Optimization design method for wings of small unmanned aerial vehicle
By optimizing the wing shape of small UAVs using an improved frog-leap algorithm, the problem of lack of systematic optimization in existing technologies is solved, resulting in faster convergence speed and higher design accuracy, thus improving the core performance of UAVs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-15
- Publication Date
- 2026-04-10
AI Technical Summary
Existing small UAV wing optimization designs lack a systematic quantitative optimization framework, rely on engineers' experience, and are difficult to effectively improve core performance such as endurance, payload capacity, flight stability, and takeoff and landing adaptability.
An improved frog-jump algorithm is used to optimize the wing shape. By randomly generating a population, performing integral analysis, updating individual velocity and position based on the frog-jump principle, and adjusting inertia weight and acceleration factor, the global optimal design is achieved.
It improved the convergence speed and accuracy of wing shape optimization, obtained the globally optimal design scheme, and enhanced the core operational capabilities of UAVs.
Smart Images
Figure CN121835006A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to a small unmanned aerial vehicle wing shape optimization method, in particular to a small unmanned aerial vehicle wing shape optimization method using a frog leap algorithm. BACKGROUND
[0002] With the rapid development of unmanned aerial vehicle technology, small fixed-wing / compound-wing unmanned aerial vehicles have been widely applied to topographic mapping, power inspection, emergency rescue, agricultural plant protection, environmental monitoring and other civil and special fields due to the advantages of long endurance and long range of fixed-wing unmanned aerial vehicles and the flexible adaptability of vertical take-off and landing (VTOL) of compound-wing unmanned aerial vehicles. The core performance (such as endurance time, load capacity, flight stability, maneuvering response speed and take-off adaptability) of the unmanned aerial vehicle directly depends on the aerodynamic characteristics of the wing, and the shape parameters of the wing determine the core aerodynamic indexes such as lift-drag ratio, lift coefficient and stall characteristics. Limited by the cost and cycle requirements of small unmanned aerial vehicle design, processing and manufacturing, current wing shape optimization mainly relies on the experience selection of engineers (such as small modification based on existing mature airfoils), lacks a systematic quantitative optimization framework, and urgently needs a simple and scientific wing shape optimization method to solve the pain points of the prior art and improve the core operation ability of small unmanned aerial vehicles. SUMMARY
[0003] The technical problem of the application is to overcome the shortcomings of the prior art, and a small unmanned aerial vehicle wing shape optimization method is provided. The method uses the frog leap algorithm, fully utilizes the advantages of simple design method, easy implementation, fast convergence speed and strong universality, improves the convergence speed by improving the algorithm, and finally obtains a globally optimal design scheme.
[0004] The application adopts the following technical scheme: a small unmanned aerial vehicle wing optimization design method, including the following steps:
[0005] S1, taking wing shape parameters as design variables, randomly generating a population consisting of individuals in a variable space formed by the upper and lower boundaries of the design variables;
[0006] S2, performing aerodynamic analysis on the wing by using the integral method to obtain individual fitness;
[0007] S3, reordering the population according to the individual fitness, updating the individual speed and position according to the sorting, and grouping;
[0008] S4, updating the worst individual according to the frog leap principle;
[0009] S5, when the maximum number of iterations is reached, the updating is terminated, and the optimal solution is output as the wing optimization scheme, otherwise, the step S2 is performed.
[0010] The wing profile parameters are taken as design variables, and random generation is performed in the variable space surrounded by the upper and lower bounds of the design variables The population is composed of individuals, and the population is specifically as follows:
[0011] The design variables include airfoil, wingspan, chord length; for The population is composed of individuals, The number of individuals is represented by the number of individuals, and the position of individuals is a vector composed of design variables, is the number of design variables.
[0012] Step S2, the integral method is used to perform aerodynamic analysis on the wing, and the individual fitness is obtained, and the specific process is as follows:
[0013]
[0014]
[0015]
[0016] =
[0017] Wherein, and are the lift-drag ratio and the lift margin of the wing in the cruise condition, respectively, , are the performance requirement weights of the lift-drag ratio and the lift margin of the wing in the cruise condition, respectively; , , is the lift coefficient of the selected airfoil of the i-th individual at different angles of attack, is the zero-lift-drag coefficient of the selected airfoil of the i-th individual, , is the Oswald efficiency factor; is the wing lift coefficient, is the wing drag coefficient, is the wing lift, is the wing drag; is the air density, is the flight speed; is the wing area;
[0018] Through the above calculation, the wing lift at different angles of attack can be obtained , and the maximum lift is taken as , is the lift in the cruise state, wherein is the weight of the unmanned aerial vehicle, is the gravitational acceleration, and further , This represents the desired lift margin.
[0019] In step S3, the update strategy is as follows;
[0020]
[0021]
[0022] in, For inertial weights, , , The acceleration factor represents the corresponding part. , , for Random numbers between, , For speed control factors, The i-th individual with the best fitness during the population iteration process is selected. The individual with the best fitness in the subgroup. The individual with the best fitness in the population; Let be the individual position at iteration t. Let be the individual velocity at iteration t.
