Navigating India’s Changing Energy Demand with Renewable Energy Generation

“India’s transition to renewable energy generation is crucial to meet the growing demand for electricity while mitigating climate change. Understanding the complex interplay between demand and generation is key to achieving a sustainable energy future.”

India is anticipating significant changes in its electricity demand, both in terms of total demand and daily demand patterns, over the model period. To account for this, the country has designed a system of 35 time-slices to capture changes in seasonal and daily demand patterns, as well as wind and solar availability. This system includes five seasons (Winter, Spring, Summer, Rainy, and Autumn), with seven representative times of day per season (Night, Sunrise, Morning, Afternoon, Sunset, Evening, and Peak).

The demand in each hour is allocated to a particular time-slice based on the corresponding season and time of day. For example, the Winter Night time-slice represents mean electricity demand between 12:00 a.m. and 6:00 a.m. from December through January. Sunrise and sunset periods are determined based on 2022 solar generation profiles, representing the first and last three hours of the day when solar generation is available, respectively.

Periods of seasonal peak load for each state are determined based on the highest 40 state-wise demand hours. After every hour of the year is allocated to one of the 35 time-slices, the time-slice load is calculated as the mean load from all hours assigned to that time-slice.

This system is designed to capture major seasonal and diurnal trends in load and wind and solar resources needed for resource adequacy planning while maintaining a manageable number of decision variables. However, it is not without its limitations.

The time-slice approximation tends to overestimate periods with both very high load and very low load. System planners concerned with resource adequacy are most concerned about high-load periods. In both 2017 and 2047, the time-slice approximation overestimates peak load by 8%, which is equivalent to 13 GW and 46 GW, respectively. The actual peak demand tends to be lower than the time-slice approximation because, within the same season and part of the day (e.g., Summer Evening), states may experience peak demand on different days or different hours. The normalized root mean square error between the actual and approximate load is 14%.

For renewable energy (RE) resources, the time-slice approximation underestimates periods of both very high wind and very high solar availability, which may result in an underestimate of RE curtailment. For both wind and solar, the normalized root mean square error is 10%. For wind, the time-slice approximation tends to underestimate wind resources during the high wind months of June and July. Approximation errors for solar follow a daily pattern where the time slices overestimate solar resources during the morning and underestimate them during the afternoon.

In conclusion, while India’s system of 35 time-slices is a useful tool for capturing major seasonal and diurnal trends in load and RE resources, it is not without its limitations. System planners and operators must be aware of these limitations and work to develop solutions that take them into account to ensure a reliable and sustainable energy system for India’s future.

NREL Report https://www.nrel.gov/docs/fy20osti/76153.pdf