Sulaiman Mohaidat

Acoustic emission (AE) monitoring in metal additive manufacturing (AM) requires compact sensors capable of high-frequency operation, broad bandwidth, and high sensitivity. However, increasing structural stiffness to achieve high resonance frequencies typically reduces electromechanical sensitivity. This work presents a finite element study of a coupled-resonator capacitive micromachined ultrasonic transducer (CMUT) designed to address this trade-off. The proposed architecture integrates three mechanically coupled silicon membranes with a stepped capacitive cavity that increases capacitance while preserving structural stiffness, enabling enhanced sensitivity without compromising high-frequency operation. COMSOL Multiphysics simulations were used to evaluate modal characteristics and frequency response under DC pre-stressed conditions. Modal coupling produced closely spaced resonances that broadened the effective bandwidth, while the stepped cavity significantly increased voltage output through improved electromechanical coupling. Compared to a single-resonator flat-cavity design, the coupled stepped-cavity configuration demonstrated nearly a threefold enhancement in output voltage while maintaining operation near 100 kHz. Additionally, adjusting the central resonator length enabled controlled frequency tuning for scalable array implementation. These results establish a proof of concept for a high-frequency, high-sensitivity micro-electro-mechanical systems (MEMS) CMUT architecture suitable for distributed AE monitoring in advanced manufacturing environments.

Accurate estimation of lithium-ion battery State of Health (SoH) is critical for emerging applications such as reconfigurable battery systems. Although data-driven machine learning methods are promising, they often rely on costly, time-intensive aging experiments and extensive feature engineering. This work proposes a lightweight SoH-prediction framework validated on both physics-informed synthetic aging data and the NASA battery aging dataset. We evaluated Random Forest (RF) and Feedforward Neural Network (FNN) models that use only a limited number of samples from an early segment of the raw discharge voltage curve as input. Results show that RF consistently outperforms FNN across input sizes in deterministic or noise-free environments, achieving an RMSE of 0.07% SoH using just 5 voltage samples. In inherently stochastic experimental data, however, FNN can achieve an RMSE 50% lower than RF (1.28 vs. 2.87), but requires 37× more mathematical operations per inference. These findings emphasize the predictive value of the early-discharge-voltage region and demonstrate that compact, low-feature-complexity models can deliver accurate SoH estimates. Overall, the approach supports a goal of combining informed synthetic data with limited real measurements to build robust, scalable SoH predictors, reducing dependence on labor-intensive degradation testing and feature-heavy pipelines.