Model Predictive Control Strategy for Residential Battery Energy Storage System in Volatile Electricity Market with Uncertain Daily Cycling Load

Abstract

This paper presents a control strategy for residential battery energy storage systems, which is aware of volatile electricity markets and uncertain daily cycling loads. The economic benefits of energy trading for prosumers are achieved through a novel modification of a conventional model predictive control (MPC). The proposed control strategy guarantees an optimal global solution for the applied control action. A new cost function is introduced to model the effects of volatility on customer benefits more effectively. Specifically, the newly presented cost function models a probabilistic relation between the power exchanged with the grid, the net load, and the electricity market. The probabilistic calculation of the cost function shows the dependence on the mathematical expectation of market price and net load. Computational techniques for calculating this value are presented. The proposed strategy differs from the stochastic and robust MPC in that the cost is calculated across the market price and net load variations rather than across model constraints and parameter variations.

Keywords

• Optimal control
• model predictive control (MPC)
• energy market
• nonlinear constrained optimization
• revenue for battery energy storage system
• Gaussian mixture model