Research in AI and sensor technology could make building upgrades smarter and more affordable
Energy-efficiency upgrades to existing buildings can be expensive, complex and time consuming. It is also difficult to measure the results from upgrades to certify that performance goals are met.
The high barrier to reducing energy usage is a key factor driving a team of researchers from the Hong Kong University of Science and Technology (HKUST), and the Massachusetts Institute of Technology (MIT) to look for better ways to make building upgrades easier, smarter, and more affordable.
The key to their research centers on the combination of local sensors and artificial intelligence, such as within smart HVAC Building Management Systems. One such example is an improvement in how these sensors are packaged and produced, enabling them to provide much more accurate air flow and energy sensing, at a much lower price when compared with today’s traditional solutions.
As Schneider Electric continues to make EcoStruxure™ a vital tool for IoT implementation across its market segments, this unified research initiative between academia and industry is crucial to the development of cutting-edge research on sensors, signal processing and analytics for building and transportation connectivity.
The team’s inspiration for such high-level technology might surprise you, because it’s likely sitting in your pocket – the sensors inside today’s smartphones. Certainly, this is advanced technology. But you’d be forgiven for wondering how a phone’s sensors could drive the future of energy-efficient buildings.
According to Professor Yi-Kuen Lee from HKUST, leader of one of the research teams focused on smart buildings, solid-state CMOS transistor production has grown to astronomical levels over the past few years. The result has been a reduction in price for each transistor to a mere billionth of one US dollar.
Thanks to this advancement in the fundamental chip technology, the burgeoning mobile sensor market has grown by about 200% per year since 2007.
Micro-electrical-mechanical-systems (MEMS) sensors are an important derivative of semiconductor devices, initially used in automotive applications (such as airbags), and later by telecommunications and consumer products like smartphones.
Now, they stand to change the course of IoT by increasing the intelligence of sensor solutions used in:
- Heating, ventilation and air conditioning (HVAC)
- Gas and water supply systems
- Temperature sensors
- Humidity sensors
- Air quality sensors
- Vibration sensors
- Early problem detection
- Variable air volume systems
Similar to Moore’s Law, selected MEMS technology will follow CMOS to undergo a 50% increase in transistor density every 18 months. Researchers from HKUST believe that their revamped MEMS sensor design based on CMOS is already far more sensitive and accurate than conventional airflow sensors and could be produced at a much lower cost.
Professor Lee said of their approach, “Research on the main sensors presents an opportunity to leverage smartphone sensors’ fabrication technology for smart buildings. Can this lead to smart buildings that are more affordable? With distributed sensors and artificial intelligence, can we make smart buildings even smarter? AI combined with the distributed sensors and actuators can result in smart HVAC control systems.”
Lee, went on to describe how his team’s sensor technology could – when combined with other complimentary IoT technology – lead to annual savings of about 20-27% for buildings.
The team is currently working on new sensor design using commercial CMOS MEMS technology. Though advanced in concept, these sensors have a great potential for mass production.
In summary, professor Lee stated, “This is a proof-of-concept application of low-cost air flow/energy sensors and actuators, including smart windows and personalized ventilation, for smart HVAC control systems. This integration will lead to significant energy savings, and we hope this technology will be implemented in more scenarios. With the collaborative effort of our team, we believe the technology will be ready for production in the near future.”