Accuracy of energy-use surveys in predicting rural mini-grid user consumption

Courtney Blodgett, Peter Dauenhauer, Henry Louie, Lauren Kickham

Research output: Contribution to journalArticlepeer-review

61 Citations (Scopus)
28 Downloads (Pure)


Mini-grids for rural electrification in developing countries are growing in popularity but are not yet widely deployed. A key barrier of mini-grid proliferation is the uncertainty in predicting customer electricity consumption, which adds financial risk. Energy-use surveys deployed in the pre-feasibility stage that capture present and aspirational consumption are intended to reduce this uncertainty. However, the general reliability and accuracy of these surveys has not been demonstrated. This research compares survey-predicted electrical energy use to actual measured consumption of customers of eight minigrids in rural Kenya. A follow-up audit compares the aspirational inventory of appliances to the realized inventory. The analysis shows that the ability to accurately estimate past consumption even in a relatively short time-horizon is prone to appreciable error—a mean absolute error of 426 Wh/day per customer on a mean consumption of 113 Wh/day per customer. An alternative data-driven proxy village approach, which uses average customer consumption from each mini-grid to predict consumption at other mini-grids, was more accurate and reduced the mean absolute error to 75 Wh/day per customer. Hourly load profiles were constructed to provide insight into potential causes of error and to suggest how the data provided in this work can be used in computer-aided mini-grid design programs.
Original languageEnglish
Pages (from-to)88-105
Number of pages18
JournalEnergy for Sustainable Development
Early online date6 Sept 2017
Publication statusPublished - 31 Dec 2017


  • micro-grids
  • mini-grids
  • rural electrification
  • energy estimation
  • energy-use survey
  • solar power


Dive into the research topics of 'Accuracy of energy-use surveys in predicting rural mini-grid user consumption'. Together they form a unique fingerprint.

Cite this