Setting the Operating Reserve Using Probabilistic Wind Power Forecasts

Abstract

In power systems with a large integration of wind power, setting the adequate operating reserve levels is one of the main concerns of system operators (SO). The integration of large shares of wind generation in power systems led to the development of new forecasting methodologies, including probabilistic forecasting tools, but management tools able to use those forecasts to help making operational decisions are still needed. In this paper, a risk evaluation perspective is used, showing that it is possible to describe the consequences of each possible reserve level through a set of risk indices useful for decision making. The new reserve management tool (RMT) described in the paper is intended to support the SO in defining the operating reserve needs for the daily and intraday markets. Decision strategies like setting an acceptable risk level or finding a compromise between economic issues and the risk of loss of load are explored. An illustrative example based on the Portuguese power system demonstrates the usefulness and efficiency of the tool.Index Terms-Multicriteria decision, operating reserve, operating risk, uncertainty, wind power forecast.

Probabilistic forecasting of the wave energy flux

“…The value of probabilistic forecasts in decision-making related to the optimal integration of renewable energy generation in power systems is increasingly recognized, as for the example case of reserve quantification [4]. This follows from a more general result: for a large class of decisionmaking problems, optimal decisions directly relate to quantiles of conditional predictive distributions [5].…”

Use of wind power forecasting in operational decisions.

“…A detailed probabilistic analysis of the impact of renewable sources in the power system of the Iberian Peninsula is given in [74]. A methodology to define operating reserve considering conventional generation outages, load forecast uncertainty, and wind power forecast uncertainty is described in [75]. Probabilistic forecasts are used for wind power.…”

“…The optimal choice of quantile range for OR wind generally depends on the cost of scheduling additional reserves compared to the cost of not meeting the regular reserve requirements. It should be noted that a more rigorous approach could be used to derive OR wind (e.g., based on the ideas in [75]). …”

“…Stochastic UC has been proposed as one approach to better handle the wind power uncertainty in market operations [55]- [58] and [66]- [68], [70]. Furthermore, new approaches to calculating operating reserve requirements under wind power uncertainty have also been proposed (e.g., based on assumed normal distributions of WPF errors [29][48], or on a probabilistic wind power forecast [75]). It is obvious that WPF can provide important inputs to the operational decisions of the system operator.…”

Development and testing of improved statistical wind power forecasting methods.

“…The pdf representation is also useful to set the required operating reserve for the current and next days using, for instance, the method presented by Matos and Bessa [24]. The pdf representation provides the full probability distribution, which allows a better characterization of the tails.…”