The field of distributed energy is humming with a lot of new buzzwords for 2005—mesh networks, smart dust, MEMS (microelectromechanical devices), and peel and stick. What’s all the buzz about?
It’s nothing less than a wireless sensor and control revolution. And it’s catching on in a wide variety of distributed energy applications. From power generation components, to building automation, to wind site testing equipment, wireless monitoring and remote control is the next wave in power and efficiency.
Though the philosophy is to make them small (and cheap) wireless sensors are a key component of a very big idea—the pervasive Internet, where virtually any device can be embedded with tiny sensors. Obviously, the potential for applications is staggering. Moreover, it’s growing fast. The financial market for wireless monitoring and control is expected to rise to more than $750 million in sales by 2006, according to market research firm Venture Development Corp.
Some technologies are still in the development stage, but others are already in the field. As in the case of remote monitoring for standby power systems, where sensors have assumed a key role in ensuring power reliability.
For example, Critical Wireless produces modular remote terminal units that monitor up to 14 discreet signals for tasks such as operating status (running, auto/manual), fluid levels and fuel tank breach, UPS failure, room temperature, starting battery voltage, and intrusion alarms. The system is ideally suited to hurricane-prone Polk County, FL, according to Bob Stanton, director of fleet management for the county.
Polk has more than 100 backup generators, with 35 of them located in critical need facilities that include county administration, jails, hospitals, emergency communication centers, and shelters. “If we were to lose electricity due to a utility wire going down and we couldn’t start a generator, we’d have a serious problem,” says Stanton. “The system also allows us to allocate maintenance crews if we had a countywide problem and needed to prioritize our resources.”
Even without an emergency, Stanton expects to see benefits in maintenance cost reductions. Polk County has a mobile fleet of 3,000 units, including cars, plus heavy equipment. Eliminating routine service visits to backup generators means higher efficiency and lower costs.
Stanton stays on top of the system with visits to his Web site. When problems develop he gets alerts by e-mail, which are also routed to the county’s facilities department, as well as an outside contractor responsible for certain repairs and maintenance.
Critical’s system offers simultaneous alert delivery to pagers, text-capable cell phones and e-mail. A web-based device management portal lets administrators create alert lists, check unit status and analyze performance data. The data can be used to document generator service/exercise intervals and run time reports to avoid fines from air quality regulatory agencies, or as proof of peak shaving runtimes for utility companies.
“Sensors are getting smaller and easier to use,” says Charles Christie, vice president for marketing and business development at Critical. “You don’t have to run wires, so you save on the installation costs, and it’s possible to spread the sensors out to set up an ad hoc network to communicate with itself.” In most cases, hardware, installation labor, and one year of remote monitoring service runs about $1,500. Monitoring service includes mobile alert messaging as well as unlimited access to the M2Web Portal for device configuration, alert recipient list configuration, remote control, and data reporting.
The benefits of a low-cost approach and potential for savings haven’t been lost on the DOE. The agency directed major support to wireless networks and recently awarded $18 million to industry heavyweights General Electric, Eaton, and Honeywell to develop low-cost wireless sensors to reduce electrical loads from industrial motors.
According to DOE surveys, industrial motors (not including facility heating and ventilation) consumed 679 billion kWh in the US in 2003. The demand accounted for 63% of all electricity used in industry and 23% of the electricity sold in the US. Analysts estimate that a 10% to 20% reduction would save 35.1 billion kWh per year. The DOE doesn’t stipulate reducing demand on distributed energy, but as described later in this article, the areas of research and technology have much in common with both distributed and centralized power generation.
At General Electric, researchers are focusing on motor monitoring with wireless mesh networks. Mesh networks differ from single-node networks, such as those used by cell phones. Instead of communicating to a central node, mesh networks allow each sensor to function as a node, as well. The result is a more robust network that provides multiple paths for data.
Under harsh industrial conditions—such as a factory with many motors—a mesh network can survive damage by reconfiguring itself. These sensors collect data on motor performance that can be used for predictive maintenance. Information on wear, stress, and operation status allow for flexible repair scheduling, rather than at point of failure.
In September 2004, a competitor to GE, wireless mesh provider Sensicast, announced an environmental monitoring system that could span multiple buildings. The company is targeting sectors of high potential for distributed energy, such as data centers, hospitals and grocery warehouses.
At Eaton, the goal is data collection from electrical switching equipment. Switches are low cost points for monitoring voltage, current, power, load, and other key process information. Analysis of electricity usage allows for some creative process scheduling, says Jose Gutierrez, principal engineer at Eaton.
“With diagnostics and prognostics you can see consumption of energy clusters in a plant and alter schedules to control peak energy demand,” Gutierrez explains. “Then you don’t need to generate so much electricity to deal with demand fluctuations. Sensor feedback could also go to the generation source for use in peak shaving plans.”
Honeywell is targeting much of its efforts toward a problem area of petrochemical manufacturing—steam traps. The traps drain condensate from steam lines and often leak steam when damaged. Lost steam can increase energy consumption and cause temperature drops that create problems throughout the manufacturing process.
