Page 279 - Fundamentals of Ocean Renewable Energy Generating Electricity From The Sea
P. 279

266 Fundamentals of Ocean Renewable Energy
























            FIG. 9.21  A map of the current PJM system (thick lines). The meteorological stations (inland and
            offshore) for the wind energy assessments are shown as asterisks. (Reproduced from C. Budischak,
            D. Sewell, H. Thomson, L. Mach, D.E. Veron, W. Kempton, Cost-minimized combinations of wind
            power, solar power and electrochemical storage, powering the grid up to 99.9% of the time, J. Power
            Sources 225 (2013) 60–74, with permission from Elsevier.)


            9.5.2 OpenTidalFarm
            OpenTidalFarm is an open source software designed to simulate and optimize
            arrays of tidal energy converters (TECs; http://opentidalfarm.org). Optimization
            of TEC arrays generally requires the running of computationally expensive 2D
            or 3D models, and so only a limited number of options can be explored; for
            example, Divett et al. [39] explored ‘only’ four TEC array configurations using
            their 2D shallow water equation model to maximize global array power output.
            Funke et al. [40] presented an automated procedure for TEC array optimization
            using the adjoint technique, based on a 2D shallow water equation model.
            In contrast to less efficient gradient-free optimization algorithms (such as the
            greedy algorithm presented in Section 9.3.1, which optimizes the objective
            function, e.g. power output, as a black box), the adjoint method is a gradient-
            based method. Gradient-based optimization algorithms update the position in
            parameter space at each iteration using derivatives of the objective function.
            However, as mentioned before, this involves differentiating through the solution
            of a partial differential equation, which can be challenging in the case of
            complex models. In the tidal energy case, we wish to optimize a single
            variable—power output—with respect to many input parameters, and ‘adjoint
            linearization’ is a method that can efficiently compute the derivative of a single
            output with respect to all inputs.
               OpenTidalFarm has, amongst other applications, been used to optimize the
            siting of 256 turbines in the Inner Sound of the Pentland Firth [40], and to
            minimize the cost of cabling in the same region [12] (Fig. 9.22).
   274   275   276   277   278   279   280   281   282   283   284