Hybrid stochastic-robust optimization for smart parking lot trading with local electricity markets under a decentralized framework with renewable energy integration

Abstract

This paper presents a hybrid stochastic-robust optimization approach for trading smart parking lots (SPL) with the local electricity market (LEM) within a decentralized scheduling framework and considering the renewable energy sources (RES) participation. In the proposed structure, the smart parking operator aims to minimize operating costs by submitting offers/bids to the LEM. Additionally, the impact of implementing smart charging strategies in the trading process of SPLs with the LEM is discussed. Dischargin profle of SPLs have been modeled using the k-means clustering method and considering the uncertain behavior of electric vehicle (EV) owners. The aim of the LEM operator is to clear the electricity market while considering the physical constraints of the electricity distribution network (EDN), fluctuations in wholesale electricity market (WEM) prices, and the uncertain behavior of RES. In this study, the uncertain behavior of WEM price and RES is modeled by robust optimization (RO) and stochastic programming (SP), respectively. To implement energy trade between SPLs and the LEM, alternating direction method of multipliers (ADMM) algorithm has been used in the framework of decentralized optimization. The proposed problem is formulated as a second-order conic programming (SOCP) model, leveraging the benefits of convex optimization and efficiently solved using the MOSEK solver. Solving the proposed hybrid optimization problem using the ADMM algorithm leads to a robust solution, which enables trade between SPLs and the LEM while respecting privacy. The results show that implementing a smart electric vehicle charging strategy leads to a 59.18%, 8.56%, and 11.23% reduction in the first to third parking lots. In addition, implementing the smart charging strategy also reduces the operating costs of LEM by 0.3%.