A methodology for reliability-based traffic signals alternative power capacity design

Traffic signals power failures can degrade the efficiency and safety at signalized intersections. Alternative Power Sources (APS), such as battery banks, can be used as backup power to maintain normal signal operations during power outages. Currently, the transportation agencies in the U.S. use engineering judgment and past experience to make the decisions on APS design placement. There is a need to define a performance metric to quantify, measure, and model the impact of APS. Statistical models can be developed for APS project planning and evaluation; advanced optimization models can be formulated for generating optimal APS designs. In recent years, some state transportation agencies have used wind and solar energy for transportation facilities. The Renewable Energy assisted APS (REAPS), if used at a signalized intersection, have the potential to improve the reliability of traffic signals operation and reduce the dependence on grid power. There is a need to study the feasibility of using REAPS as backup power for traffic signals. This dissertation assesses the power production potential of REAPS, and develops a reliability-based methodology to identify the optimal capacity of REAPS. The following transformational contributions to the field of transportation engineering are included: (1) This research develops a new performance metric, traffic signal power reliability, to quantify the impacts of providing REAPS at a signalized intersection. (2) This research develops original power failure models for signalized intersections using real-world data. (3) This research, for the first time, evaluates the feasibility of using REAPS at a signalized intersection. Weibull and uniform distributions are calibrated to predict the wind and solar resources available at the height of a traffic pole. A Gaussian-exponential model is developed to estimate the power output of a small wind turbine. (4) A Monte-Carlo based stochastic simulation tool is coded to fuse the estimates from the power failure, renewable resource, and renewable power generation models to assess the traffic signal power reliability. (5) This research formulates two stochastic optimization schemes as decision support tools