Multi-objective optimization design of linear oscillating actuator based on hybrid surrogate model.

Area of Science:

• Engineering
• Mechanical Engineering
• Control Systems

Background:

• Thrust fluctuations from electromagnetic and cogging forces limit linear oscillating actuator (LOA) efficiency and performance.
• Optimizing LOAs presents a complex multi-objective challenge due to these limitations.

Purpose of the Study:

• To develop an innovative optimization strategy for enhancing LOA performance and controllability.
• To address the challenge of reducing force fluctuations in LOAs.

Main Methods:

• Utilized a hybrid surrogate model (HSM) integrating four data-driven models with a heuristic weighting scheme.
• Employed an improved nondominated sorting genetic algorithm II (NSGA-II) with a modified crowding degree model.
• Integrated empirical validation (static/dynamic tests) and finite element analysis for performance assessment.

Main Results:

• Achieved significant enhancements in average LOA performance.
• Demonstrated a notable reduction in force fluctuations, improving system controllability.
• The hybrid surrogate model (HSM) accurately predicted motor performance and facilitated rapid candidate solution generation.

Conclusions:

• The proposed optimization strategy, combining HSM and improved NSGA-II, offers a robust, data-driven framework for high-performance LOA design.
• The method effectively addresses complex multi-objective optimization problems in engineering.
• Empirical validation confirmed the efficacy of the proposed optimization approach for LOAs.