A Context-Aware Street Light System Based on Multi-variate Forecast and Fuzzy Logic

Abstract

The rapid growth in urbanization increases the number of street lamps worldwide in rural areas and cities. This places a heavy demand on electric energy, which stems from fossil fuels, thus increasing greenhouse emissions. By using stand-alone photo-voltaic street lamps, gas emissions are lowered. However, the vital aspect of street lamps is the safety of users at nighttime. The batteries of these systems are quickly depleted due to adverse weather conditions. In order to increase safety and smartly manage energy consumption, in this work we suggest using a fuzzy controller with traffic and solar radiation forecasts in this work. We suggest using a fuzzy controller with traffic and solar radiation forecasts in this work. The controller adapts light according to traffic demand, solar radiation in the upcoming three days, and battery level. First, we tested and validated several multi-step forecast models to predict solar radiation. Then the description of the system, along with simulations, was carried out. The obtained results indicate that the designed light controller is capable of lowering energy consumption, thus prolonging the system’s autonomy while at the same time assuring road safety.