ARMAX, OE and SSIF model predictors for power transmission and distribution predictions in Akure and its environs in Nigeria

  • Reginald O. A. Osakwe
  • Vincent A. Akpan
  • Sylvester A. Ekong
Keywords: Auto-regressive moving average with exogenous input (ARMAX) model, output-error (OE) model, state-space innovations form (SSIF) model, mathematical modeling, model predictor, model structure, power distribution, power transmission.

Abstract

Three mathematical model structures, namely: ARMAX, OE and a SSIF are first formulated followed by the formulation of their respective model predictors for the model identification and prediction of power transmission and distribution within Akure and its environs. A total of 51,350 data samples from the Power Holding Company of Nigeria were collected for thirteen different parameters that influences the evaluation and analysis in the case study area. The performances of these three model predictors are validated by one-step and five-step ahead prediction methods as well as the Akaike’s final prediction error (AFPE) estimates. The results obtained from the application of these three model structures and their predictors for the modeling and prediction of power transmission and distribution as well as the validation results show that the OE model predictor outperforms the ARMAX and SSIF model predictors with much smaller prediction errors, good prediction and tracking capabilities and that the OE model structure and its predictor structure can be used for power transmission and distribution modeling and predictions in real scenarios.

Keywords: Auto-regressive moving average with exogenous input (ARMAX) model, output-error (OE) model, state-space innovations form (SSIF) model, mathematical modeling, model predictor, model structure, power distribution, power transmission.

Published
2018-01-05
Section
Articles

Journal Identifiers


eISSN: 1596-0862