Vol. 1 No. 1 (2023): Dynamic Science and Engineering Infrastructure-base for AchievingĀ Innovations

Day Ahead Forecasting of Photovoltaic Power Output in Maiduguri Using Feedforward Neural Network

Abdullahi Bukar
Physics Department, Federal University Gusau, Zamfara State
cover page

Published 2023-10-11


  • Feed Forward Neural Network (FFNN),
  • APE (Absolute Percentage Error),
  • MAPE (Mean Absolute Percentage error)

How to Cite

Bukar, A., Musa, A., Abbagana, M., & Umar Mustapha, B. M. . (2023). Day Ahead Forecasting of Photovoltaic Power Output in Maiduguri Using Feedforward Neural Network. Nigerian Journal of Science and Engineering Infrastructure, 1(1). Retrieved from https://njsei.naseni.gov.ng/index.php/nsjei/article/view/65


With the increasing use of large scale grid-connected PV system, accurate forecasting approach for the power output PV system has become an important issue. Power planning is necessary for cost efficiency of power generation in which power forecasting is an essential part. For the Feed Forward Neural Network (FFNN) models, pre-processing techniques was used on the relevant variables. In this research work, FFNN based model was designed to forecast the next day PV power output of Maiduguri depending on available variables (Time of the day, day of the week, temperature, relative humidity, wind speed, cloud cover and PV power output) for the model. The FFNN based models were designed in the MATLABĀ® (R2015a) environment. Hence based on the results obtained from this research, it can be concluded that the FFNN based model performance is satisfactory for predicting next day PV power output. From the mean absolute percentage error (MAPE) value of 8.9093 on the test data set, it can be concluded that the Model had a better MAPE value which is an indication that the system performance is better than ANFIS which had 15.0048 as MAPE.