Parametric design method for aero-engine particle separator

Abstract

The aero-engine particle separator parameterization design method comprises the following specific steps that firstly, air parameters of dust-containing airflow are determined, and meanwhile material parameters of sand dust are determined; 2, designing initial geometric parameters of the particle separator; 3, a parameter optimization range is given, an optimization algorithm is utilized, the total loss delta P serves as an optimization target, loss minimization serves as an optimization direction, new delta P is compared with original loss, and a new geometric parameter combination with the minimum total loss is obtained; and 4, modeling by using a new geometric parameter combination, checking calculation by using three-dimensional simulation, carrying out iteration with the step 3 according to the simulation result satisfaction degree, and otherwise, finishing the design. According to the method, the parameterization design method is adopted, the main geometric parameters and the performance parameters of the separator are associated, the loss model is formed, the model is optimized and solved, the geometric parameter combination with the minimum loss is obtained, the experience dependence degree of the prior art can be relieved, and the design efficiency is greatly improved.

Description

Technical field:

[0001] The invention relates to a parametric design method for an aero-engine particle separator, which is suitable for the initial design of a helicopter engine particle separator with an intake dust particle filtering requirement, and belongs to the field of aero-engines. Background technology:

[0002] Particle separators are mainly used in helicopter engines. Since helicopters need to perform take-off and landing missions in mountainous areas, deserts, ice and snow, and often need to hover at low altitude or fly close to the ground, a large amount of sand, dust, ice and snow and other foreign objects on the ground will be sucked into the engine, which will reduce the engine performance, shorten its life and even cause failure. For this reason, a particle separator needs to be installed at the inlet of the helicopter engine to enhance the engine's air intake protection capability. Common particle separators generally include arresting type, multi-tube type and integral type. Among them, the integral particle separator is currently the most widely used. The integral particle separator is integrated with the engine and integrates multiple functions. The working principle of the integral particle separator is to use the inertia of the movement of foreign objects contained in the air to separate the foreign objects from the main airflow before the air enters the engine compressor, and discharge them to the outside of the machine through the scavenging flow.

[0003] Figure 1 The flow path schematic diagram of the conventional particle separator is given. As shown in the figure, taking the common engine intake flow path as an example, there are two flow paths in the engine intake duct, namely the main flow path and the scavenging flow path. The main flow path is the clean air after the sand and dust are filtered out by the particle separator, that is, the airflow that actually enters the engine to work, and the filtered sand and dust are collected through the outlet of the scavenging flow path. This conventional particle separator needs to be assisted by a blower in the scavenging flow path to further enhance the negative pressure of the flow path, so that the sand and dust are more easily sucked into the scavenging flow path. Its working principle is: after passing through the pre-swirl blades at the front end of the intake duct, the dust-containing air flow rotates after passing through the pre-swirl blades, and under the action of centrifugal force, the air and sand and dust are inertially separated, thereby filtering the sand and dust out of the air; the separated sand and dust and clean air pass through the diversion lip and are divided into the main airflow and the scavenging flow. The main airflow passes through the anti-swirl blades, eliminates the swirl, and enters the compressor. The scavenging flow containing sand and dust is guided by the scavenging blades and discharged from the engine through the sand discharge volute.

[0004] The above-mentioned existing particle separator has a relatively complex structure, requiring the design of pre-rotation blades and the use of a blower for exhaust. It has certain disadvantages in terms of both volume and weight. For aircraft engines, both weight and volume should be minimized, and the structure should be simplified as much as possible to enhance reliability and maintainability.

[0005] At the same time, the design of existing particle separators mostly relies on manual experience, and the design results need to be verified by experiments and iterated with the design before they can be finalized.

[0006] Therefore, it is necessary to improve the existing technology to solve the shortcomings of the existing technology. Summary of the invention:

[0007] The present invention provides a parametric design method for an aero-engine particle separator in order to solve the problems existing in the above-mentioned prior art. The method adopts a parametric design method to associate the main geometric parameters of the separator with the performance parameters to form a loss model. The model is optimized and solved to obtain a combination of geometric parameters that minimizes the loss, which can alleviate the degree of experience dependence of the existing technology and greatly improve the design efficiency.

