Distributed MPC of Residential Energy Storage for Voltage Regulation and Peak Shaving Along Radial Distribution Feeders
Brendan Banfield, Duane Robinson, Ashish P. Agalgaonkar
Abstract
This article presents two low bandwidth distributed model predictive control (MPC) based algorithms for the coordinated control of residential energy storage (ES) to mitigate overvoltage and reduce peak demand along LV radial distribution feeders. Each ES unit consists of a low level controller (LLC) that utilises MPC with three distinct objective functions that can optimise for financial benefit, voltage regulation and peak shaving. These modes are activated via binary signals sent by the high level controller (HLC). The HLC, operating at the feeder level, determines if any network violations such as overvoltage or overloading are predicted to occur. Using inter-bus <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\boldsymbol{P}$</tex-math></inline-formula> , <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\boldsymbol{|V|}$</tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\boldsymbol{\Delta P}$</tex-math></inline-formula> data, the HLC determines the minimum ES units to dispatch to mitigate these issues. The result is a very low bandwidth distributed control architecture with low computational complexity, that can achieve acceptable network operating conditions while dispatching the minimum number of ES units. The proposed distributed MPC algorithms were shown to outperform decentralised control strategies for voltage regulation and peak shaving from the Australian Standards and also had significantly less computational requirements than a centralised MPC controller to achieve the same network overvoltage and peak load objectives.