Council on Energy, Environment and Water Integrated | International | Independent

What Smart Meters Can Tell UsInsights on Electricity Supply and Use in Mathura and Bareilly Households

Shalu Agrawal, Sunil Mani, Karthik Ganesan, Abhishek Jain
February 2020 | Power Sector

Suggested citation: Agrawal, Shalu, Sunil Mani, Karthik Ganesan, and Abhishek Jain. 2020.  What Smart Meters Can Tell Us? Insights on Electricity Supply and Use in Mathura and Bareilly Households. New Delhi: Council on Energy, Environment and Water.


This study provides insights into urban household electricity consumption patterns by collecting high-frequency data from nearly 100 smart meters installed in Mathura and Bareilly districts of Uttar Pradesh. It investigates the gaps in the quality of supply and discusses how power distribution companies can utilise the smart meter data for effective service delivery and demand management. Further, it discusses the developments in smart metering technology and how Indian power utilities are tapping into the opportunities offered by this technology. The study also outlines the policy implications of the insights derived and proposes recommendations. The data presented in the study was collected between May and October 2019.

Appliance penetration among the sample households varies across the two districts

Source: Author’s analysis

Key findings

  • Sampled households in Mathura and Bareilly received electricity supply for 22 hours on average. However, there was a variation in the duration and quality of electricity supply.
  • Households in notified town areas (NTAs) reported power outages of around 3.5 hours/day, with an average of 6 interruptions. Households in district headquarters had shorter (1.3 hours/day) and fewer interruptions (3.5 times/day).
  • Outages recorded at the household level were significantly high when compared with the outage duration reported by the discoms at the feeder-level (11 kV). This suggests the need of smart meters to monitor supply quality at the end-user level..
  • A few residential areas reported low voltage issues, mainly due to inadequate capacity, with voltages dropping by 25 to 30 per cent during the peak load.
  • Households using only fans for space cooling spend up to 200 units per month between May and October. Those with coolers and air conditioners used power in the range of 200–1000 units a month.
  • Richer households are more likely to own high-end appliances like air conditioners and/or coolers. Even among the households that own high-end appliances, hours of use and its contribution to peak demand tends to rise further with an increase in household income levels.
  • Using an analysis on consumption/current patterns at every three-minute interval, we find that majority of the households tend to underestimate their AC usage. This implies that people have limited information to optimise their energy usage.

Key recommendations

  • Leverage smart meter network to monitor the quality of power supply on a real time basis. This can then help in predictive maintenance and infrastructure planning.
  • Utilise data obtained from smart meters to track electricity consumption at the household level and identify consumers driving the peak demand. This can help the discoms in planning the load curve flattening interventions, and reducing the additional generation cost, which is currently required to meet the peak demand.
  • Use the data on energy consumption to engage with individual consumers to provide them with customised periodic feedback on self-consumption, promote conservative energy use, and implement demand response mechanisms in a targeted manner.
"Discoms can use smart meters for better service delivery, load management and to enable low carbon transition in India. For this, it is crucial that this intricate technology is deployed in a systematic manner." – Shalu Agrawal

Sign up for the latest on our pioneering research