| Popis: |
This study presents a statistical approach that allows estimating the sea wave power (Pmax) by using the joint probability distribution of the wave’s significant height (Hm) and period (T), pHm,T. Unlike other works in the literature, the goal of the present paper is not to find the best locations for wave farming or to extrapolate the data in space or time, but to achieve a more suitable design of wave electric generators in terms of a reduced delivery uncertainty. When the raw values of Hm and T are used, the estimations of Pmax end up spanning over several orders of magnitude owing to the nonlinear amplification of the fluctuations through the calculations. This typically results in large overestimations of Pmax that preclude their use for a modest design. Another associated issue is the low temporal resolution of the data available, which forbids using measurements of Hm and T in numerical simulations directly. These issues have been overcome with the proposed methodology, whose core feature consists of generating a new series of random Hm,T pairs, based on pHm,T, which allow estimating the wave power potential realistically while reducing the chances of over/underrating. In this new series, large fluctuations are suppressed, but the essential statistics of the measurements are preserved. Additionally, since the new series of Hm,T pairs is time-independent, it can be fed directly into numerical simulations. To illustrate our proposal, we used one-year long datasets of hourly measurements of Hm and T recorded at a maritime buoy, for the design and simulation of a linear electric generator with a single degree of freedom. |
