A conditional model of wind power forecast errors and its application in scenario generation

Abstract

A B S T R A C TIn power system operation, characterizing the stochastic nature of wind power is an important albeit challenging issue. It is well known that distributions of wind power forecast errors often exhibit significant variability with respect to different forecast values. Therefore, appropriate probabilistic models that can provide accurate information for conditional forecast error distributions are of great need. On the basis of Gaussian mixture model, this paper constructs analytical conditional distributions of forecast errors for multiple wind farms with respect to forecast values. The accuracy of the proposed probabilistic models is verified by using historical data. Thereafter, a fast sampling method is proposed to generate scenarios from the conditional distributions which are non-Gaussian and interdependent. The efficiency of the proposed sampling method is verified.

Multi-Time Scale Rolling Economic Dispatch for Wind/Storage Power System Based on Forecast Error Feature Extraction

“…For each time point, λj,t and et W are calculated by using Equations (1)−(5) when Nj = 2~96. After normalization of λj,t and et W , correlation coefficients Rj are calculated through Equation (6). After each step, 95 groups of correlation coefficients are calculated.…”

“…The factors, including wind power forecast output fluctuation (λ 1 ), short-term wind power output stability (λ 2 ), wind power output Energies 2018, 11, 2124 3 of 27 amplitude (λ 3 ) and short-term wind power forecast output accuracy (λ 4 ) are listed in Equations (2)- (5). The degree of relationship between these factors and e W is calculated by correlation coefficients, which is defined in Equation (6). e w t = w a t − w f t , …”

“…However, due to the strong uncertain and volatile characteristics of wind power, it is difficult to forecast and model accurately [3][4][5]. Overestimation or underestimation of wind power would lead to reserves under-committed, causing wind curtailment and load shedding [6]. Therefore, how to handle the wind power forecast error (WPFE) becomes a research hotspot [7][8][9].…”

“…This type of model needs to set a large capacity of spinning reserve to ensure the safety of the system operation, which is too conservative and not economical. The second type is the uncertain model, such as probabilistic model [14][15][16][17], fuzzy set theory [18,19], scenario analysis [6,20], stochastic programming theory [21][22][23][24], etc. For example, the authors in Reference [14] regard the forecast error of load demand and wind power as normal distribution, and propose a probabilistic model to estimate the reserve capacity demand.…”

A Distributed Probabilistic Modeling Algorithm for the Aggregated Power Forecast Error of Multiple Newly Built Wind Farms

“…To construct the conditional PDF of the aggregated WFE in a DN for a given aggregated FWO, the joint PDF of the aggregated AWO and FWO must first be constructed. Then, the conditional PDF can be derived from the joint PDF using simple algebraic calculation [19]. For constructing the joint PDF, common MAP estimation methods must gather the raw data pairs of AWO and FWO from all the NWFs to form the aggregated AWO and FWO historical data for training.…”