Accurate load forecasting constitutes a fundamental component of power system planning, operational optimization, stability assessment, and effective network management. In this study, Artificial Neural Networks (ANN) and an Adaptive Neuro-Fuzzy Inference System (ANFIS) were implemented within the MATLAB/Simulink environment to predict electrical load demand for the study area over a 13-year horizon (2023–2035). The forecasting framework utilized month and year as input variables, while aggregated load consumption served as the output parameter. During model training, the ANFIS achieved a minimum mean squared error (MSE) of 0.307983, converging rapidly at epoch 2. In comparison, the ANN model recorded an MSE of 0.3201 at epoch 10. The lower MSE and faster convergence rate indicate superior predictive performance of the ANFIS model for the Abuloma 33 kV distribution network. For the ANN model, the regression coefficients (R-values) obtained during training, validation, and testing phases were 0.9289, 0.9262, and 0.9501, respectively, demonstrating a strong correlation between the predicted outputs and the measured load data. Historical load data from the Abuloma 33 kV Injection Substation (2010–2022) revealed a steady increase in demand from 4.9 MW to 9.1 MW. Based on the ANFIS forecasting results, the projected load demand is expected to rise from 10.076 MW in 2023 to approximately 13.73 MW by 2035, indicating sustained growth in electricity consumption within the network.

This paper focuses on electrical load forecasting utilizing ANFIS and ANN. Abuloma 33/11kV Injection Substation serves as the distribution network for this investigation. A 13-year forecast of electricity usage for 2023–2035 was completed. In order to forecast future power usage, the ANFIS and ANN models were designed and simulated. The minimal mean squared error (MSE) for the training of ANFIS is 0.307983 and the training was completed at epoch 2 whereas, the MSE for ANN model is 0.3201 at epoch 10. ANFIS model performed better in predicting the power consumption training, validation and testing were 0.9289, 0.9262 and 0.9501. The regression results showed that there is a close fit between the actual data and the neural network results. The actual power consumption data of Abuloma 33kV Injection Substation from 2010 to 2022, showed that there was an increase in power consumption from 4.9MW to 9.1MW and the predicted power consumption rose from 10.076MW in 2023 to 13.73MW in 2035 using ANFIS model.

[1] Amit, J., Srinivas, E., & Rasmimayee, R. (2010). Short-term load forecasting using fuzzy adaptive inference and similarity. Nature & Biologically Inspired Computing. [2] Amlabu, C. A., Amber, J. U., Onah, C. O., & Mohammed, S. Y. (2020). Electric load forecasting: A case study of the Nigerian power sector. International Journal of Engineering and Innovative Technology. [3] Al-Kandari, M., Soliman, S. A., & El-Hawary, M. E. (2004). Fuzzy short-term electric load forecasting. International Journal of Electrical Power and Energy Systems, 26. [4] Aslan, Y., Yavasca, S., & Yascar, C. (2019). Long term electric peak load forecasting of Kutahya using different approaches. International Journal on Technical and Physical Problems of Engineering (IJTPE), 3(2), 87–91. [5] Babatunde, D. E., Babatunde, O. M., Akinbilire, T. O., & Oluseyi, P. O. (2018). Hybrid energy systems model with the inclusion of energy efficiency measures: A rural application perspective. International Journal of Energy Economics and Policy. [6] Cullen, K. A. (2019). Forecasting electricity demand using regression and Monte Carlo simulation under conditions of insufficient data. Division of Resource Management, Morgantown, West Virginia. [7] Eneje, I. S., Fadare, D. A., Simolowo, O. E., & Falana, A. (2020). Modelling and forecasting periodic electric load for a metropolitan city in Nigeria. African Research Review, 6(1), 101–115. [8] Ibe, A. O. (2020). Power system analysis (1st ed.). Odus Press. [9] Kucukdeniz, T. (2020). Long-term electricity demand forecasting: An alternative approach with support vector machines. Istanbul University of Engineering Sciences, 1, 45–53. [10] Moradzadeh, A., Mohammadi-Ivatloo, B., Abapour, M., Anvari-Moghaddam, A., & Roy, S. S. (2021). Heating and cooling loads forecasting for residential buildings based on hybrid machine learning applications: A comprehensive review and comparative analysis. IEEE Access, 10, 2196–2215. [11] Mitchell, B. M., Ross, J. W., & Park, R. E. (2020). A short guide to electric utility load forecasting. Rand. [12] Samuel, I., Adetiba, E., Odigwe, I. A., & Felly-Njoku, F. C. (2017). A comparative study of regression analysis and artificial neural network method for medium term load forecasting. Indian Journal of Science and Technology, 10(10), 1–7.