[0008] The present invention adopts the following technical scheme: a parametric design method for an aircraft engine particle separator, the particle separator comprises a straight tube and a conical tube which are connected and internally interpenetrating, the end of the straight tube away from the conical tube is connected with an air outlet tube interpenetrating with the internal space of the straight tube, one side of the straight tube is provided with a separator inlet interpenetrating with the internal space of the straight tube, the end of the straight tube away from the air outlet tube is formed with an inlet, the end of the conical tube away from the straight tube is formed with a first outlet, and the end of the air outlet tube away from the straight tube is formed with a second outlet, the parametric design method for the aircraft engine particle separator, the specific steps are as follows:

[0009] Step 1: Determine the air parameters of the dust-laden airflow and the material parameters of the sand and dust;

[0010] Step 2: Design the initial geometric parameters of the particle separator;

[0011] Step 3: Given the parameter optimization range, use the optimization algorithm, take the total loss ΔP as the optimization target, take the loss minimization as the optimization direction, compare the new ΔP with the original loss, and obtain the new geometric parameter combination with the minimum total loss;

[0012] Step 4: Use the new combination of geometric parameters to build a model and verify it with 3D simulation. Depending on the degree of satisfaction of the simulation results, iterate with step 3 as appropriate. Otherwise, the design is completed.

[0013] Furthermore, the air parameters of the dust-laden airflow include air parameters such as flow rate, gas viscosity value, and sand content ratio.

[0014] Furthermore, the material parameters of the sand and dust include density and particle size.

[0015] Furthermore, step 2 is as follows:

[0016] Determine the particle separator cylinder diameter D and the main flow path diameter D x , Sand discharge port diameter D d 、Separator total height H t , cone part height H c, a combination of geometric parameters of tangential inlet height a and tangential inlet width b, wherein the ratio of the main flow path diameter to the particle separator cylinder diameter is 0.4-0.6, the ratio of the cross-sectional area of ​​the main flow path to the sand-containing airflow inlet area is 0.9-1.0, a / b≈2, and the cone angle is 20-25°;

[0017] Use the following formula to find the flow loss of the initial solution;

[0018] Inlet velocity:

[0019] Main flow outlet speed:

[0020] Tangential velocity of the inner wall of the separation cylinder:

[0021] Rotation speed of the inner vortex in the separation cylinder:

[0022] Resistance loss in the separation cylinder:

[0023] Rotating vortex core and main airflow resistance loss:

[0024] Total loss: ΔP = ΔP b +ΔP x

[0025] in, is the flow rate of sand-laden air, R in is the average swirl radius at the inlet, A r is the total inner surface area of ​​the separator, and ρ is the air density.

[0026] Furthermore, step three is optimized using a genetic algorithm, as follows:

[0027] Step 1: Define the variables as the particle separator cylinder diameter D and the main flow path diameter D x 、Separator total height H t , cone part height H c , tangential inlet height a, tangential inlet width b, the constraint condition is the given parameter optimization range, and the objective function f(x1,x2,x3…) is the total loss relationship;

[0028] Step 2: Encode the variables appropriately to facilitate operation in the genetic algorithm. Represent the xi value in f(x1,x2,x3…) as a binary string and encode it into a binary gene string.

[0029] Step 3: Heuristically generate an initial population of size N, where each individual is an xi value represented by a set of binary gene strings;

[0030] Step 4: Decode the binary gene string of each individual into the corresponding xi, calculate the value of the objective function, and use it as the fitness function value score;

[0031] Step 5: Select individuals according to the value of the fitness function, and use selection operations to select individuals with high fitness so that they can be repeated in the next generation;

[0032] Step 6: In each generation, new individuals are generated through crossover and mutation operations, where the crossover operation is to combine the gene strings of two individuals to create a new individual, and the mutation operation is to change a gene in an individual to create a new individual;

[0033] Step 7: Repeat steps 4 to 6. After multiple iterations, the best individual in the population represents the optimal solution of f(x1, x2, x3…); finally, decide when to stop iterating based on the convergence of the algorithm or other evaluation criteria.

[0034] The present invention has the following beneficial effects:

[0035] (1) Reduce the structural complexity of conventional particle separators and improve reliability and maintainability;

[0036] (2) Eliminating the pre-rotation blades and blower of conventional particle separators, optimizing volume and weight, and saving costs;

[0037] (3) The tangential air intake method is adopted to utilize the self-rotation of air and dust, eliminating the need for pre-rotation blades; the powerful suction capacity of the engine itself is utilized to increase the rotation speed of the sand-containing airflow in the separator, greatly enhancing the separation efficiency and eliminating the need for a blower; the separated sand and dust and the clean airflow move in opposite directions, and the sand and dust no longer need to be collected by a specially designed sand discharge volute. Description of the drawings:

[0038] Figure 1 Schematic diagram of the flow path of a conventional particle separator.

[0039] Figure 2 Schematic diagram of the particle separator structure.

[0040] Figure 3 This is the working principle diagram of the particle separator.

[0041] Figure 4 Schematic diagram of variable parameters for particle separator design.

