Energy plays an important role in every stage of human life in different forms and variations. Along with the developments in economic, social and industrial fields, the amount of energy that countries need is increasing day by day. Therefore, it is significant to estimate the energy demand for a country's economic activities accurately. In this study, the energy demand forecast (EDF) application optimization problem of Turkey, one of the real-world optimization problems, was performed by MTSA (Modified Tree Seed Algorithm) and TSA (Tree Seed Algorithm) methods. From 1979 to 2005, gross domestic product (GDP), population, export and import values were used as parameter data. Thus, in the presence of three different possible scenarios, Turkey's energy demand from 2006 to 2025, which was estimated by MTSA and TSA methods. To demonstrate the success of MTSA and TSA in the problem of energy demand forecasting (EDF), they are compared with Ant Colony Algorithm (ACO), Particle Swarm Optimization (PSO), Bat Algorithm (BA), Differential Evolution Algorithm (DEA) and Artificial Algae Algorithm (AAA) methods which are in the literature. According to the results of the analysis, it was observed that the MTSA method was a successful estimation tool for energy demand.