Forecasting Geothermal Installation Capacity Worldwide to 2030: Application of an Improved Grey Prediction Model to the Top 10 User Countries | Chapter 4 | Geography, Earth Science and Environment: Research Highlights Vol. 8

Geothermal energy is gaining strong interest from both the private and public sector. Analysts expect that its use will grow rapidly over the next several decades at many places in the world. The number of annual geothermal unit installations reflects this growing demand. Foresight of geothermal energy installation is valuable for energy decision-makers, allowing them to readily identify new capacity units, improve existing energy policies and plans, expand future infrastructure, and fulfill consumer load needs. Therefore, in this chapter, an improved grey prediction model (IGM (1,1)) was applied to perform the annual geothermal energy installation capacity prediction for the top 10 countries based on installed power generation capacity evaluated at the end of 2021, namely the United States, Indonesia, Philippines, Turkey, New Zealand, Mexico, Italy, Kenya, Iceland, and Japan, for the next nine years for the period from 2022 through 2030. The grey prediction model (GM) is a mathematical model that can be used to characterize an unknown system behaviour using a small data size, which contains a minimum of four data points. Due to the GM (1,1) model producing an inaccurate prediction, its accuracy prediction needs to be improved. Thus, the IGM (1,1) model was proposed to reduce prediction error and improve the overall GM (1,1) model performance. All these data in the investigation can be used by future researchers in the field. Separately, datasets from 2000 to 2021 were collected for each country’s geothermal energy installation capacity to build a model which can accurately predict the annually geothermal energy installation capacity by 2030. The IGM (1,1) model used a small dataset of 22 data points, with one point denoting one year (i.e., 22 years), to predict the capacity of geothermal energy installations for the next nine years. Following that, the model was implemented for each dataset in MATLAB, where appropriate, and the model accuracy was evaluated. In this study, three metrics were used to assess the prediction accuracy of the applied models: mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE). Ten separate geothermal energy installation capacity datasets were used to validate the improved model, and these datasets further demonstrated the overall improved model’s accuracy. According to the IGM (1,1) model prediction results, the current ranking of countries utilizing geothermal energy is expected to change by 2030. The results prove that the prediction accuracy of the IGM (1,1) model outperforms the benchmark conventional GM (1,1) model, thereby enhancing the overall accuracy of the GM (1,1) model. The IGM (1,1) model ensures error reduction, suggesting that it is an effective and promising tool for accurate short-term prediction. The results reveal the 2030 geothermal energy installation capacity rankings from the United States having the larger capacity of 3.925 GW to Japan having the smaller capacity of 0.481 GW.

 

Author (s) Details

Khaled Salhein
Department of Electrical and Computer Engineering, Oakland University, Rochester, MI 48306, USA and Department of Control Engineering, College of Electronic Technology, Bani Walid 322, Libya.

 

C. J. Kobus
Department of Electrical and Computer Engineering, Oakland University, Rochester, MI 48306, USA.

 

Mohamed Zohdy
Department of Electrical and Computer Engineering, Oakland University, Rochester, MI 48306, USA.

 

 

Please see the book here:- https://doi.org/10.9734/bpi/geserh/v8/4632