Efficient Control Algorithm for a Smart Solar Street Light

This paper describes a model of a smart street light that can work autonomously, save energy and can be seamlessly integrated with the existing grid. The street light incorporates a Light Emitting Diode array and an ARM Cortex M0 based microcontroller, both of which are powered by a combination of a solar panel and battery pack. The microcontroller executes an astronomical time switch based estimation algorithm to determine the sunrise and sunset time daily. This is done using the Real Time Clock running on the controller along with the hard coded coordinates and time zone of any given location. Upon evaluation of the sun times, the controller can routinely regulate the lamp at sunrise and sunset. The algorithm operates with an accuracy of up to 10 seconds in its estimation of sun times. The controller also monitors the voltage from the battery at sunset and switches to mains to power up the LED array in case power is not enough. A coin cell is also connected to the controller to ensure RTC keeps on running in the scenario that battery is completely discharged and there is no power from the mains. The results conclude that the use of such a technology would provide high energy efficiency, increases the operating life and also proves to be cost effective as compared to prevalent lamp technologies. Analysis is performed using HOMER, a microgrid simulation software. The boundary conditions are set for Indian scenarios; however the result applies to many geographical locations.