WEB is interested in developing a load forecasting model that can predict WEB’s peak load and annual generation requirement between 2025 and 2029. The model should forecast electricity generation based on forecasts for key independent variables such as temperature, tourism arrivals, distributed generation, EV usage, and others agreed with WEB.
Under this assignment, K&M was responsible for collecting and validating historical data for Aruba, including hourly temperature, humidity, tourism arrivals, distributed solar PV additions, and others. K&M processed the data by cleaning missing values, aligning different time resolutions, and applying smoothing techniques. Using historical data, K&M analyzed the correlation between the dependent variable, Peak Load (MW), and each of the independent variables to determine which variables to include in the polynomial regression analysis. Based on the regression analysis, K&M developed the forecasting model, which can predict WEB’s peak load based on forecasted data for the period 2025 – 2029. K&M will provide this model as a workbook to WEB, which includes a user-friendly dashboard for data visualization and scenario analysis, as well as a user manual.