Page - 66 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

Improving Daily Electricity Loads Mohcine Bakhat a,b,1 and Jaume Rossello Nadal c a Research fellow, Economics for Energy, Vigo (Spain) E-mail: mohcine@uvigo.es b Business economics department, University Sultan Moulay Slimane (Morocco) E-mail: mohcine.bakhat@usms.ma c Department of Applied Economics, Universitat de les Illes Balears (SPAIN) E-mail: jrossello@uib.es Abstract. Electricity demand forecasting is becoming an essential tool for energy management, maintenance scheduling and investment planning. In small and isolated electric systems like in many islands, the power system is usually not extensively interconnected with enough number of electric generators and loads, hence it provokes an electric load shedding or forces the electric utility to take control actions such as temporary power outage. Despite this trouble, isolated territories are fortunate to have the possibility of estimating the population pressure in a very accurate manner, even at a daily level. This is possible thanks to the fact that entries to these territories are limited to ports and airports which in turn facilitate keeping the record of in/out flows to/from the territory. Investigating this problem, using the most classical and standard prediction techniques applied to the case of Balearic Islands (Spain) authors demonstrate how daily arrivals and people stocks improve accuracy of forecast. Keywords. electricity load forecasting, isolated electric systems, daily models, population pressure, tourism 1. Introduction The need of having an accurate electricity load forecast is of crucial importance to electricity planning in both short-term and long-term. The former is relevant because the national grid requires a balance between the electricity produced and consumed at any instant of the day, in other words it is essential to balance the generation capacity with the system demand. The latter is important for staffing, maintenance, and capital investment planning. As the operation and control processes for an electric utility are quite sensitive to forecasting errors, their costs can be easily increased when these errors are large [16]. Therefore, forecasting accuracy is a basic tool for determining the optimal utilization of generators and power stations, especially that electricity is not a storable item and some facilities are more efficient than others. In fact, overestimating the future load results in unused spinning reserve which is being burnt for nothing, in the other side underestimating future load is equally detrimental because high starting costs are incurred if the cold reserve has to be used. Furthermore, in situations when the option to buy at last minute from other suppliers remains obviously expensive; and when demand is exceeding power supply, electric companies may intentionally cut power partially or totally in order to avoid a total blackout of the power system [7]. 1 Corresponding Author, E-mail: mohcine.bakhat@usms.ma Forecasting in Tourist Isolated Areas Intelligent Environments 2019 A. Muñoz et al. (Eds.) © 2019 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/AISE190025 66