Optimizing storage operations in medium- and long-term power system models

In this paper, we propose a new methodology to formulate storage behavior in medium- and long-term power system models that use a load duration curve. Traditionally in such models, the chronological information among individual hours is lost; information that is necessary to adequately model the operation of a storage facility. Therefore, these models are not fully capable of optimizing the actual operation of storage units, and often use pre-determined data or some sort of peak-shaving algorithm. In a rapidly changing power system, the proper characterization of storage behavior and its optimization becomes an increasingly important issue. This paper proposes a methodology to tackle the shortcomings of existing models. In particular, we employ the so-called system states framework to recover some of the chronological information within the load duration curve. This allows us to introduce a novel formulation for storage in a system states model. In a case study, we show that our method can lead to computational time reductions of over 90% while accurately replicating hourly behavior of storage levels.