‘Good’ or ‘bad’ wind power forecasts: a relative concept

Abstract

This paper reports a study on the importance of the training criteria for wind power forecasting and calls into question the generally assumed neutrality of the 'goodness' of particular forecasts. The study, focused on the Spanish Electricity Market as a representative example, combines different training criteria and different users of the forecasts to compare them in terms of the benefi ts obtained. In addition to more classical criteria, an information theoretic learning training criterion, called parametric correntropy, is introduced as a means to correct problems detected in other criteria and achieve more satisfactory compromises among confl icting criteria, namely forecasting value and quality. We show that the interests of wind farm owners may lead to a preference for biased forecasts, which may be in confl ict with the larger needs of secure operating policies. The ideas and conclusions are supported by results from three real wind farms. Copyright

Development and testing of improved statistical wind power forecasting methods.

“…Further information on the underlying criteria can be found in [10] and [11]. However, the cMCC criterion as a difference between the MCC and MEE criteria is introduced in the scope of this project -this criteria aims to exploit the benefits of the MCC and MEE criteria, avoiding the worsening of bias when MEE is used while also keeping the robustness MEE criterion offers.…”

“…Replacing (3)(4)(5)(6)(7)(8)(9)(10)(11) in (3-6), we have the following conditional density estimator:…”

“…Now, it is necessary to build an estimator for (3)(4)(5)(6)(7)(8)(9)(10)(11)(12). The idea proposed by Bouezmarni and Rombouts [38] was a semiparametric approach, where a parametric model is considered for the copula, and the marginal distributions are represented by a nonparametric model (empirical distribution function).…”

“…Fig. 3-4 depicts the copula pdf computed with (3-13) for the quantile transform of the wind speed and wind power, using (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14), from a real wind farm. This copula density function represents the probability density associated to each point plotted in the wind speed versus the wind power scatter of Fig.…”

“…Moreover, the variables u and v in (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15) are bounded between [0,1]; therefore, the Chen beta kernels will be used for these variables.…”