Sizing and economic assessment of energy Storage with real-time pricing and ancillary services

Energy storage units (ESU) can generate profit through providing multiple services. In order to maximize profit while meeting battery lifetime constraints, it is necessary to select the capacity and charge/discharge schedule of the battery. This paper proposes a methodology to calculate the necessary ESU battery capacity for a given power rating and the lifetime profitability of the ESU. Because the price of electricity is not known more than 24 hours in advance, forecasting is necessary when scheduling the ESU. The reduction in profitability of the ESU because of forecasting errors must be taken into account. The methodology applies ridge regression for price forecasting to overcome the problem of having many highly correlated inputs, which lead to poor performance of the forecaster. The methodology demonstrates the encouraging result that the profitability of an ESU is not highly sensitive to forecaster error, as the scheduler is insensitive to the types of errors introduced by the forecaster. This is because of the nature of the variance in electricity market price.