No abstract

No abstract

Abstract:This research aims to estimate the long-term financial benefits of using smart grids to mitigate and adapt the power sector to climate change. In order to do that, twelve scenarios were analyzed applying an energy accounting model (LEAP (Long-range Energy Alternatives Planning System)) that was developed using Brazilian historical data from 1970 to 2015. To conduct the analysis, the Sathaye and Ravindranath's three steps methodology was used. The main final results include a long-term cost-benefit analysis that is developed for each considered scenario. The initial phase includes the analysis of the projections for the power sector up to 2030. The following phase consists on the estimation of costs for operation, maintenance, losses and new electrical projects investments. And finally, all scenarios' results were compared and the benefits of implementing smart grids in the sector were estimated. The attained results show that smart grid implementation would contribute to reduce electricity tariffs, the generation costs as well as the costs associated with theft and fraud.

No abstract

Evolutionary programming (EP) is a metaheuristic method developed as an alternative approach to artificial intelligence. The aim of this paper is to bring an introduction to EP algorithms through the implementation of the basic D. B. Fogel’s Evolutionary Programing approach of 1988 and the emulation of his results in order to analyze the performance of the evolutionary programming method on solving a benchmark test case. The EP approach is implemented thru a simple simulation of natural evolution and the allowance of probabilistic survival of individuals. The novelty of this paper relies on testing the algorithm performance in some problems of well-known benchmark instances of the Traveling Salesman Problem, where that 1988 evolutionary approach was not tested. The reproduction of 1988 D. B. Fogel’s approach was possible, the found average error of this method for 200000 offspring applied to the benchmark instances was found to be in the order of the 10%.