A Dynamic Programming Approach for the Decentralized Control of Energy Retrofit in Large-Scale Street Lighting Systems

Abstract

This article proposes a decision-making procedure that supports the city energy manager in determining the optimal energy retrofit plan of an existing public street lighting system throughout a wide urban area. The proposed decision model aims at simultaneously maximizing the energy consumption reduction and achieving an optimal allocation of the retrofit actions among the street lighting subsystems, while efficiently using the available budget. The resulting optimization problem is formulated as a quadratic knapsack problem. The proposed solution relies on a decentralized control algorithm that combines discrete dynamic programming with additive decomposition and value functions approximation. The optimality and complexity of the presented strategy are investigated, demonstrating that the proposed algorithm constitutes a fully polynomial approximation scheme. Simulation results related to a real street lighting system in the city of Bari (Italy) are presented to show the effectiveness of the approach in the optimal energy management of large-scale street lighting systems. Note to Practitioners —This article addresses the emerging need for decision support tools for the energy management of urban street lighting systems. The proposed decision-making strategy allows city energy managers and local policy makers taking retrofit decisions on an existing public street lighting system throughout a wide urban area. The presented strategy can be implemented in any engineering software, providing decision makers with a low-complexity and scalable Information and Communication Technology (ICT) tool for the optimization of the energy efficiency and environmental sustainability of street lighting systems.