Abstract Electricity plays a crucial role in the development of countries, serving as the foundation for the functioning of modern societies and as a key indicator of a nation’s level of modernization and overall strength. Predicting a country’s electricity demand provides valuable insights into the evolving patterns of electrical energy provision and usage and helps to improve the flexibility of the energy distribution network. In addition, it provides early warnings and decision support for governments, businesses, and individuals about future electricity consumption. This plays a pivotal function in optimizing power distribution, adjusting the proportion of power generation sources, and ensuring a continuous and secure electricity supply. This paper uses the example of electricity demand in Panama and uses the LSTM model and Prophet model to compare their predictions for electricity demand in lower-income countries, with the aim of improving accuracy. Furthermore, it introduces a data set representing the electricity demand of higher-income countries, using the United Kingdom as a representative, to compare the forecast results between the two countries. The results indicate that within the Panama data set, the Prophet model exhibits better accuracy than the LSTM model. On the contrary, in the data set representing higher-income countries with the United Kingdom as a representative, the LSTM model outperforms the Prophet model. This leads to the conclusion that the selection of a predictive model is specific to each country, influenced by factors such as development status, geographic location, policies, energy markets, and more. A specific model may not be universally applicable due to these unique conditions, requiring tailored-modeling approaches for precise and efficient allocation and scheduling of electricity resources. This approach ensures the improvement of residents’ living standards and quality of life while meeting the typical demands of a country’s economic and social development.
Read full abstract