TL;DR
This paper tackles the challenge of accurately modeling electromechanical actuators (EMAs) for aircraft, specifically a trimmable horizontal stabilizer actuator (THSA). By employing an incremental modeling approach and parameter identification, they demonstrate how individual nonlinear effects like friction and backlash significantly impact system performance, providing a methodology for creating models with appropriate complexity for system engineering.
What Problem Does This Paper Address?
The aerospace industry is increasingly moving towards "more electric aircraft," replacing traditional hydraulic systems with electromechanical actuators (EMAs). While EMAs offer advantages in power management and integration, accurately modeling their complex nonlinear behavior is crucial for ensuring system performance, safety, and efficient design. This paper focuses on a trimmable horizontal stabilizer actuator (THSA), a less safety-critical component, to investigate how to build system models that capture these nonlinearities. The challenge lies in developing models with the right level of complexity – not too simple to miss critical effects, and not too complex to be computationally prohibitive. This requires a systematic way to introduce and evaluate individual nonlinear effects, such as friction, backlash, and compliance, which are prevalent in the mechanical transmission components of EMAs.
Methods
The core of this paper is its incremental modeling methodology. Instead of building a single, highly complex model from the outset, they propose a step-by-step approach:
- Multi-Level Modeling: They define four levels of model complexity, starting from a basic linear model (Level I) and progressively adding nonlinear effects like friction (Level II), friction and backlash (Level III), and finally a fully advanced model with complete nonlinear effects (Level IV). This allows for a systematic evaluation of each nonlinear effect's impact.
- Parameter Identification: For each nonlinear effect (friction, backlash, compliance), they employ parameter identification techniques. This involves using experimental data from a dedicated test rig to tune model parameters, ensuring the models accurately represent the physical behavior. The Particle Swarm Optimization (PSO) algorithm is used for this purpose.
- Focus on Key Nonlinearities: The paper specifically targets nonlinearities in the mechanical power transmission, namely:
- Friction: Modeled using a Stribeck friction model, accounting for velocity and load dependency.
- Backlash: Modeled in the no-back mechanism, which is crucial for locking the surface position.
- Compliance: Modeled in the mechanical load path, affecting stiffness and potentially causing oscillations.
Datasets
Not discussed in the paper. The paper relies on experimental data collected from a custom-built test rig for the THSA.
Key Insights and Ideas
- Incremental Modeling is Key: The paper strongly advocates for an incremental approach to modeling complex electromechanical systems. This allows engineers to understand the contribution of each nonlinear effect and select an appropriate model complexity for different engineering tasks (e.g., early design vs. detailed performance analysis).
- Friction Dominates Current Demand: The study reveals that no-back friction significantly impacts motor current demand, potentially tripling it under aiding load conditions. This highlights the importance of accurate friction modeling for power sizing.
- Backlash Causes Back-Driving: The presence of backlash in the no-back mechanism leads to a "back-driving" phenomenon where the aerodynamic load can move the actuator even when disabled. This is a critical finding for understanding actuator behavior during fault conditions or trim stops.
- Compliance Affects Speed Stability: Nonlinear compliance in the mechanical load path, particularly due to preload springs and backlash, can lead to significant speed oscillations and overshoots, impacting the precision of the actuator's movement.
- Symmetry and Unilaterality Matter: The paper notes that clearances in the no-back mechanism are unilateral and not perfectly symmetrical, leading to opposite variation trends under the same load direction, which is an important detail for accurate modeling.
Key Results
- Model Validation: The simulation results from models with increasing complexity (Levels II, III, and IV) are compared against experimental data, showing good agreement and validating the incremental modeling approach.
- Friction Impact: Model (b) (friction only) shows a significant difference in motor current compared to the basic model (a), with aiding load conditions requiring approximately three times more current. A speed perturbation of about 0.1s is also observed.
- Backlash Impact: Model (c) (friction and backlash) demonstrates a back-driving phenomenon when the motor is disabled, consistent with experimental results. This back-driven speed can reach up to 40% of the command speed amplitude.
- Compliance Impact: Model (d) (fully advanced) shows speed oscillations and an overshoot of around 30% of the command signal due to compliance effects in the mechanical load path, which is also observed experimentally.
Limitations and Future Work
- Model Complexity Trade-off: While the paper demonstrates the value of incremental modeling, the optimal complexity level for different applications (e.g., control design vs. system integration) still needs to be determined based on specific requirements.
- Parameter Identification Data: The accuracy of the models relies heavily on the quality and comprehensiveness of the experimental data used for parameter identification.
- Future Work: The authors suggest that this methodology can be extended to other electromechanical actuators and more complex systems. Further research could focus on integrating these models into a Model-Based Systems Engineering (MBSE) framework for more comprehensive aircraft system design and analysis. The paper also implicitly points to the need for robust control strategies that can mitigate the effects of these nonlinearities.
Summary updated