ABSTRACT
Smart cities offer new technological solutions to meet the increased energy demand. Smart lighting systems are one of these solutions to reduce energy consumption. In this study, a smart street lighting system is modeled for the Istanbul Technical University campus. Standard street lighting fixtures are substituted with light-emitting diode (LED) fixtures that require less energy together with a smart control system that enables dimming adjustment. This work also develops a Seasonal Autoregressive Integrated Moving Average – Long Short-Term Memory (SARIMA-LSTM) hybrid model to estimate total energy consumption. The total electricity consumption for two scenarios (old and new systems) was calculated. It is observed that the hybrid system gives better results. The purposed smart system achieves an 18% energy efficiency.
