Energy Intelligence

One of our bets at BSI WGD is that people can make better decisions about energy use when they have ready access to usage information.  For many of us, that statement seems obvious, maybe too obvious to bother writing down. For people trained to interpret data as stories about actions and outcomes, cause and effect, simple data can be highly motivating.

There are two important leaps here: (1) data tells a story and (2) motivations aren’t too conflicted.  At BSI WGD, one of our goals is to create a platform for Web access to real-time energy use information that tells the story clearly.  The other is to put the data and the story of energy use into a context where motivations are clear.

This is why we aren’t in the residential energy monitoring business.  The motivations in this space aren’t clear. The utilities have a motive to sell energy at capacity, but not over.  So demand shedding during high-use periods isn’t related to conservation of energy, but rather, to controlling outage and capital costs (new power generation capabilities).  At other times, the utility does better by providing energy use incentives.

Another misalignment exists on the consumer side of residential energy monitoring. Users may be motivated to save energy in order to save the environment or to save money.  But the amount of energy used in an average household hardly justifies the cost of energy monitoring and “the smart grid” for an individual user. And users have been pretty clear that they don’t really want the wash cycle to stop in the middle because the utility needs a break.

A friend recently suggested that one clear commercial benefit of the smart grid will be smart, networked appliances sharing information about consumer behavior for targeted advertising–that seems like uncomplicated motivation.  Maybe an idea like this has motivated some large players to get in middle of smart appliance networks in the home even in the face of not payback on energy savings for the comsumer. Imagine a networked refrigerator with a bar code reading laser at the door.  Every item in your refrigerator, the number of times it is removed, now long it lasts, when you are running low, etc all in a database for King Soopers to mine for your next promotional coupon…

Energy Output from Solar Panel for a single day (somewhat cloudy)

At BSI WGD, we are working to provide direct results so consumers of energy for changing their behavior. In a few weeks, we will release BSI WGD beta 0.90 of the online application to our current customers. Among the capabilities: Real time energy use 5 min and daily averages Power Factor and other diagnostic information Peak use periods/devices mining User defined data roll up for any number and combination of monitored devices Compare use profiles over time

Where's the USB jack on this thing?

WattsGoingDown Building Systems Insight requires a low cost solution for putting energy measurements on the Internet. Our short-term plan is to use off-the-shelf hardware to create and deploy systems that send energy use data to our servers every minute or two.  A key driver of the software-as-a-service (SaaS) energy monitoring business model will be the balance of the savings realized by customers offset by the cost of the initial installation and WGD’s BSI’s monthly monitoring fees.

When the costs are justified purely by energy savings, monitoring system costing a few hundreds of dollars per monitored circuit only makes sense for higher-voltage, high current systems, or for circuits that are representative of many in the building or facility.

This is because, for a single circuit, the power used scales as voltage and current,  P= V x I . This is why business models centered on home energy monitoring are difficult to justify based on energy savings alone–customers have to be data geeks to make them work.  A typical home circuit runs at 120 volts and up to 20 amps.  This gives a maximum capacity of 2400 Watts per circuit (about two hair dryers).

Over a year, a home owner might run this circuit at capacity a quarter of the time (365 days x 6 hrs/day = 2200 hours).  At $0.10 per kWH, the total cost of running this example circuit is 2200 hours x 2400 Watts x 0.10 cents/kWH = $528.  If energy monitoring enables savings of 25% (aggressive!), you can pay no more than $132 per circuit for the hardware and monitoring to reach 1-year break even.

Without utility or government subsidy, the Tendril system costs quite a bit more. The Kill-A-Watt can be purchased for about $35 but provides no logging, network or analytics.  There is a clever hack, the tweet-a-watt, that adds a wireless Xbee module to the Kill-A-Watt, but you’ll need to be comfortable with soldering and analyzing the raw data yourself. The typical commercial network/analytic-enabled solutions run $750+ (e.g. WattNode + PLC).

So one challenge will be to create low-cost, dependable web-enabled measuring devices, devices so cheap that there will be no hesitation to install them and no motivation to remove them.