Sunday, January 12, 2014

Week 2 - The Future of Sensor Capabilities

Sensors are a key aspect of intelligent building systems, providing the measurements and data on which system responses are based. Though most modern buildings already utilize sensors to some degree (typically in thermostats, smoke, CO, and CO2 detectors, and utility meters, but humidity, occupancy, and light sensors are becoming more commonplace), William Healy of the National Institute of Standards and Technology notes that the response of the building sector to technological advances is generally slow. As an example, he cites the automotive industry: three decades ago, automobiles contained five sensors on average, while a typical new vehicle has forty to fifty. In comparison, the number and type of sensors in standard commercial buildings has changed relatively little over the same period of time. This is partly due to the fact that sensor systems have a high initial cost and are often difficult to install, meaning that the return on investment period is too long to be cost-effective for the average commercial structure (Castilloux). The growth of intelligent building practices, then, relies on sensor technology becoming inexpensive, noninvasive, and easy to install and future advances will likely focus on achieving these objectives.

One area that has seen a lot of recent development is wireless sensor technology, which eliminates the cost and intrusion associated with wiring. However, despite being faster and easier to install, the initial cost of the wireless components themselves is still prohibitive. Two other concerns are reliability related to signal attenuation over larger distances and security issues with transmitting data (Healy). These and other challenges such as accuracy and size reduction will be key points to address as sensor technology evolves.

For sensor installation to be truly cost-effective, not only initial expenses but also maintenance and operation costs must be lowered. With sensor networks of any significant size, energy consumption is increasingly an issue, creating a push to develop components that have the ability to harvest energy from the surrounding environment (Castilloux). Photoelectric, thermoelectric, and vibrational energy "scavenging" technologies have potential for integration into intelligent building sensors. Similarly, in order to reduce maintenance costs, self-diagnostic and self-calibrating sensors that can identify component damage and automatically correct for sensor "drift" over time are a major "wish list" item for the future (Healy).

Other students addressed advances in computer hardware technology in their blog posts. One interesting topic that was mentioned is the development of smart computer chips that mimic neural processing to "learn" better decision-making. This technology has some exciting implications for implementation in intelligent sensors that are capable of processing data prior to transmission to provide a higher quality of information (e.g., distinguishing noise and signal or performing biometric recognition).

Sources

Castilloux, Ryan. "Controls, Sensors Use in Building Energy Management Systems Forecast to Grow." Green Manufacturer. Web. 24 July 2012. <http://www.greenmanufacturer.net/article/facilities/controls-sensors-use-in-building-energy-management-systems-forecast-to-grow>


Healy, William. "Building Sensors and Energy Monitoring Systems." Building America Meeting on Diagnostic Measurement and Performance Feedback for Residential Space Conditioning Equipment. Web. 26 April 2010. <http://www1.eere.energy.gov/buildings/residential/pdfs/5_building_sensors_healy.pdf>.

2 comments:

  1. It would be very interesting for the wireless sensors you mentioned to be connected to the "smart" computer system that I wrote about via a cloud system. One smart system that learned trends and monitored several buildings via their wireless sensors could learn and optimize buildings based on how the other buildings in its system acted. It would be really cool to see one "brain" taking care of all the buildings on a campus and working to make them as efficient and cost effective as possible. Of course, like Mike said in my post, that would probably eventually lead to a robot revolution where everyone got locked out of the buildings...

    ReplyDelete
  2. I agree with everything you've mentioned in this post. I think the next step would be to look at if these sensors are performing a task or if they will be collecting information. If it is performing a task, what is the cost benefit rather than the traditional method currently used. If it is collecting information, what is the information being used for and how/who is analyzing it.

    ReplyDelete