Design and Fabrication of a Magnetic Actuator for Torque and Force Control Estimated by the ANN/SA Algorithm

Heris, Pooriya Kazemzadeh; Khamesee, Mir Behrad

doi:10.3390/mi13020327

Cited by 9 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prior demonstrations indicate that artificial neural networks can be trained to outperform linear multipole electromagnet modeling relying on fundamental physics, particularly in cases in which iron cores demonstrate nonlinear behavior in the presence of varying applied magnetic fields from electromagnets [21]. Using arrays of electromagnets controlled by ANNs, researchers have positioned millimeter-scale neodymium magnet (NdFeB) disc agents in 2D using an eight-coil array [22], generated real-time predictions of motion dynamics on polymer-based soft magnetic manipulators [23], as well as guided helical microswimmers in 3D [24] and through uncharacterized biomimetic environments [25]. Magnetically guided wheeled robots have been controlled using neuro-fuzzy networks [26], and researchers have made significant progress in guiding endoscopy instruments via intelligent controllers [27][28][29].…”

Section: Introductionmentioning

confidence: 99%

A Control Interface for Autonomous Positioning of Magnetically Actuated Spheres Using an Artificial Neural Network

Huynh,

Mutawak,

et al. 2024

Robotics

View full text Add to dashboard Cite

Electromagnet arrays show significant potential in the untethered guidance of particles, devices, and eventually robots. However, complications in obtaining accurate models of electromagnetic fields pose challenges for precision control. Manipulation often requires the reduced-order modeling of physical systems, which may be computationally complex and may still not account for all possible system dynamics. Additionally, control schemes capable of being applied to electromagnet arrays of any configuration may significantly expand the usefulness of any control approach. In this study, we developed a data-driven approach to the magnetic control of a neodymium magnets (NdFeB magnetic sphere) using a simple, highly constrained magnetic actuation architecture. We developed and compared two regression-based schemes for controlling the NdFeB sphere in the workspace of a four-coil array of electromagnets. We obtained averaged submillimeter positional control (0.85 mm) of a NdFeB hard magnetic sphere in a 2D plane using a controller trained using a single-layer, five-input regression neural network with a single hidden layer.

show abstract

Section: Introductionmentioning

confidence: 99%

A Control Interface for Autonomous Positioning of Magnetically Actuated Spheres Using an Artificial Neural Network

Huynh,

Mutawak,

et al. 2024

Robotics

View full text Add to dashboard Cite

show abstract

“…Magnetic levitation is defined as the suspension of an object using only non-contact forces induced through interactions of magnetic fields. Magnetic levitation and manipulation have been used in several applications, such as drug delivery within the human body [1,2], energy harvesting techniques [3,4], and the micromanipulation of microrobots [5] amongst several others.…”

Section: Introductionmentioning

confidence: 99%

Experimental Implementation of a Magnetic Levitation System for Laser-Directed Energy Deposition via Powder Feeding Additive Manufacturing Applications

et al. 2023

Self Cite

View full text Add to dashboard Cite

Magnetic levitation and additive manufacturing (AM) are two fields of significant interest in academic research. The use of non-contact forces for magnetic levitation techniques provides opportunities for adoption within the AM environment. The key goal of this article is to experimentally validate the implementation of a magnetic levitation system for Laser-Directed Energy Deposition via Powder Feeding (LDED-PF) Additive Manufacturing applications. Through simulations (conducted in ANSYS Maxwell) and experimental implementation, the levitation system’s stability is tested under a variety of different conditions. The experimental implementation highlights the feasibility of a magnetic levitation system for LDED-PF applications. The levitation system developed is capable of the suspension of non-magnetic materials. The system is also able to maintain stable levitation for extended periods of time. The incorporation of the levitation system into the AM environment may result in an increased maneuverability of non-clamped structures for AM deposition operations.

show abstract

“…In this regard, various machine learning techniques, including sparse identification of nonlinear dynamical systems (SINDy) [35] and artificial neural network (ANN) [37], have been successfully applied to model the systems. However, the application of neural network algorithms in MEMS modeling poses several challenges for real-world scenarios.…”

Section: Introductionmentioning

confidence: 99%

Modeling of Magnetoelectric Microresonator Using Numerical Method and Simulated Annealing Algorithm

Sadeghi,

Bazrafkan,

Rutner

et al. 2023

Micromachines

View full text Add to dashboard Cite

A comprehensive understanding of the linear/nonlinear dynamic behavior of wireless microresonators is essential for micro-electromechanical systems (MEMS) design optimization. This study investigates the dynamic behaviour of a magnetoelectric (ME) microresonator, using a finite element method (FEM) and machine learning algorithm. First, the linear/nonlinear behaviour of a fabricated thin-film ME microactuator is assessed in both the time domain and frequency spectrum. Next, a data driven system identification (DDSI) procedure and simulated annealing (SA) method are implemented to reconstruct differential equations from measured datasets. The Duffing equation is employed to replicate the dynamic behavior of the ME microactuator. The Duffing coefficients such as mass, stiffness, damping, force amplitude, and excitation frequency are considered as input parameters. Meanwhile, the microactuator displacement is taken as the output parameter, which is measured experimentally via a laser Doppler vibrometer (LDV) device. To determine the optimal range and step size for input parameters, the sensitivity analysis is conducted using Latin hypercube sampling (LHS). The peak index matching (PIM) and correlation coefficient (CC) are considered assessment criteria for the objective function. The data-driven developed models are subsequently employed to reconstruct/predict mode shapes and the vibration amplitude over the time domain. The effect of driving signal nonlinearity and total harmonic distortion (THD) is explored experimentally under resonance and sub-resonance conditions. The vibration measurements reveal that as excitation levels increase, hysteresis variations become more noticeable, which may result in a higher prediction error in the Duffing array model. The verification test indicates that the first bending mode reconstructs reasonably with a prediction accuracy of about 92 percent. This proof-of-concept study demonstrates that the simulated annealing approach is a promising tool for modeling the dynamic behavior of MEMS systems, making it a strong candidate for real-world applications.

show abstract

Design and Fabrication of a Magnetic Actuator for Torque and Force Control Estimated by the ANN/SA Algorithm

Cited by 9 publications

References 38 publications

A Control Interface for Autonomous Positioning of Magnetically Actuated Spheres Using an Artificial Neural Network

A Control Interface for Autonomous Positioning of Magnetically Actuated Spheres Using an Artificial Neural Network

Experimental Implementation of a Magnetic Levitation System for Laser-Directed Energy Deposition via Powder Feeding Additive Manufacturing Applications

Modeling of Magnetoelectric Microresonator Using Numerical Method and Simulated Annealing Algorithm

Contact Info

Product

Resources

About