Although over a thousand scientific papers address the topic of load forecasting every year, only a few are dedicated to finding a general framework for load forecasting that improves the performance, without depending on the unique characteristics of a certain task such as geographical location. Meta-learning, a powerful approach for algorithm selection has so far been demonstrated only on univariate time-series forecasting. Multivariate timeseries forecasting is known to have better performance in load forecasting. In this paper we propose a meta-learning system for multivariate time-series forecasting as a general framework for load forecasting model selection. We show that a meta-learning system built on 65 load forecasting tasks returns lower forecasting error than 10 well-known forecasting algorithms on 4 load forecasting tasks for a recurrent real-life simulation. We introduce new metafeatures of fickleness, traversity, granularity and highest ACF. The meta-learning framework is parallelized, component-based and easily extendable.
This paper analyzes a price-taker hydro generating company which participates simultaneously in day-ahead energy and ancillary services markets. An approach for deriving marginal cost curves for energy and ancillary services is proposed, taking into consideration price uncertainty and opportunity cost of water, which can later be used to determine hourly bid curves. The proposed approach combines an hourly conditional value-at-risk, probability of occurrence of automatic generation control states and an opportunity cost of water to determine energy and ancillary services marginal cost curves. The proposed approach is in a linear constraint form and is easy to implement in optimization problems. A stochastic model of the hydro-economic river basin is presented, based on the actual Vinodol hydropower system in Croatia, with a complex three-dimensional relationship between the power produced, the discharged water, and the head of associated reservoir.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.