Road lighting efficiency improvement through data processing and automated computation

October 21, 2020

Road lighting design requires extraction of crucial parameters i.e. lighting situations from the data (both road and lighting infrastructure) regarding the area intended for the retrofit. As the commercially available software is not capable of batch processing geospatial data, this process is currently done by hand, which inevitably leads to many simplifications and omissions, especially in large-scale projects. For that reason, an attempt to automatically derive the lighting situations from the official data concerning the area of Washington, D.C has been performed. The process assumed determination of spatial relationships between three sets of data (street lights, street segments, and roads) in order to obtain the luminaires and their corresponding road segments. The procedures, by means of which the best ascription candidate was chosen, were formed by initially reviewing the problematic cases manually and extending the algorithms accordingly, until the satisfactory outcomes were achieved. Therefore, a set of criteria has been determined to minimize the ambiguity of the assignments. As a result, 99% of street lights have been automatically assigned to the road segments. The results suggest that the lighting design performed with the help of the developed solution yields significantly more accurate outcomes in comparison with the other considered approaches. As the tool is versatile, it also allows to automatically estimate the possible power savings for any other area without the need for human assistance. This is especially helpful during the strategic planning phase of the lighting retrofit projects.

Authors in the community:

• 

  J

  Jakub Marek Starczewski

  ist194396

Supervisors of this institution:

• 

  C

  Claudia Martins Antunes

  ist14046

mechanical-engineering - Mechanical engineering

eng - English

Instituto Superior Técnico