[0042] Figure 5 The present invention is a flow chart of the parametric design method of an aircraft engine particle separator. Specific implementation method:

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] The particle separator adopted in the present invention comprises a straight cylinder 2 and a conical cylinder 3 which are connected and internally interconnected. An end of the straight cylinder 2 away from the conical cylinder 3 is connected to an air outlet cylinder 4 which is interconnected with the internal space of the straight cylinder 2. A separator inlet 1 which is interconnected with the internal space of the straight cylinder 2 is provided on one side of the straight cylinder 2. An inlet 41 is formed at the end of the straight cylinder 2 away from the air outlet cylinder 4. A first outlet 31 is formed at the end of the conical cylinder 3 away from the straight cylinder 2. A second outlet 42 is formed at the end of the air outlet cylinder 4 away from the straight cylinder 2.

[0045] The working principle of the particle separator is as follows:

[0046] The dust-laden airflow enters the straight cylinder 2 tangentially from the separator inlet 1 at a certain speed V1, and the sand-laden airflow moves in a circular motion around the straight cylinder 2. Since the mass of the sand and dust is much greater than that of the gas, under the action of inertia, the rotation radius of the sand and dust particles will be greater than that of the gas, so the sand and dust and the gas will be separated. To enhance this separation effect, the circumferential radius is further gradually reduced below the straight cylinder 2 to form a cone cylinder 3. The sand and dust are separated from the first outlet 31 below the cone cylinder 3 at a certain speed V2, and the clean airflow flows out from the inlet 41 of the air outlet cylinder 4 to the second outlet 42 at a certain speed V3.

[0047] The parametric design method of an aircraft engine particle separator of the present invention specifically comprises the following steps:

[0048] Step 1: Determine the air parameters of the dusty airflow, such as flow rate, gas viscosity, sand content ratio, etc., and determine the material parameters of the sand and dust, such as density, particle size, etc. Generally speaking, the material parameters of the sand and dust used can be determined according to the specifications;

[0049] Step 2: Design the initial geometric parameters of the particle separator and preliminarily determine the particle separator cylinder diameter D and the main flow path diameter D x , Sand discharge port diameter D d 、Separator total height H t , cone part height H c , a combination of geometric parameters of tangential inlet height a and tangential inlet width b, wherein the ratio of the main flow path diameter to the particle separator cylinder diameter is 0.4-0.6, the ratio of the cross-sectional area of ​​the main flow path to the sand-containing airflow inlet (referred to as the inlet) area is 0.9-1.0, a / b≈2, and the cone angle is 20-25°;

[0050] Use the following formula to find the flow loss of the initial solution;

[0051] Inlet velocity:

[0052] Main flow outlet speed:

[0053] Tangential velocity of the inner wall of the separation cylinder:

[0054] Rotation speed of the inner vortex in the separation cylinder:

[0055] Resistance loss in the separation cylinder:

[0056] Rotating vortex core and main airflow resistance loss:

[0057] Total loss: ΔP = ΔP b +ΔP x

[0058] in, is the flow rate of sand-laden air, R in is the average swirl radius at the inlet, A r is the total inner surface area of ​​the separator, and ρ is the air density.

[0059] Step 3: Given the parameter optimization range, use the optimization algorithm, take the total loss ΔP as the optimization target, take the loss minimization as the optimization direction, compare the new ΔP with the original loss, and obtain a new geometric parameter combination with the minimum total loss; This patent uses genetic algorithm for optimization. The first step defines the variables as the particle separator cylinder diameter D and the main flow path diameter D x 、Separator total height H t , cone part height H c , tangential inlet height a, tangential inlet width b, the constraints are the given parameter optimization range, and the objective function f(x1, x2, x3...) is the total loss relationship; the second step is to encode the variables appropriately to facilitate operations in the genetic algorithm. For example, the xi value in f(x1, x2, x3...) can be represented as a binary string, so that it can be encoded into a binary gene string; the third step is to heuristically generate a set of initial populations of size N, in which each individual is an xi value represented by a set of binary gene strings; the fourth step is to decode the binary gene string of each individual into the corresponding xi and calculate the value of the objective function , and use it as the fitness function value score; the fifth step is to select individuals according to the value of the fitness function, and use the selection operation to select individuals with high fitness so as to be repeated in the next generation; the sixth step is to generate new individuals in each generation through crossover and mutation operations, where the crossover operation is to combine the gene strings of two individuals to create a new individual, and the mutation operation is to change a gene in an individual to create a new individual; repeat steps 4 to 6, and after multiple iterations, the best individual in the population represents the optimal solution of f(x1,x2,x3…); finally, decide when to stop the iteration based on the convergence of the algorithm or other evaluation criteria.

