Análise estatística de séries temporais em demanda de energia elétrica da UTFPR câmpus Curitiba, sede centro

Abstract

The statistical analysis of time series has been used a lot in the approach of problems that try to construct models for the series, with determined purposes. Among the main tools of this analysis are the integrable autoregressive and moving average models (ARIMA), the recurrence plots (RP) together with the quantitative recurrence analyzes (QRA), which are used in the analysis of time series supposedly coming from of stochastic models. Such tools have been applied in a wide range of phenomena, from the daily stock market indexes to the record of tides in a port. In this work, the ARIMA models were used to study a time series of measured electric energy demand in Curitiba, UTFPR, where the behavior of the demand series, the periodicity of the data, the patterns formed in the RP well as the quantity values in the RQA, analyzed possible predictions of future values of the series, comparing the results obtained with the original and predicted series. From these results, he sought to establish if there is any kind of correlation between the campuses of Curitiba, UTFPR, and their intensity. In addition to, carry out the study of the series of measured demand for electric power.