Tuesday, 27 January 2015

DECC Workshop: Specifying and Costing Monitoring Equipment for a Longitudinal Energy Study

Yesterday I attended a DECC workshop aiming to specify and estimate the costs of monitoring equipment for a longitudinal energy study. The scale of the study would be to use questionnaires and monitoring equipment to study energy use in 10,000s of homes. However, only £500 per household could be spent on hardware at such scale. It was concluded that this budget would not go very far beyond aggregate gas, electricity and water monitoring equipment.

In contrast to this, the LUKES project proposes aggregate monitoring of 10,000s of homes, while extensively monitoring up to 400 homes for up to 4 years. Although not the primary aim of the study, such a data set would have considerable impact for the energy disaggregation community. I was keen to point out that such uses have arisen due to the Household Energy Study, and it is important to take this data set as a case study when designing new surveys.

However, it is also important that lessons are learned from HES, such that the same mistakes are not repeated. In particular, I’d hope that a new data set would:

  • Collect both aggregate-level and circuit-level data as well as appliance level data
  • Specify and maintain each household’s metering hierarchy and appliance names using a consistent metadata schema, such as the NILM Metadata project
  • Collect aggregate data at a higher resolution than 2 minute energy readings. Ideally, I believe 1 second power data would be best trade-off between cost and frequency


  1. Thanks loads for the write up.

    Do you know if DECC are going to investigate whether they can afford to do appliance-by-appliance monitoring per homes within their £500 budget? Or have they already ruled that out?

    I'm sure that £500 wouldn't have gone far back in the days of HES; but today I would imagine that they should be able to assemble a fairly comprehensive measurement system for £500. I think the hardware for my dataset (UK-DALE) cost about £500 for House1 and that's recording 53 channels at 6 second resolution and mains voltage and current at 16 kHz. And, of course, some hardware manufacturers might be interested in collaborating with DECC. Intel, for example, are interested in deploying their Galileo platform (a £50 board, a little like an x86 Raspberry Pi) into IoT sensing applications. Oh, and of course if the survey is done in homes with smart meters already installed then the cost of measuring the aggregate utility consumption should be very low (just the cost of the consumer access device).

    I'd hazard a guess that there are ways to deploy pretty comprehensive appliance-by-appliance sampling hardware, even given the £500 per home budget.

    1. I think they want a complete solution for £500, rather than it covering just the hardware and requiring them to commission a project to build the required software. Do you know of any existing conplete solutions that are ready to be deployed at scale that could do appliance level monitoring at scale?

    2. The system we designed for UK-DALE works well (in fact, it hasn't fallen over in years). They'd need to design the backhaul infrastructure but that shouldn't be hard. or just leave the data on the local logger's hard disk. But they'd also have to flash the EPROM of a bunch of Nanodes (one per house) so that might put them off.

      Failing that, I think the EDF EcoManager off-the-shelf hardware (about £13 per IAM IIRC) can handle 14 channels and spits out data over its USB connection every minute. So just add a RPi & USB HDD and you're set.

      And I haven't looked closely at the market for a year or so. I bet there are some exciting new sensors out there. I certainly hope DECC are doing a thorough search of the available systems and not just brushing it off.

      For a project of this scale and size, surely they can afford a few grand to commission some backhaul software?