Torque Transmission Modeling of Two Coaxial Electrorheological Clutches for Reciprocating Actuation
Shouren Huang, Masatoshi Ishikawa
AI summary
Problem
Conventional Bingham-based models simplify ER fluid viscosity to a constant, failing to capture complex electric-field-dependent nonlinear dynamics and hindering accurate real-time torque prediction for coaxial clutch systems.
Approach
The authors develop data-driven regression models that incorporate electric-field-dependent viscosity nonlinearity, alongside direct neural network estimators that bypass traditional rheological assumptions to predict transmission torque.
Key results
- HEM and RBFM models capture electric-field-dependent viscosity nonlinearity, surpassing constant-viscosity baselines
- Direct neural estimators (RBFN and FNN) accurately predict torque without relying on the Bingham model
- The FNN achieves superior accuracy with a minimal single hidden layer, enabling low-computational real-time implementation
- Real-time validation across diverse operating conditions confirms the robustness and feasibility of the proposed method
Why it matters
Enables precise, real-time control of high-speed reciprocating actuators for robotics and manufacturing by providing an accurate, computationally lightweight torque model for ER fluid clutches.
Abstract
This study focuses on modeling the transmission torque of two coaxial electrorheological (ER) fluid clutches through a data-driven approach. Instead of simplifying the viscosity term in the Bingham model to be a constant as shown in conventional methods, we propose the method of introducing electric field-dependent nonlinearity into the viscosity term to better capture the complex rheological behavior of ER fluids. Based on this framework, we developed a heuristic explicit model (HEM) and a radial basis function model (RBFM) that incorporate the mechanical characteristics of the coaxial clutch structure. Furthermore, we explored direct estimation methods using a radial basis function network (RBFN) and a feedforward neural network (FNN) without relying on the Bingham model. Comparative evaluations with traditional ER models validated the effectiveness of our nonlinear formulations. Notably, the FNN approach demonstrated superior accuracy even with a single hidden layer containing only a few neurons, making it well-suited for real-time implementation with minimal computational over- head. Real-time validation across diverse operating conditions further confirmed the feasibility and robustness of the FNN- based method. These findings contribute new insights into ER fluid applications.