SINDy + ANN for Magnetoelectric Sensors

Thesis SINDy + ANN Magnetoelectric sensors (Duffing Oscillator) R² SINDy: 0.986 (deriv) R² SINDy: 0.991 (signal) DNNS: 99.85–100%

What: Designed the excitation & acquisition pipeline (MATLAB + audio interface) **and** built an ML stack (SINDy + small ANN residual) to learn nonlinear EM-sensor dynamics.

Why: Prototype toward an implantable magnetoelectric sensor; set excitation frequency and boundary conditions to respect brain-field constraints and safety.

How: Discovered governing equations with SINDy; ANN residual captured leftover nonlinearity.

Results: Validated via energy consistency, stiffness (oscillator-type), and damping (phase portraits); interpretable model predicts the signal **and its derivatives**.

GitHub

Stiffness force — **SINDy** capturing the EM-sensor’s stiffness force, i.e., non-dimensional force-deflection characteristics.

Damping force — **SINDy** capturing the EM-sensor’s damping behavior (phase portraits, i.e., non-dimensional deflection (y1, U) vs. its derivative (y2, Udot)).