Page 252 - Fundamentals of Ocean Renewable Energy Generating Electricity From The Sea
P. 252

242 Fundamentals of Ocean Renewable Energy


               A simple analogy to optimization is climbing towards the summit of a
            mountain (maximum elevation). Here, the objective function is elevation, and
            latitude and longitude are the decision variables. Assume that a helicopter leaves
            you at a random location on the mountain. Based on the best local information,
            you move towards a direction for a specific distance (step size), and then you
            change your direction. The distance that you move in each direction is not
            necessarily the same: the step size can also vary within each iteration. An
            obvious direction is going up, and in particular in the steepest ascent direction.
            The steepest ascent is mathematically the direction of the gradient, which can
            be written as

                                       ∂f   ∂f     ∂f
                                ∇f =      ,   , ... ,                   (9.4)
                                       ∂x 1 ∂x 1   ∂x n
            and the method of steepest ascent becomes

                                    x k+1 = x k + α k ∇f k              (9.5)
               This method almost guarantees that you reach a summit that is closest to
            your initial location but, as you can imagine, it does not guarantee that you
            will reach the highest summit of a mountain range. Many classical optimization
            techniques are based on gradient. The method of steepest ascent (or descent in
            minimization) is not very effective, especially near the solution. Referring to
            Fig. 9.1, it is clear that the gradient of the function (slope of the tangent) near
            the local and global minimums approaches zero. This causes some issues for
            methods that only use gradient for the direction. Other techniques are called
            conjugate gradient methods that use second derivatives (Hessian matrix) as well
            as first derivative (gradient) for direction.



            9.2 INTRAARRAY OPTIMIZATION
            Although understanding the natural (undisturbed) resource, and the operation of
            single devices, is essential, it is only when devices are installed in arrays that
            significant levels of electricity generation can be achieved. In this section, we
            discuss how devices within such arrays can be optimized for wind, tidal stream,
            and wave energy conversion.



            9.2.1 Micro-Siting of Offshore Wind Farms
            Referring to Section 4.4, marine spatial planning is the standard procedure used
            to find suitable and potential sites for the development of offshore wind farms in
            a region. Marine spatial planners, after a comprehensive study of technical, eco-
            nomic, environmental, sociopolitical, legal, and regulatory aspects, recommend
            areas that have minimum conflicts with other users of the ocean, in addition to
            a feasible energy resource. Macro-siting is the selection of a location for a wind
   247   248   249   250   251   252   253   254   255   256   257