Measurement is not (necessarily) controlling - part 2

A little over a year ago, I posted a short reflection on the difference between measurement and control, here.  This month, two things have made me return to that blog.

The first was the discovery in a shop of a drinking water bottle with a simple measuring device built into the cap.  The cap could be clicked through four positions, and a window displayed, successively, one dot, two dots, three dots and four dots.  The selling line was that by clicking the cap after every time you emptied the bottle, you could count the number of times you drank a bottle of water.  Maybe the designer couldn't count to four?  Maybe the company didn't think that the system could be fooled - a clever child fooling her/his parent's urging to increase the child's water consumption?  I won't rush out to buy one of these bottles!

But in a more serious vein, I found this article by Anthony Vickers of City Science in Exeter.  The article describes using a smart meter and recording the electrical power consumption of a house through a day.  As I wrote in my reflection, it is easy to spot the peaks in demand, and step changes as particular systems are started or stopped.

(In our house, the equipment on standby - radios, router, clocks, fridge and freezer - consume about 0.1 kilowatts per hour.  As soon as the central heating pump comes on, this doubles, and a few lights increase the base load when it is dark.)

But it was the next stage that I found fascinating; Anthony Vickers and his team superimposed the plot of consumption against time of day with the power that solar panels would produce.  Now we enter the possibility of control.  If your home is fitted with solar panels, don't have an electrically heated shower.  Use the solar power to heat the water.  There are peaks for two showers at about 7am below.  And running the dishwasher can be delayed from immediately after the meal to a time when there is more power being produced.

But, this is a home where the two occupants were out during the day.  What would the minute-by-minute power consumption look like if they were around all day, working from home?  Or at weekends?  Or if there were more than two people in the home?  What further decisions could be made by examining such plots?   Time for some interesting research, some examination of choices, and some Operational Research models!


Popular Posts