[0023] The grouping method uses the frog-jumping principle for grouping: set the number of subgroups to A, sort them according to fitness, and fill the individuals into different subgroups in turn.
[0024] The inertia weight w gradually decreases as the number of iterations increases, and its calculation method is as follows:
[0025]
[0026] in, and These are the maximum and minimum weights, respectively. The maximum number of iterations, This represents the current iteration number.
[0027] Step S4, updating the worst-fit individual according to the frog-jumping principle, includes the following steps:
[0028] (1) The individual with the worst fitness in each subgroup is counted as The update strategy is as follows:
[0029]
[0030] In the formula, for Random numbers between; The individual with the best fitness in the subgroup; t is the current iteration number;
[0031] (2) First, calculate the fitness of the new individual. If the worst-fit individual is found to be fit, then the update is valid; otherwise, the update strategy changes to:
[0032]
[0033] in, The individual with the best fitness in the population; This indicates the updated fitness of the individual. This represents the fitness of the individual with the best fitness in the subgroup;
[0034] Recalculate the fitness of the new individual; if If this is true, then the update of the worst-fit individual is valid. This represents the fitness of the individual with the best fitness in the population; otherwise, the update strategy changes to:
[0035]
[0036] in, , … For independence Random numbers between, , Independent scaling factors; , … Elements representing individuals in the current population.
[0037] A small unmanned aerial vehicle (UAV) wing optimization design system, comprising:
[0038] The initialization module is used to randomly generate variables within a variable space bounded by the upper and lower bounds of the wing shape parameters as design variables. A population composed of individuals;
[0039] The fitness calculation module is used to perform aerodynamic analysis on the wing using the integral method to obtain the individual fitness.
[0040] The individual update module is used to reorder the population according to individual fitness, update the individual speed and position according to the order, and group them; and update the worst-fit individuals according to the frog jumping principle.
[0041] The results output module terminates the update when the user reaches the maximum number of iterations, and outputs the optimal solution as the wing optimization scheme.
[0042] The advantages of this invention compared to the prior art are:
[0043] 1. This invention provides a new approach to wing shape optimization. Based on the advantages of the frog leap algorithm, the algorithm is improved to achieve faster convergence speed, make it easier to obtain the global optimal design, and effectively improve the accuracy of the optimal solution.
[0044] 2. This invention can optimize the shape of UAVs by multiple parameters, thereby improving design efficiency. Attached Figure Description
[0045] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation
[0046] The present invention will now be described in detail with reference to the accompanying drawings.
[0047] like Figure 1 As shown, this invention proposes a method for optimizing the wing shape of a small unmanned aerial vehicle (UAV), comprising the following steps:
[0048] (1) Using the wing shape parameters as design variables, establish constraints based on rotor characteristics; randomly generate within the variable space enclosed by the upper and lower bounds of the design variables. A population composed of individuals The individual's ID number; the individual's location. It is a vector consisting of design variables. To determine the number of design variables, take... , .
[0049] (2) The aerodynamic analysis of the wing was performed using the integral method to obtain the individual fitness. , and These are the wing lift-to-drag ratio and wing lift margin under cruise conditions, respectively. , The performance requirements weights for wing lift-to-drag ratio and wing lift margin under cruise conditions are respectively used. The individual fitness calculation method is as follows:
[0050]
[0051]
[0052] =
[0053] in, , , and The parameters can be obtained from a table. , This is the Oswald efficiency factor, with a value ranging from 0.7 to 0.8.
[0054] The above calculations yield the wing lift at different angles of attack. ,in, To maximize lift, For lift during cruise, To determine the desired lift margin, take .
[0055] The performance requirement weight is set to [value]. , .
[0056] (3) Reorder the population according to individual fitness, update individual velocity and position according to the order, and group them;
[0057] The update strategy is as follows;
[0058]
[0059]
[0060] in, The inertia weight is calculated as follows:
[0061]
[0062] and And set the maximum and minimum weights; To determine the maximum number of iterations, take... ; Current iteration number; , , The acceleration factor representing the corresponding part is taken as... , , Speed control factor , ; , , for Random numbers between; The i-th individual with the best fitness during the population iteration process is selected. The individual with the best fitness in the subgroup. It is the individual with the best fitness in the population.
[0063] The grouping method uses the frog-leap principle for grouping: setting the number of subgroups. Individuals are sorted by fitness and then placed into different subgroups in turn.
[0064] (4) The worst (smallest) fittest individual is updated according to the frog jumping principle. The update steps are as follows:
[0065] The individual with the worst fitness in each subgroup is counted as... Its fitness score is The update strategy is as follows:
[0066]
[0067] In the formula for Random numbers are generated between the given numbers, and the fitness of the new individual is calculated. If the update is successful, then the update is successful; otherwise, the update strategy changes to:
[0068]
[0069] Calculate the fitness of the new individual, if If the update is successful, then the update is successful; otherwise, the update strategy changes to:
[0070]
[0071] in, , to For independence Random numbers between, , to As an independent scaling factor, take .
[0072] (5) Termination condition judgment. If the maximum number of iterations is met, the update is terminated and the optimal solution is output as the wing optimization scheme; otherwise, proceed to step (2).
