Browsing by Author "Oliva, Diego"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Fractional Lévy flight bat algorithm for global optimisation
    (Inderscience Publishers, 2020) Boudjemaa, Redouane; Oliva, Diego; Ouaar, Fatima
    A well-known metaheuristic is the bat algorithm (BA), which consists of an iterative learning process inspired by bats echolocation behaviour in searching for prays. Basically, the BA uses a predefined number of bats that collectively move on the search space to find the global optimum. This article proposes the fractional Lévy flight bat algorithm (FLFBA), which is an improved version of the classical BA. In the FLFBA the velocity is updated through fractional calculus and a local search procedure that uses a random walk based on Lévy distribution. Such modifications enhance the ability of the algorithm to escape from local optimal values. The FLFBA has been tested using several well-known benchmark functions and its convergence is also compared with other evolutionary algorithms from the state-of-the-art. The results indicate that the FLFBA provided in several cases better performance in comparison to the selected evolutionary algorithms
  • No Thumbnail Available
    Item
    A multi-objective approach to weather radar network architecture
    (Springer, 2019) Boudjemaa, Redouane; Oliva, Diego
    This paper proposes a multi-objective optimization approach for the optimal placement of a weather radar network. Assuming a finite geographical region and a limited number of weather radars, a network is produced by considering the minimization of the total partial beam blocking percentage of the network and the minimization of network installation and maintenance costs. Several constraints on the solutions are considered such as terrain topography, radar beam elevation, distance between radars and distance from the power grid and roads. In order to reduce the number of possible combinations of radar networks, the solution space is discretized into a gridded system. The multi-objective optimization problem is solved by four different evolutionary algorithms, and the obtained results are used in a land clutter simulation of the whole network. The presented approach can serve as an analysis tool for a decision support system by providing meteorologist a set of Pareto optimal solutions to facilitate the selection of future prime sites for the installation of weather radars