1 SHORT-TERM LOAD FORECAST USING TREND INFORMATION AND PROCESS RECONSTRUCTION Santos, P.J. a, Martins, A.G. b , Pires, A.J. a , Martins, J. F. a , Mendes, R. V. c a LabSei-Department of Electrical Engineering, ESTSetúbal/InstitutoPolitécnico de Setúbal , Rua do Vale de Chaves Estefanilha 2914-508 Setúbal Portugal: e-mail: psantos@est.ips.pt ; apires@est.ips.pt jmartins@est.ips.pt : tel:+351 265 790 000 fax:+351 265 721 869 b Department of Electrical Engineering FCTUC/INESC, Polo 2 University of Coimbra, Pinhal de Marrocos, 3030 Coimbra Portugal; e-mail: amartins@deec.uc.pt tel: +351 239 796 279 fax: +351 239 796 247 c Laboratório de Mecatrónica, Instituto Superior Técnico, Av. Rovisco Pais, 1096 Lisboa Codex, Lisboa Portugal e-mail: vilela@cii.fc.ul.pt : tel +3510218417435 fax: +351 218417167 ABSTRACT The algorithms for short-term load forecast (STLF), especially within the next-hour horizon, belong to a group of methodologies that aim to render more effective the actions of planning, operating and controlling electric energy systems (EES). In the context of the progressive liberalisation of the electricity sector, unbundling of the previous monopolistic structure emphasizes the need for load forecast, particularly at the network level. Methodologies such as artificial neural networks (ANN) have been widely used in next-hour load forecast. Designing an ANN requires, amongst other things, the proper choice of input variables, avoiding overfitting and an unnecessarily complex input vector (IV). This may be achieved by trying to reduce the arbitrariness in the choice of endogenous variables. At a first stage, we have applied the mathematical techniques of process-reconstruction to the underlying stochastic process, using coding and block entropies to characterize the measure and memory range. At a second stage, the concept of consumption trend in homologous days of previous weeks has been used. The possibility to include weather-related variables in the IV has also been analyzed, the option finally being to establish a model of the non-weather sensitive type. The paper uses a real-life case study. Keywords Distribution Systems, Load forecasting, Measure, Memory range, Consumption trend, Artificial neural networks. 1. INTRODUCTION Distribution companies (DISCO) operating on a scenario of complete or partial unbundling of the electricity sector are confronted with increasing demands on planning, management and operation of the networks. Relations with generation, transmission and retail companies (GENCO, TRANSCO, RESCO) are now becoming increasingly complex (Gross, 1987). Therefore, DISCOs play a major role in the managing and planning of distribution, with an emphasis on the quality of the supply.