[0060] Step 4: Use a new combination of geometric parameters to build a model, and use 3D simulation for verification. Iterate with step 3 according to the degree of satisfaction of the simulation results, otherwise the design is completed. This patent simulation uses commercial fluid simulation software for 3D simulation. First, grid division is performed. The second step is to perform pre-calculation processing such as boundary condition setting and interface definition. The third step is to perform 3D numerical simulation. Finally, the results are analyzed, mainly analyzing the flow field quality and total pressure loss. The simulation method here is common knowledge and does not belong to the focus of the invention. These will not be repeated.

[0061] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A parametric design method for an aircraft engine particle separator, characterized in that: The particle separator comprises a straight cylinder (2) and a conical cylinder (3) which are connected and internally interpenetrating, the end of the straight cylinder (2) away from the conical cylinder (3) is connected to an outlet cylinder (4) interpenetrating with the internal space of the straight cylinder (2), one side of the straight cylinder (2) is provided with a separator inlet (1) interpenetrating with the internal space of the straight cylinder (2), the end of the straight cylinder (2) away from the outlet cylinder (4) is formed with an inlet (41), the end of the conical cylinder (3) away from the straight cylinder (2) is formed with a first outlet (31), and the end of the outlet cylinder (4) away from the straight cylinder (2) is formed with a second outlet (42), and the parametric design method of the aircraft engine particle separator comprises the following specific steps: Step 1: Determine the air parameters of the dust-laden airflow and the material parameters of the sand and dust; Step 2: Design the initial geometric parameters of the particle separator; Step 3: Given the parameter optimization range, use the optimization algorithm, take the total loss ΔP as the optimization target, take the loss minimization as the optimization direction, compare the new ΔP with the original loss, and obtain the new geometric parameter combination with the minimum total loss; Step 4: Use the new combination of geometric parameters to build a model and verify it with 3D simulation. Depending on the degree of satisfaction of the simulation results, iterate with step 3 as appropriate. Otherwise, the design is completed.

2. The parametric design method for an aircraft engine particle separator according to claim 1, characterized in that: The air parameters of dust-laden air flow include air parameters such as flow rate, gas viscosity value, and sand content ratio.

3. The parametric design method for an aircraft engine particle separator according to claim 2, characterized in that: The material parameters of sand and dust include density and particle size.

4. The parametric design method for an aircraft engine particle separator according to claim 3, characterized in that: Step 2 is as follows: Determine the particle separator cylinder diameter D and the main flow path diameter D x , Sand discharge port diameter D d 、Separator total height H t , cone part height H c , a combination of geometric parameters of tangential inlet height a and tangential inlet width b, wherein the ratio of the main flow path diameter to the particle separator cylinder diameter is 0.4-0.6, the ratio of the cross-sectional area of ​​the main flow path to the sand-containing airflow inlet area is 0.9-1.0, a / b≈2, and the cone angle is 20-25°; Use the following formula to find the flow loss of the initial solution; Inlet velocity: Main flow outlet speed: Tangential velocity of the inner wall of the separation cylinder: Rotation speed of the inner vortex in the separation cylinder: Resistance loss in the separation cylinder: Rotating vortex core and main airflow resistance loss: Total loss: ΔP = ΔP b +ΔP x in, is the flow rate of sand-laden air, R in is the average swirl radius at the inlet, A r is the total inner surface area of ​​the separator, and ρ is the air density.

5. The parametric design method for an aircraft engine particle separator according to claim 4, characterized in that: Step three uses genetic algorithm for optimization, as follows: Step 1: Define the variables as the particle separator cylinder diameter D and the main flow path diameter D x 、Separator total height H t , cone part height H c , tangential inlet height a, tangential inlet width b, the constraint condition is the given parameter optimization range, and the objective function f(x1,x2,x3…) is the total loss relationship; Step 2: Encode the variables appropriately to facilitate operation in the genetic algorithm. Represent the xi value in f(x1,x2,x3…) as a binary string and encode it into a binary gene string. Step 3: Heuristically generate an initial population of size N, where each individual is an xi value represented by a set of binary gene strings; Step 4: Decode the binary gene string of each individual into the corresponding xi, calculate the value of the objective function, and use it as the fitness function value score; Step 5: Select individuals according to the value of the fitness function, and use selection operations to select individuals with high fitness so that they can be repeated in the next generation; Step 6: In each generation, new individuals are generated through crossover and mutation operations, where the crossover operation is to combine the gene strings of two individuals to create a new individual, and the mutation operation is to change a gene in an individual to create a new individual; Step 7: Repeat steps 4 to 6. After multiple iterations, the best individual in the population represents the optimal solution of f(x1, x2, x3…); finally, decide when to stop iterating based on the convergence of the algorithm or other evaluation criteria.

Images