Multi-Objective Optimization of PLTMG Duri Engine 4-7 Operations Using RSM (Response Surface Methodology) – MOGA (Multi-Objective Genetic Algorithm) Methods for Efficiency Improvement and Emission Reduction

Abstract

The Balai Punggut-Duri Gas Engine Power Plant (PLTMG) plays an important role in meeting the electricity needs in Central Sumatra. However, its flexible operation causes significant variations in efficiency and emissions between units. This study aims to optimize the operation of units 4-7 of the Duri PLTMG by considering efficiency, NPHR, and NOx emissions. The Response Surface Methodology (RSM) method models the relationship between load and power factor variations with plant performance. Furthermore, the Multi-Objective Genetic Algorithm (MOGA) is applied to find the optimal solution that simultaneously considers efficiency, NPHR, and NOx emissions. The results show that optimization with MOGA-RSM successfully identified the optimal start-up pattern for each unit. The priority order of start-up based on efficiency, NPHR, and NOx is units 5, 7, 4, and 6. This study provides a new contribution by integrating RSM and MOGA for PLTMG operation optimization, which has not been widely done in previous studies. This approach improves energy efficiency and reduces environmental impacts by reducing NOx emissions. The resulting start-up pattern recommendations are expected to support more efficient and environmentally friendly PLTMG Duri operations.