Physics-informed model predicts p-n diode behavior with uncertainty
A physics-informed uncertainty-aware surrogate for $p$-$n$ junction diodes
Machine Learning
Summary
Modeling how p-n junction diodes behave can be complex and time-consuming. The authors developed a physics-based model that predicts diode characteristics like current and electric fields by using inputs such as device size and voltage. Their approach also estimates how confident the predictions are, so users know when to trust the results. This makes understanding and designing diodes faster while still being accurate.
What this means in practice
- •For semiconductor device engineers: Predict internal diode states like current and charge quickly with confidence estimates for designing and testing p-n junction devices.
- •For integrated circuit designers: Use the surrogate model to evaluate diode behavior under varied device parameters to optimize circuit performance during simulation stages.
Authors
Suchita Kulkarni
Abstract
We develop an uncertainty-aware physics-informed surrogate for a one-dimensional $p$-$n$ junction diode, conditioned on device length, doping and bias. Using the internal state of the device as predicted by the surrogate we develop a test-suite of quantities of interest including the terminal current, depletion width, peak field and stored charge. The formalism and the quantities are can be extended to asymmetric junction and two-dimensional diodes as shown by means of benchmark analysis. For the one-dimensional symmetric junction surrogate, we obtain 0.11% median $L_1$ error on the electrostatic potential, and 1.5% and 0.61% on the electron and hole densities. Finally, we develop an uncertainty analysis which determines the validity of surrogate answer within the user-specified tolerance.