[0073] In summary, this invention proposes a novel method for optimizing the wings of small unmanned aerial vehicles (UAVs). It constructs a simple and effective fitness function and uses an improved frog-jumping approach combined with particle swarm optimization to update individual components, avoiding local optima and increasing the algorithm's global search capability. Furthermore, it improves the particle swarm update strategy by designing parameters such as inertia weights and acceleration factors that can vary with the number of iterations, enabling strong dynamic adjustment capabilities. This balances global search and local exploitation, thereby enhancing the algorithm's performance.
[0074] The above are merely specific steps of the present invention and do not constitute any limitation on the scope of protection of the present invention; it can be extended to the field of UAV operating condition optimization, and all technical solutions formed by equivalent transformation or equivalent substitution fall within the scope of protection of the present invention.
Claims
1. A method for optimizing the wing design of a small unmanned aerial vehicle (UAV), characterized in that, The steps include the following: S1. Using the wing shape parameters as design variables, randomly generate within the variable space bounded by the upper and lower boundaries of the design variables. A population composed of individuals; S2. The integral method is used to perform aerodynamic analysis on the wing to obtain individual fitness. S3. Reorder the population according to individual fitness, update the individual velocity and position based on the order, and group them accordingly; S4. Update the individual with the worst fitness according to the frog jumping principle; S5. When the maximum number of iterations is reached, terminate the update and output the optimal solution as the wing optimization scheme; otherwise, go to step S2.
2. The method for optimizing the wing design of a small unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step S1, using wing shape parameters as design variables, randomly generates within the variable space enclosed by the upper and lower bounds of the design variables. A population consisting of individuals is as follows: The design variables include airfoil, wingspan, and chord length; for A population composed of individuals The individual's ID number, the individual's location It is a vector composed of design variables. The number of design variables.
3. The method for optimizing the wing design of a small unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step S2, the aerodynamic analysis of the wing is performed using the integral method to obtain the individual fitness, as follows: ； ； ； = ； in, and These are the wing lift-to-drag ratio and wing lift margin under cruise conditions, respectively. , These are the performance requirement weights for wing lift-to-drag ratio and wing lift margin under cruise conditions, respectively. , , Let be the lift coefficient of the airfoil selected by the i-th individual at different angles of attack. Let be the zero-lift drag coefficient of the airfoil selected for the i-th individual. , Oswald efficiency factor; For wing lift coefficient, For wing drag coefficient, For wing lift, For wing drag; For air density, For flight speed; Wing area; The wing lift at different angles of attack can be obtained through the above calculations. Take the maximum lift as , For lift during cruise, ,in For the weight of the drone, For gravitational acceleration, we can obtain , This represents the desired lift margin.
4. The method for optimizing the wing design of a small unmanned aerial vehicle (UAV) according to claim 1, characterized in that, In step S3, the update strategy is as follows; ； ； in, For inertial weights, , , The acceleration factor represents the corresponding part. , , for Random numbers between, , For speed control factors, The i-th individual with the best fitness during the population iteration process is selected. The individual with the best fitness in the subgroup. The individual with the best fitness in the population; Let be the individual position at iteration t. Let be the individual velocity at iteration t.
5. The method for optimizing the wing design of a small unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The grouping method uses the frog-jumping principle for grouping: set the number of subgroups to A, sort them according to fitness, and fill the individuals into different subgroups in turn.
6. The method for optimizing the wing design of a small unmanned aerial vehicle according to claim 4, characterized in that, The inertia weight w gradually decreases as the number of iterations increases, and its calculation method is as follows: ； in, and These are the maximum and minimum weights, respectively. The maximum number of iterations, This represents the current iteration number.
7. The method for optimizing the wing design of a small unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step S4, updating the worst-fit individual according to the frog-jumping principle, includes the following steps: (1) The individual with the worst fitness in each subgroup is counted as The update strategy is as follows: ； In the formula, for Random numbers between; The individual with the best fitness in the subgroup; t is the current iteration number; (2) First, calculate the fitness of the new individual. If the worst-fit individual is found to be fit, then the update is valid; otherwise, the update strategy changes to: ； in, The individual with the best fitness in the population; This indicates the updated fitness of the individual. This represents the fitness of the individual with the best fitness in the subgroup; Recalculate the fitness of the new individual; if If this is true, then the update of the worst-fit individual is valid. This represents the fitness of the individual with the best fitness in the population; otherwise, the update strategy changes to: ； in, , … For independence Random numbers between, , Independent scaling factors; , … Elements representing individuals in the current population.
8. A small unmanned aerial vehicle (UAV) wing optimization design system, characterized in that, include: The initialization module is used to randomly generate variables within a variable space bounded by the upper and lower bounds of the wing shape parameters as design variables. A population composed of individuals; The fitness calculation module is used to perform aerodynamic analysis on the wing using the integral method to obtain the individual fitness. The individual update module is used to reorder the population according to individual fitness, update the individual velocity and position based on the order, and group them. The worst-fitting individual is updated according to the frog-jumping principle; The results output module terminates the update when the user reaches the maximum number of iterations, and outputs the optimal solution as the wing optimization scheme.