Monday, January 14, 2013

Future Analysis Techniques


I would define the future as being anything within the next 5 years. But ultimately it depends on what is being considered. In terms of analysis techniques, it takes a few years to thoroughly check algorithms for as many probable scenarios and to find and fix all malfunctions.

Recently, examination and improvement in the design of sustainable intelligent buildings has gained attention. One of the things I found interesting is the utilization of the collected data from the building operation management that can be used to analyze the real time building information for hazard detection. In the article, the authors installed an indoor wireless sensor network to collect the air temperature data. A machine-learning algorithm was developed to analyze the collected data and automatically generate early warning signals for hazards such as overheat/fire. The system was designed to detect all possible potential hazards while allowing as few false alarms. The experimental results validate the proposed machine-learning algorithm and the effectiveness of using a wireless sensor network for early detection of building hazards.

As discussed by my fellow group members, similar algorithms can be used in analyzing more dynamic modeling software to help integrate all the different electrical  and mechanical components in a building. Thus, allowing buildings to be more energy efficient.

No comments:

Post a Comment