Key Takeaways
- Digital twins combined with edge AI can significantly reduce energy waste in smart buildings by managing “phantom load” efficiently.
- A recently developed Edge-Enabled Digital Twins (EEDT) system achieved up to 82% reduction in phantom power usage in tested environments.
- Implementation of the system can lead to annual savings exceeding £9,000 for 500 devices, aiding organizations in minimizing costs and environmental impact.
The Impact of Phantom Load
Phantom load refers to energy consumed by devices left in standby mode, which can account for a staggering 32% of a building’s energy expenditure. In office buildings and student housing, this figure is significant, often representing one-third of total electricity use. Recognizing this issue, Dr Ahmad Taha from the University of Glasgow is leading efforts to develop solutions to mitigate energy waste without hindering business operations.
Traditional methods of managing energy consumption have often been limited, particularly concerning decentralized device management. Many enterprises rely on basic metering, failing to address “always-on” assets, which leaves potential savings untapped.
Revolutionary Digital Tool Development
Dr Taha’s team is creating a digital tool that targets phantom load efficiently. This approach utilizes Edge-Enabled Digital Twins (EEDT) to create virtual replicas of real-world assets on a local server, employing AI for real-time insights. This local processing not only enhances performance by reducing latency but also addresses privacy concerns associated with monitoring device usage.
A key aspect of the EEDT is its shift from traditional binary and rule-based automation to a “fuzzy logic” approach, which incorporates degrees of truth. The technology pulls data from smart energy sensors and processes it using a framework of 27 optimized rules. Three crucial metrics are established for energy management:
- User Habit Score: Evaluates user behavior patterns.
- Device Activity Score: Measures the duration a device remains inactive.
- Confidence Score: Assesses data reliability for informed decision-making.
Based on these metrics, the system can efficiently decide whether to power down devices, notify users, or maintain current settings, thereby minimizing energy waste without disrupting user activities.
Operational Results and Financial Benefits
Initial deployment in a university research lab showed promising results, with power consumption reduced by approximately 40.14% per workspace. Notably, focus on phantom loads resulted in a staggering 82% reduction during idle times. Financial projections indicate that deploying this system across 500 devices could result in annual savings exceeding £9,000, based on current UK electricity prices.
Beyond immediate energy savings, Dr Taha emphasizes long-term benefits in asset lifecycle management. By minimizing energy use, organizations could delay the necessity of replacing older devices with newer, more energy-efficient models, ultimately contributing to cost reduction in a challenging economic landscape.
Future Implications and Scalability
The transition from passive energy monitoring to sophisticated edge AI-driven solutions represents a significant leap forward for smart buildings. This technology can seamlessly fit into various environments, including corporate offices and healthcare facilities, where energy management remains crucial.
Dr Taha advocates for a comprehensive approach to energy efficiency, citing the EEDT tool’s potential to enhance institutional strategies aimed at achieving net-zero goals. However, for broader applicability, it must evolve alongside legacy infrastructures and the diverse range of assets found in various environments.
There is an ongoing challenge to leverage existing data effectively. The future of smart buildings will depend on the integration of AI with edge assets to actively manage energy use, making them not just smart but genuinely efficient.
The content above is a summary. For more details, see the source article.
