Use of Multiple Linear Regression Techniques to Predict Energy Storage Systems' Total Capital Costs and Life Cycle Costs

Abstract

In the United States, legislative and regulatory requirements are the primary drivers for the use of grid-level energy storage. In some cases, state-level legislative mandates called Renewable Portfolio Standards (RPSs) necessitate the use of storage to support renewable generation (e.g., solar, wind energy). At the federal level, the Federal Energy Regulatory Commission (FERC) has issued final Orders that stipulate fair and equitable competition rules for regional interstate transmission markets. In both instances, it is the investor-owned utility (IOU) entities that are financially responsible to either satisfy the state and FERC Orders or face noncompliance fines or tariffs. Unfortunately, utility investors do not have a reliable tool to assist in understanding the front-end installation costs or whole life cycle costs for electrical storage systems that service the electric grid. This paper proposes the use of multiple linear regression (MLR) techniques using R-Script to predict the total capital cost (TCC) and life cycle cost (LCC) of real-world energy storage systems (ESSs) derived from manufactures' design specifications and intrinsic characteristics of lead-acid, lithium-ion, sodium sulfur; and vanadium and Ainc-based Aow Aatteries.