The rapid evolution of modern warfare toward unmanned and intelligent systems has positioned Unmanned Aerial Vehicles as core assets for battlefield situational awareness and precision strikes, owing to advantages such as zero personnel casualties and enhanced survivability. Concurrently, performance demands for long endurance and high reliability impose severe challenges on structural design. Statistical data indicate that the average accident rate for Unmanned Aerial Vehicles is approximately four times that of manned aircraft of similar class, with system-related faults accounting for up to 69.14% of incidents. Among mechanical failures, control surface-related issues rank second only to power system faults, manifesting as surface jamming, transmission looseness, and dynamic response instability. For instance, in November 2018, a U.S. Air Force MQ-9 “Reaper” Unmanned Aerial Vehicle experienced a left elevator jam during a training flight, leading to an uncontrolled crash. The control surface actuation system, as the end effector for Unmanned Aerial Vehicle attitude control, directly influences maneuverability and flight safety. Traditional design approaches often analyze components independently to shorten development cycles, but this simplification sacrifices accuracy, potentially resulting in insufficient structural safety margins and flight accidents. Thus, there is an urgent need for a structural analysis and optimization method that balances computational precision and efficiency for Unmanned Aerial Vehicle control surface actuation systems.
This paper addresses the fault diagnosis of control surface anomalies in Unmanned Aerial Vehicles by developing an integrated component model that includes the control surface rocker, linkage, actuator rocker, tail wing, and control surface. High-speed aerodynamic loads are incorporated into the model to achieve high-fidelity simulations. The analysis reveals failure locations and modes consistent with actual incidents, such as a maximum von Mises stress of 449.14 MPa at the junction of the control surface rocker support and lug, indicating a critical safety margin deficiency. Parametric modeling techniques are introduced to automate the simulation workflow via Python scripting, enabling efficient sensitivity analysis and optimization of geometric parameters. The optimized design reduces the overall maximum von Mises stress, maximum control surface displacement, and mass by 79.08%, 72.32%, and 8.98%, respectively, while increasing the safety margin from -0.02 to 3.68. This methodology offers significant engineering value for the structural design, verification, and optimization of Unmanned Aerial Vehicle control surface actuation systems, particularly for platforms like the JUYE UAV.
The control surface actuation system in Unmanned Aerial Vehicles typically employs a rigid mechanical transmission mechanism, comprising components such as the control surface rocker, linkage, and actuator rocker. Traditional analysis methods often isolate these components for individual checks, but this approach overlooks complex interactions and load transfer paths, leading to inaccuracies. To address this, an integrated finite element model is developed that includes simplified homogeneous representations of the tail wing and control surface, while preserving critical connection features. The model neglects internal structures like spars and ribs to focus on global load-bearing behavior, and connection bolts are replaced with equivalent constraints that allow relative rotation about common axes. The geometric model is depicted schematically, highlighting key components and their interactions.
Loads and boundary conditions are derived from extreme flight scenarios, such as symmetric maneuvers where the control surface is deflected to its maximum angle during a transition from level flight to climb at maximum velocity. Aerodynamic loads obtained from wind tunnel tests or computational fluid dynamics simulations are applied as chord-wise distributions, amplified by a factor of 1.25 to account for dynamic effects like control mechanism gaps and discrete gusts. The load distribution is fitted using polynomial functions and applied to the model surfaces. Boundary conditions include fixed constraints at the model ends and symmetry constraints on the symmetric plane, leveraging model symmetry to reduce computational cost.
Material properties are assigned based on actual selections, with aluminum alloy 7050-T7451 used for the rocker and linkage, and titanium alloy Ti6Al4V for the tail wing and control surface. Key material parameters are summarized in Table 1.
| Material | Elastic Modulus, $E$ (GPa) | Tensile Strength, $\sigma_b$ (MPa) | Yield Strength, $\sigma_s$ (MPa) | Poisson’s Ratio, $\nu$ | Density, $\rho$ (g/cm³) |
|---|---|---|---|---|---|
| 7050-T7451 | 71 | 510 | 440 | 0.33 | 2.83 |
| Ti6Al4V | 114 | 1080 | 862 | 0.33 | 4.43 |
Mesh generation employs hexahedral structural elements, with finer discretization in critical regions like the rocker and linkage. A mesh independence study ensures convergence, and the finite element model comprises thousands of elements, as detailed in Table 2.
