zubairumar339-sketch/Nigeria-Regional-Electricity-Demand-Prediction

The project was developed as part of research on applying machine learning techniques to regional electricity demand forecasting in Nigeria. It demonstrates how data-driven models can support more efficient energy planning and decision-making. # Nigeria Regional Electricity Demand Prediction # Overview This project develops a machine learning-based regional electricity demand prediction model for Nigeria. The objective is to accurately forecast electricity demand across the country's six geopolitical zones using weather, demographic, economic, and regional characteristics. Accurate demand forecasting supports effective energy planning, efficient power distribution, and informed policy-making. The project applies Linear Regression and Random Forest Regressor models to predict electricity demand, with Linear Regression selected as the final model due to its superior predictive performance. # Objectives • Predict regional electricity demand in Nigeria. • Analyse the factors influencing electricity consumption across regions. • Compare the performance of different machine learning regression models. • Identify the most suitable predictive model for electricity demand forecasting. # Dataset The dataset represents regional electricity demand observations across Nigeria's six geopolitical zones. # Features Date Observation date Region Nigerian geopolitical zone Temperature_C Average regional temperature (°C) Humidity_pct Relative humidity (%) Population_Density Population density Industrial_Index Industrial activity index Commercial_Index Commercial activity index Residential_Index Residential electricity consumption index GDP_Index Regional economic performance indicator Electricity_Price_NGN_kWh Electricity tariff (₦/kWh) Holiday Public holiday indicator Weekend Weekend indicator Rainfall_mm Rainfall amount (mm) Peak_Hour Time-of-day electricity usage category Month Month of observation Day_of_Week Day of the week Electricity_Demand_MWh Target variable (Electricity demand in MWh) # Technologies Used • Python • Pandas • NumPy • Scikit-learn • Matplotlib • Seaborn (optional) • Jupyter Notebook # Machine Learning Models The following regression models were implemented: • Line …