Part 1 — Exploring the Power of Physics-Based Digital Twins

From asset monitoring stations in remote locations to industrial sensors tracking equipment health, many modern systems rely on small, self-powered IoT devices operating far from human reach.

These devices are typically equipped with a solar panel, a battery, and a communication module. Their job is simple: collect data, stay powered, and reliably transmit information to the cloud, often for months or years without maintenance.

While their purpose seems straightforward, ensuring they operate continuously under changing environmental conditions is far more complex.

This article is Part 1 of a three‑part series on applying physics‑based digital twins to solar‑charged, battery‑powered IoT monitoring devices. In this opening part, we examine the operational challenges these devices face and why physics‑based modeling becomes essential for predicting availability, State of Health (SOH), and Remaining Useful Life (RUL). Part 2 will explore the modeling foundation built with GT‑SUITE and GT‑AutoLion. Part 3 will show how these models integrate into an IoT cloud environment for daily operational execution.

Across many industries, solar‑charged, battery‑powered IoT monitoring devices have become critical to large-scale, distributed measurement networks . Deployed in remote and often harsh environments, they operate as part of connected ecosystems where reliability is essential and maintenance is limited

Figure 1: IoT Monitoring Device (AI generated impression)

With small solar panels, lithium‑ion battery storage, and a 4G/5G communication subsystem, every device experiences a constant tug‑of‑war between energy harvested from the sun and energy consumed by electronics and communication bursts.

Over time, two operational questions dominate. The first concerns availability: will a device have sufficient State of Charge (SOC) over the next several days to remain operational during upcoming measurement windows or does it have to be put into an energy savings mode? The second concerns battery longevity: how quickly is the device’s battery degrading, and how long will it remain serviceable before capacity fade or resistance growth forces a replacement?

Telemetry data such as outside temperature, voltage, current, and other BMS states alone cannot provide the answers. It reflects the present but does not reveal the internal aging processes or the physics behind future performance. Simple heuristics or statistical regressions often work initially but begin to drift as seasons change, usage evolves, or environmental conditions differ from the datasets on which those heuristics were built.

Figure 2: The digital twin extending physical device data to decision making

This gap has led operators of distributed device networks to adopt physics‑based digital twins. Here the digital twin encapsulates a physics‑driven representation of the system and updates itself continuously with real telemetry. Rather than inferring behavior from patterns, it simulates behavior based on underlying physics of the system. For solar‑charged, battery‑powered IoT monitoring devices, the twin becomes an executable model that understands how irradiance, temperature, BMS constraints, and electronic load interact with the electrochemical processes inside the battery.

The modeling foundation begins with GT‑SUITE, which represents the full energy and thermal chain including solar module behavior, MPPT charging, BMS logic, load electronics, and heat transfer interactions. The battery is represented using GT‑AutoLion, a high‑fidelity electrochemical model calibrated to the specific lithium‑ion cell chemistry used in the device. The combination provides both system‑level realism and detailed insight into how the battery ages under real-world exposure.

Once validated, the model transitions into a digital‑twin‑oriented form that receives telemetry instead of simulating the entire environment. Temperature, SOC, current, and BMS data from the real device become direct inputs. This creates a continuous lineage: the same model used to validate the design becomes the engine that evaluates and predicts performance once the device is deployed.

When integrated into a modern IoT and digital‑twin cloud architecture, the twin becomes part of a daily computational workflow. Telemetry is stored, processed, fed into the twin, and the resulting availability predictions, SOH updates, and RUL forecasts are written back into the same databases powering dashboards, analytics, and automation logic. GT-Play operates as the execution layer that allows this to happen reliably and at scale.

Figure 3: Digital Twin Cloud Setup

Part 2 examines how the underlying models are constructed and how aging is captured accurately using GT‑SUITE and GT‑AutoLion.

To explore how physics-based digital twins can be applied to your systems, connect with our team to see how these models can be integrated into your workflow. Visit our Digital Twin solution pages to learn more about system-level modeling capabilities, and follow us on LinkedIn to stay updated on the latest developments in simulation-driven engineering.