Improving data center energy efficiency using a cyber-physical systems approach: integration of building information modeling and wireless sensor networks

The increase in data center operating costs is driving innovation to improve their energy efficiency. Previous research has investigated computational and physical control intervention strategies to alleviate the competition between energy consumption and thermal performance in data center operation. This study contributes to the body of knowledge by proposing a cyber-physical systems (CPS) approach to innovatively integrate building information modeling (BIM) and wireless sensor networks (WSN). In the proposed framework, wireless sensors are deployed strategically to monitor thermal performance parameters in response to runtime server load distribution. Sensor data are collected and contextualized in reference to the building information model that captures the geometric and functional characteristics of the data center, which will be used as inputs of continuous simulations aiming to predict real-time thermal performance of server working environment. Comparing the simulation results against historical performance data via machine learning and data mining, facility managers can quickly pinpoint thermal hot zones and actuate intervention procedures to improve energy efficiency. This BIM-WSN integration also facilitates smarter power management by capping runtime power demand within peak power capacity of data centers and alerting power outage emergencies. This paper lays out the BIM-WSN integration framework, explains the working mechanism, and discusses the feasibility of implementation in future work. (C) 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license.