No matter the sensor’s location, expect the sensor to be small. Crossbow Technology currently markets a line of smart dust (about the size of a quarter) wireless sensors. Areas of applications include environmental, power, and structural monitoring. So far, temperature monitoring is the most popular use across all industries, according to Crossbow’s CEO, Mike Horton. However, Crossbow conducts training seminars for application developers, and Horton expects third-party integrators to hit the market with a variety of solutions.
“If you look at the applications we’ve got so far you’ll see environmental outdoor monitoring, security and tracking, health and wellness, power monitoring, and inventory and location,” Horton explains. “The last piece is packaged solutions. “Today you don't see any complete turnkey solutions other than temperature environmental monitoring, but shortly you'll see much more vertically oriented solutions from us and developers for specific applications.”
Horton views cost and user-friendly software as more important than size, but researchers at the University of California, Berkeley, take the word “dust” literally. They have developed a wireless sensor just 5 square millimeters in size (comparable to a fleck of glitter). The sensors could eventually cost less than $1 per unit. Both Crossbow and Berkeley support the TinyOS (operating system), an open-source (royalty-free) modular embedded software platform for building reliable wireless mesh networks. There are over 500 active groups using TinyOS.
Whether they’re quarter-sized or like dust in the wind, manufacturing isn’t the only place the DOE is pushing for wireless sensors. At the department’s Pacific Northwest National Laboratory (PNNL), researchers recently developed proposals for systems to deal with HVAC energy usage. Considering the fact that there are 4.7 million office buildings in the US, the potential for savings is huge, and the technology is transferable. “We’re working to build a wireless monitoring and controlling system for small commercial buildings,” explains Srinivas Katipamula, a PNNL project manager. “It’s a general-purpose monitoring and control system and you could also apply it to CHP systems.”
Katipamula recently demonstrated a wireless network of 32 sensors dispersed throughout a 70,000-square-foot office building. The results were quite impressive. The network revealed a number of faulty variable air volume (VAV) boxes that were causing uneven air distribution. Correcting the problem allowed a reduction of the supply air temperature by 2°F during cooling periods.
Additionally, the sensors provided data for implementing a chilled-water cost reduction strategy. Rather than keeping the water’s temperature at a fixed setting of 45°F, the engineers were able to set a range of 45°F to 55°F. Total cost of the hardware and labor came in at $9,390. Energy savings allowed for a payback within eight months. Future reduction measures call for wireless sensors to detect unoccupied space (presence sensing), monitoring of emergency lighting, and rollup door sensors to suspend air conditioning during loading operations.
Katipamula has developed another wireless sensor and controller system for rooftop-mounted HVAC units. “Most of the units are not connected to central systems, and wireless is more economical when the typical 10,000-square-foot building has four or five units on a rooftop,” he notes. Having the units networked to a central system would make it easy for utilities to implement “load shedding” to temporarily reduce demand when the grid is stressed. Many utilities offer load shedding programs for residential customers and Carrier Corp. developed its programmable ComfortChoice thermostat to work with such programs. ComfortChoice uses off-the-shelf wireless—Skytel two-way pagers. If wireless sensors and controls can optimize HVAC systems for load shedding, can they do the same to manage distributed energy solutions? As a matter of fact, they can, and PNNL partnered with the Bonneville Power Administration—a federal agency under the DOE, headquartered in Portland, OR, and serving the Pacific Northwest—in a demonstration to prove the advantages of supplying onsite energy with a microturbine.
Bonneville Power sees distributed energy as a way to avoid new transmission line construction, where expenses can run upward of $100 million. The project’s 30-kW Capstone microturbine is currently hard-wired to the Internet and operated from a Web site, but Katipamula expects to see wireless applications of the technology. “There are constraints on certain geographical locations that BPA serves and it’s faster than getting permits for construction,” says Katipamula. It’s possible to visit the projects Web site and see a record of the microturbine’s performance. https://bpanws.pnl.gov/mt/frames.asp?location=apel
Wireless applications are getting broad support from utilities and state agencies like the BPA. Some areas include testing wireless sensors in the chain of power delivery, wind generation site data collection, and transformer monitoring.
The New York State Energy Research and Development Authority (NYSERDA) has funded projects to achieve its goal to develop sensor concepts that exploit MEMS, and wireless technologies for other energy and environmental applications. The first project is in progress at Columbia University’s Earth Institute, where Prof. Vijay Modi is directing the development of a wind speed sensor for micrositing studies with wind-based systems. The project’s ultimate objective is to combine sensing and wireless transmission on a single, application-specific integrated circuit chip. Such a configuration would keep power consumption low enough to operate for years.
In New Jersey, the New Jersey Institute of Technology partnered with Public Service Electric & Gas on research with MEMS-based acoustic sensors to monitor transformers. Their ultimate goal is to transfer the technologies to other failure-prone links in the transmission chain—cables and power lines.
We can expect the trend to accelerate in the future, according to the Engineering Research Center for Wireless Integrated MicroSystems (WIMS ERC) in Michigan. States are aggressively competing for the business that wireless technology provides. The stakes are high for the WIMS partnership, which combines Michigan's university programs with funding from the National Science Foundation, plus additional contributions from the State of Michigan, federal agencies, and a consortium of some 20 companies. Their goal is to merge micropower circuits, wireless interfaces and sensors into microsystems that will have a pervasive impact on society. So the revolution has clearly begun, and it’s already reshaping the nature of distributed energy applications.