loading page

A Machine Learning Framework for Predicting the LCOE of PV Systems using Demographic, Energy and Policy Data
  • +2
  • Satyam Bhatti ,
  • Ahsan khan ,
  • Ahmed Zoha ,
  • Sajjad Hussain ,
  • Rami Ghannam
Satyam Bhatti
Author Profile
Ahsan khan
Author Profile
Ahmed Zoha
Author Profile
Sajjad Hussain
Author Profile
Rami Ghannam
University of Glasgow

Corresponding Author:[email protected]

Author Profile

Abstract

The Levelized Cost of Electricity (LCOE) is a widely used economic parameter that determines a power plant’s unit cost of energy over its lifetime. In fact, it facilitates economic decisions and quantitative comparisons between different energy generation technologies. Previous methods for calculating the LCOE were based on fixed singular input values that do not capture the uncertainty associated with determining the financial feasibility of a PV project. Instead, we propose a dynamic model that takes into account important demographic, energy and policy data that includes interest rates, inflation rates and energy yield. All these parameters will undoubtedly vary during a photovoltaic (PV) system’s lifetime and will help determine a more accurate LCOE value. Furthermore, comparisons between different ML algorithms revealed that the ARIMA model gave an accuracy of 93.8% for predicting the consumer price of electricity. Moreover, we validated our proposed model with two case studies from the United States and the Philippines. Results from these case studies revealed that LCOE values for the State of California can be almost 30% different (5.03 cent/kWh for singular values in comparison to 7.09 cent/kWh using our ML model), which can distort the risk or economic feasibility of a PV power plant. Additionally, our ML model predicts the ROI of a grid-connected PV plant in the Philippines to be 5.37 years instead of 4.23 years.