A Stochastic Analysis of The Economics of Solar and Storage

Abstract

As the world becomes more environmentally friendly, an emphasis is made on transitioning from fossil-fuel electricity generation to renewable sources. However, the integration of renewable resources such as solar poses a challenge both on a residential and grid-level due to high costs and an intermittent nature of weather conditions. To ameliorate the problem of intermittence, a commonly proposed solution is energy storage. In this thesis we explore the economic viability of storage and solar power from two distinct settings; first behind the meter, where we observe economics of solar and storage for a single household. We then shift in front of the meter, and model a revenue optimization problem for a grid-scale generator with a significant solar portfolio. On a residential level, we find that not only does storage exhibit an increasing marginal cost, but that a solar and storage portfolio can never outperform the grid on a month-by-month cost basis. For our grid-level problem, we implement an exhaustive search to optimize over a parametric policy function approximation. The results show that charging from the grid at lower prices has a higher impact on maximizing revenue than does discharging the battery at higher prices.