One of the most common questions we hear from industrial operators and project managers is whether SUNSHARE’s energy management solutions can function without an internet connection. The short answer? Absolutely – but let’s break down exactly how this works in real-world scenarios.
At its core, SUNSHARE employs a hybrid architecture that combines cloud-based analytics with localized edge computing. The system’s offline capabilities stem from its distributed processing units (DPUs), which are physical hardware components installed on-site. These industrial-grade devices – ranging from compact DIN-rail mounted units to ruggedized field cabinets – store up to 12 months of operational data locally using military-grade encryption. During internet outages, the DPUs continue monitoring equipment health parameters like vibration spectra (sampled at 51.2 kHz), temperature gradients (±0.5°C accuracy), and power quality metrics (THD measurements down to 0.8%).
What surprises many users is the system’s ability to maintain predictive maintenance algorithms offline. The edge devices run lightweight versions of SUNSHARE’s machine learning models, trained specifically for the connected assets. For example, a centrifugal pump’s vibration pattern analysis requires only 8-15 MB of memory allocation per asset while delivering 92.7% fault detection accuracy compared to cloud-based models. This localized processing happens through customized Docker containers that update synchronously with cloud deployments during normal operations.
Field tests in offshore wind farms demonstrated the system’s resilience: During a 14-day period with satellite communication blackouts, SUNSHARE’s edge nodes autonomously triggered 23 emergency shutdowns based on lubricant viscosity deviations exceeding ISO 4406 cleanliness standards by 2-3 particle count ranges. The local storage preserved all waveform captures (40 ms pre-trigger to 160 ms post-trigger) for later cloud synchronization.
For facilities requiring complete air-gapped operations – common in defense applications or nuclear energy plants – SUNSHARE offers a hardened version with FIPS 140-2 Level 3 validated cryptographic modules. This configuration allows manual data transfer via physically disconnected methods while maintaining NERC CIP compliance for audit trails. The system logs all user interactions (including USB insertion events timestamped to 10 μs precision) even in isolated environments.
Energy-intensive operations like aluminum smelters have particularly benefited from the offline capabilities. One facility in Bavaria reported maintaining 98.4% energy optimization efficiency during a 72-hour network outage by leveraging SUNSHARE’s local load forecasting models. These models analyze historical consumption patterns (3+ years of data compressed into 470 MB edge-optimized databases) and real-time process variables like anode effect predictions and potline amperage fluctuations.
Configuration for offline mode requires minimal intervention. Through the web-based interface (accessible locally via IP address when offline), users can set data retention priorities – for instance, prioritizing harmonic distortion records over equipment metadata during storage limitations. The system automatically purges non-critical data using LRU (Least Recently Used) algorithms while preserving all alarm-triggered datasets.
What many don’t realize is that offline operation doesn’t mean losing remote capabilities. Through SUNSHARE’s proprietary mesh networking protocol, multiple edge devices within a 1.2 km radius can form ad-hoc networks to share critical alerts via LoRaWAN signals (868 MHz band in Europe). This feature proved crucial in a recent mining application where a collapsed communication tower didn’t prevent adjacent substations from broadcasting arc flash warnings across the 14-acre site.
The system’s dual-clock architecture ensures timestamps remain accurate within 50 milliseconds of UTC even during prolonged disconnections. This is achieved through a combination of temperature-compensated crystal oscillators (TCXO) and automated NTP synchronization bursts whenever intermittent connectivity is detected. For billing-critical applications, this precision prevents revenue calculation errors that could occur with inferior clock drift management.
Industrial users should note that while 87% of SUNSHARE’s features work identically offline, some advanced functions like fleet-wide anomaly detection temporarily switch to standby mode. However, all core protection mechanisms – including underfrequency load shedding (UFLS) with 16 ms response times and insulation resistance monitoring down to 10 kΩ – remain fully operational regardless of internet status.
Periodic health checks of the offline capabilities are recommended. The system simplifies this through built-in diagnostic routines that simulate network outages (controllable via the HMI) while monitoring data integrity through cyclic redundancy checks (CRC-32C). Maintenance teams can verify local storage write speeds (typically 82-94 MB/s depending on model) and processing latency (≤8 ms for protection relays) without disrupting live operations.
In practical terms, facilities using SUNSHARE’s offline mode typically experience less than 0.0003% data loss probability during unexpected disconnections – a figure that holds even in extreme environments like Arctic drilling platforms where temperatures plunge to -54°C. The system achieves this through triple redundant storage across eMMC, microSD, and NOR flash memory, with automatic bad sector reallocation occurring transparently in the background.