Step 3: Based on the parametric 3D model and CFD simulation calculation, optimize the objective function to obtain the optimal combination of curve parameters.

2. The centrifugal fan design method based on curve collaborative optimization according to claim 1, characterized in that, The Bézier curve is a third-order Bézier curve, represented as: ,in, Determine the starting position for the connection between the air inlet duct and the volute. , Used to adjust the curvature variation of the transition profile. Determine the connection endpoint between the transition profile and the volute body, where t is an auxiliary parameter.

3. The centrifugal fan design method based on curve collaborative optimization according to claim 2, characterized in that, The gradual expansion curve is represented as: , The starting radius, The angle at which the volute unfolds. , , The smoothness of the initial section of the flow channel, the expansion rate of the middle section, and the convergence trend of the final section are controlled respectively.

4. The centrifugal fan design method based on curve collaborative optimization according to claim 1, characterized in that, The objective function is optimized into a multi-objective function.

5. The centrifugal fan design method based on curve collaborative optimization according to claim 4, characterized in that, Multi-objective optimization of the total pressure efficiency and outlet air volume of the fan is performed.

6. The centrifugal fan design method based on curve collaborative optimization according to claim 5, characterized in that, The total pressure efficiency is calculated as follows: η is the total pressure efficiency; P is the total pressure; Q is the outlet air volume; T is the torque; and ω is the angular velocity.

7. The centrifugal fan design method based on curve collaborative optimization according to claim 6, characterized in that, A multi-objective genetic algorithm with an elite retention strategy is used for optimization.

8. The centrifugal fan design method based on curve collaborative optimization according to claim 7, characterized in that, The optimization process is as follows: Genetic algorithm-based generation of optimal variable combinations; Update the wind turbine geometry based on a parameterized model driven by optimized variable combinations; Perform CFD mesh generation and flow field calculation; The calculated total pressure efficiency and outlet air volume are fed back into the genetic algorithm for iterative optimization.

9. The centrifugal fan design method based on curve collaborative optimization according to claim 7, characterized in that, The population size is 80, the maximum number of iterations is 150, the crossover probability is 0.8, and the mutation probability is 0.

05.

10. A centrifugal fan, comprising an inlet duct, an outlet duct, and a symmetrical guide volute connecting the inlet duct and the outlet duct, characterized in that, The transition profiles of the air inlet duct, air outlet duct, and volute are configured using a Bézier curve, and the flow channel profile of the volute is configured using a gradually expanding curve. The Bézier curve and the gradually expanding curve are determined using the centrifugal fan design method based on curve co-optimization as described in any one of claims 1-9.

Images