by Lutz Ehrig and Danilo Hollosi
The project “S4EeB” (Sounds for Energy Efficient Buildings), which has been running since October 2011, aims to optimize the performance of existing building management systems by taking into account a building’s occupancy rate using audio sensor networks as a new source of information. The overall goal of this demand driven approach is to reduce unnecessary consumption of energy for heating, ventilation, air conditioning, and lighting. In the course of the project the Fraunhofer Institute for Digital Media Technology IDMT, located in Ilmenau and Oldenburg in Germany, has been developing procedures and methods for analysing audio data in order to gain information about the occupancy rate of buildings, on the basis of which the energy consumption of a building can be optimized automatically.
A major proportion (35%) of electricity consumed in commercial and public office buildings is attributed to heating, ventilation, and air conditioning (HVAC). Together with street and commercial lighting this accounts for more than 60% of the electricity consumed in office buildings across the European Union [1]. With 50 million public buildings existing across Europe, this sector has a huge potential for improving energy efficiency.
While modern buildings and public spaces use sensing technologies on a broad scale, for instance motion detection, video surveillance, temperature measurements and gas detection, the potential of sound and noise has not yet been utilized for the purpose of building automation. Given the importance that humans attribute to sound and noise in their indoor and outdoor environments, sound based sensing seems an obvious choice to provide valuable information, such as estimates of the occupancy rate in buildings. Furthermore, audio sensors are unobtrusive and have a higher user acceptance compared to video surveillance systems, for example. Unlike cameras, acoustic monitoring does not require a free line of sight. Also, sound based systems require less computing power, which is appropriate given our goal of increased energy efficiency in buildings.
The S4EeB system consists of three main components that have already been, or will be, developed, as depicted in Figure 1: the audio system for sound and noise recording, the acoustic processing system for detecting acoustic events using a machine learning approach, and the management system for monitoring occupancy rates and controlling the building automation system.
Figure 1 system components for the S4EeB project
The audio system consists of microphones to be installed inside the respective building. Ideally, the acoustic sensors should be distributed evenly across the space. However, due to architectural or infrastructure conditions and limitations, this may not always be possible. Hence, microphone arrays will be used, allowing analysis of the spatial acoustics of the respective areas in order to localize sound sources.
All acoustic sensors are connected to an acoustic processing unit (APU). The APU combines the microphone signals of one area or a particular part of it. Based on suitable signal parameters extracted from the audio signals in the APU, sound source localization and sound source separation are performed and the occupancy rate of a building is determined. Initial experiments have shown that the occupancy rate can be estimated with high accuracy by using approaches for machine learning based acoustic event detection. Furthermore, the modular layout of the APU allows users to easily modify the system so that it is capable of detecting security related acoustic events, such as cries for help or breaking glass. Hence, using the S4EeB system will not be limited to the project’s objectives.
All the data and semantic information from the APU are collected and analyzed by the building management system optimizer, which is the interface to the “classic” building management system. Based on the building’s occupancy rate, its thermal characteristics, outside weather conditions, and other parameters, the optimal settings with respect to energy efficiency and user comfort will be determined. Thermal modelling of the building is done beforehand, providing the basis for the best strategy considering the building’s energy consumption rate and the interaction of the building management and automation system with the HVAC system.
The main contributions of Fraunhofer IDMT to the project are: sound recording, audio signal processing, and acoustic event detection. In particular, audio data captured is analysed by algorithms developed by Fraunhofer IDMT, allowing the building’s occupancy rate to be determined for the purpose of integrating this data into the building energy management system.
This three-year project is funded by the European Union, and its consortium comprises research institutes and industry partners from four European countries, who have long-standing experience in building control strategies, audiovisual applications, microelectronics and mechanical components as well as in consulting and dissemination of results. The project recently finished its first year of collaborative development of a prototype system and the corresponding components. In the first quarter of 2013, field tests will start at the S4EeB demo sites, namely Milano-Linate airport and two shopping malls in Spain, Principe Pio in Madrid and Maremagnum in Barcelona.
Links:
http://www.s4eeb.org
http://www.idmt.fraunhofer.de
Reference:
[1] P. Bertoldi and B. Atanasiu: “Electricity Consumption and Efficiency Trends in the Enlarged European Union”, European Commission, Institute for Environment and Sustainability, 2007.
Please contact:
Lutz Ehrig, Danilo Hollosi, Fraunhofer IDMT, Germany
E-mail:
