A Sensing-Aware Simulation-Based Digital Twin Framework for Firmware-Level Validation of Photovoltaic MPPT Controllers

Abstract

As the photovoltaic (PV) generation sector expands, ensuring reliable maximum power point tracking (MPPT) in embedded controllers becomes increasingly important. However, many MPPT studies rely on idealized simulation assumptions that neglect practical sensing limitations, including ADC quantization, finite measurement resolution, sensor noise, offset, ripple, and scaling constraints. This paper presents a sensing-aware simulation-based digital twin framework for photovoltaic arrays that integrates dynamic environmental excitation with a virtual instrumentation layer and firmware-level execution on an ESP32 microcontroller (Espressif Systems, Shanghai, China). Unlike conventional model-in-the-loop simulations, the proposed framework ensures that the embedded control algorithm operates on reconstructed and quantized measurements rather than on ideal internal model states. The platform is evaluated using an Incremental Conductance MPPT implementation as a representative embedded workload. The results demonstrate that the proposed simulation-based digital twin framework enables repeatable firmware-level experimentation while exposing instrumentation-induced effects on tracking stability, control dynamics, and measurement-driven behavior under non-ideal sensing conditions. Furthermore, it provides structured datasets to support early-stage development, tuning, and validation of embedded MPPT algorithms before physical laboratory deployment.