Conventional black-box models for electric vehicle (EV) battery State-of-Health (SOH) prediction achieve high accuracy but lack interpretability, limiting their practical deployment in Battery Management Systems (BMSs). To circumvent these limitations, this study proposes a novel Grey-Box Residual-Driven Framework (GBRDF) that synergizes Deep Symbolic Regression (DSR) with a residual-learning BiLSTM network with two contributions: (1) the DSR component derives explicit, interpretable mathematical expressions governing global degradation trajectories based on electrochemical features, and (2) the BiLSTM network models the residual errors to capture high-frequency nonlinearities and complex sequential dependencies not addressed by the symbolic baseline. By fusing the physics-informed transparency of DSR with the data-driven refinement of BiLSTM, the GBRDF significantly enhances forecasting precision. Experimental validation across four independent EV datasets shows that the GBRDF achieves the highest coefficient of determination (R2) of 0.982, and the lowest mean absolute error (MAE) of 0.1398 and root mean square error (RMSE) of 0.3176, significantly outperforming existing methods. Furthermore, the DSR-derived SOH equation shows that battery degradation is primarily driven by high voltage exposure and charging time, with mathematical transformations reflecting how degradation accelerates initially then slows, matching real-world aging patterns where voltage stress dominates over temperature and usage variations.
In this work, a novel fully integrated glucose sensor based on organic electrochemical transistor (OECT) has been proposed through feasible printing on a flexible substrate. The fully integrated glucose sensor incorporates: 1) a screen-printed OECT featuring poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) channel functionalized with a chitosan–titanium dioxide (CS-TiO₂) nanocomposite and glucose oxidase (GOx) enzyme to enable glucose detection; 2) a 3D-printed 3x3 hollow microneedle (HMN) platform with reservoir pad for minimally invasive sampling; and 3) a flexible compact wireless interfacing circuit for glucose analysis. Mechanical and electrical stability are investigated on the OECT device after 100 bending cycles and over a week of testing on a curved surface. With respect to glucose, the OECT showed a strong linear response in between 0.01 to 30 mM (r² = 0.993), a high sensitivity of 6.01 μA·mM⁻¹·cm⁻², and good selectivity against ascorbic acid and uric acid. Upon integration, the integrated glucose sensor achieved a rapid response time of under 20 s when tested on an agar gel-based skin model. The output currents were calibrated and then verified within the glucose range of 1–10 mM, with a total reading time of less than 120 s, demonstrating good accuracy. Overall, the implementation of OECT coupled with HMN offers a promising platform for a fully integrated glucose sensor, with great potential towards commercialization.