| Component | Element Type | Number of Elements | Element Size (mm) |
|---|---|---|---|
| Control Surface Rocker | C3D8R | 14510 | 0.5 |
| Linkage | C3D8R | 22536 | 0.5 |
| Actuator Rocker | C3D8R | 13568 | 0.5 |
| Control Surface | C3D8R | 10932 | 10.0 |
| Tail Wing | C3D8R | 8536 | 25.0 |
Structural failure is assessed using the von Mises yield criterion, where the equivalent stress $\sigma_{\text{von}}$ is computed as:
$$\sigma_{\text{von}} = \sqrt{\frac{1}{2}\left[(\sigma_1 – \sigma_2)^2 + (\sigma_2 – \sigma_3)^2 + (\sigma_3 – \sigma_1)^2\right]}$$
Here, $\sigma_1$, $\sigma_2$, and $\sigma_3$ are the principal stresses. Failure occurs when $\sigma_{\text{von}} > \sigma_s$, with $\sigma_s$ being the material yield strength.
Parametric modeling is implemented to automate the finite element analysis process, which typically involves pre-processing, solving, and post-processing stages. Pre-processing and post-processing consume over 95% of the time, hindering design efficiency. By leveraging Abaqus’s Python script interface, the workflow is encapsulated into functions that drive model creation, adjustment, and result extraction. For example, complex geometries like NACA0012 airfoils are generated using mathematical functions, and component actuation is handled through kinematic solutions. Pseudocode illustrates the algorithms for model creation and adjustment, enabling rapid parameter-driven analysis.
In a case study involving a JUYE UAV, an anomaly occurred during high-speed flight where pitch control failed, leading to an uncontrolled climb. Post-flight inspection revealed fracture at the control surface rocker support-lug junction. The integrated model was subjected to极限 loads, and results showed a maximum von Mises stress of 449.14 MPa at the critical location, with a safety margin of -0.02. The stress concentration aligned with the actual damage, confirming the fault mechanism. Further analysis indicated low-cycle fatigue under alternating stresses due to high-frequency vibrations from control gaps and gust loads, culminating in structural failure and control surface misalignment.
For optimization, design variables are defined as $X = (d_1, d_2, t_1, t_2, t_3, t_4, h_1, h_2, \theta_1, \theta_2)$, representing geometric parameters such as distances, thicknesses, heights, and angles. The objective functions aim to minimize the maximum von Mises stress $\sigma_{\text{von-max}}$, maximum control surface displacement $u_{\text{max}}$, and mass $m$:
$$\begin{aligned}
&\min f_1(X) = \sigma_{\text{von-max}} \\
&\min f_2(X) = u_{\text{max}} \\
&\min f_3(X) = m \\
&\text{subject to } X \in [X_L, X_U]
\end{aligned}$$
Here, $\sigma_{\text{von-max}} = \max(\sigma_{\text{von-a}}, \sigma_{\text{von-r}}, \sigma_{\text{von-w}})$, where subscripts denote the actuator rocker, linkage, and control surface rocker, respectively.
Sensitivity analysis employs orthogonal experimental design to explore parameter effects. Figures and tables summarize the relationships between design variables and responses. Key findings include:
- Parameters $t_2$, $h_1$, and $h_2$ significantly influence $\sigma_{\text{von-max}}$, with stress concentrations mitigated by avoiding hollow supports or adding stiffeners.
- $u_{\text{max}}$ correlates strongly with $\sigma_{\text{von-max}}$, especially near yield strength, and is most affected by $t_2$ and $\theta_2$.
- Mass $m$ is reduced by minimizing $h_1$, which also benefits strength and stiffness.
Optimization adjustments include increasing $t_2$ from 2.00 mm to 3.00 mm, decreasing $t_3$ and $t_4$ to 4.00 mm and 4.50 mm, reducing $h_1$ to 28.00 mm, and setting $h_2$ to 9.50 mm. Results show $\sigma_{\text{von-max}} = 93.95$ MPa, $u_{\text{max}} = 1.32$ mm, and $m = 26.86$ g, representing improvements of 79.08%, 72.32%, and 8.98%, respectively. The safety margin rises to 3.68, ensuring robust performance for the JUYE UAV and similar Unmanned Aerial Vehicles.
This study presents a comprehensive methodology for structural analysis and optimization of Unmanned Aerial Vehicle control surface actuation systems. The integrated modeling approach accurately captures failure mechanisms, while parametric automation enhances efficiency. Sensitivity insights guide design improvements, and optimization results demonstrate substantial gains in strength, stiffness, and lightweight performance. The framework is adaptable to other Unmanned Aerial Vehicle components, such as wings and landing gear, and can be extended with advanced algorithms for further refinement. Future work will focus on software encapsulation and broader applications to advance Unmanned Aerial Vehicle design capabilities.