Search

This paper presents the implementation of a new online real-time hybrid load-forecasting model based on an autoregressive model and neural networks. This new system is currently running at the Spanish Transport System Operator (REE) and provides an hourly forecast for the current day and the next nine days timely every hour for the national system as well as 18 regions of Spain. These requirements impose a heavy computational burden that needs to be considered during the design phase. The system is developed to improve forecasting accuracy specifically on difficult days like hot, cold and special days. In order to achieve this goal, a deep analysis of the temperature series from 59 stations is made for each region and the relevant series are included individually in the model. Special days are also analyzed and a thorough classification of days is proposed for the Spanish national and regional system. The model is designed and tested with data from 2005 to 2015. The results provided for the period from December 2014 to October 2015 show how the addition of the proposed model to the TSO’s ensemble causes a 5% RMSE overall error reduction and a 15% reduction on the 59 difficult days considered in the testing period. [ABSTRACT FROM AUTHOR]

Abstract: Demand response has proven to be a useful mechanism that produces important benefits for both the customer and the power system. In the context of an increasingly competitive electricity market, where prices are constantly rising and the presence of renewable energy resources is gaining prominence, this paper analyzes the flexibility potential of customers in the meat industry, based on the management of the most energy consuming process in this type of segment: cooling production and distribution. The effectiveness of the proposed actions has been successfully tested and validated in an active factory that produces cured ham in Spain, where savings of about 5% in the total annual cost of electricity have been assessed, together with power reductions in the range of 50% of the total peak demand of the studied facilities. Such results demonstrate the efficacy of these techniques, and they open the door to an innovative perspective on the evaluation of flexibility among customers which are traditionally considered rigid, providing a novel approach to the management of customer infrastructures in order to exploit their flexibility in electricity markets. [Copyright &y& Elsevier]

We examine price by season and by year in 14 deregulated markets, by looking at diurnal price patterns and price volatility. Power price volatility is measured by price velocity, the daily average of the absolute value of price change per hour, which is broken into a component expected from the average diurnal pattern, and an unexpected component. Deregulated markets can be categorized into three groups: stable markets, markets that experienced one bad period or season of high price excursion, and chaotic markets. Britain, Spain and Scandinavia show consistent price patterns and low unexpected price velocity; a thoughtful consumer could reasonably implement demand side management (DSM) by shaping consumption patterns to reflect price. California, New Zealand and Alberta are examples of markets that had a period of very high price excursion; we discuss factors affecting public reaction to this. Australian power markets have inconsistent price patterns from season to season and year to year, and very high unexpected price velocity. Planned DSM in these markets would be very difficult. We offer four policy considerations for markets considering deregulation in the future. [ABSTRACT FROM AUTHOR]

• Holidays affect electricity load greatly, causing inaccurate forecasts. • This effect varies widely among holidays, special days and adjacent days. • Wide and detailed classification of each special day can reduce forecasting error. • The proper methodology eliminates the analisys effort needed for classification. • An automatic classification can outperform an expert-level classification. Electricity demand presents a repetitive pattern following daily, weekly and seasonal patterns. However, factors like temperature or social events tend to disrupt these patterns introducing outlying data that is difficult to forecast. This paper introduces a new methodology to classify special days without any prior knowledge of the database. Simple classification of special days into two or three categories is insufficient as the consumers' behavior has many shades on these days. However, classifying special days in a wide range of categories required a deep understanding of the consumers' behavior on different days and periods of the year. The methodology proposed describes an algorithm to automate this classification starting from a simple day-of-the-week classification and branching into as many categories as needed to fit a real database. Categories with similar profiles are merged to avoid overfitting and actual outliers are detected to ensure that no false categories are created. The methodology is developed using data from 2010 to 2017 and tested in three different systems. The benchmark used is the classification used by the Transmission System Operator in Spain and the test show that the proposed methodology provides more accurate results without the need of an expert to develop the classification. [ABSTRACT FROM AUTHOR]