Prediction of the performance of industrial gas turbine based on different environmental conditions.

This study examines the effect of ambient environmental conditions on the performance and efficiency of General Electric GE-9171E gas turbines used for electricity generation at the Nandipur power plant. The research emphasizes the importance of improving gas turbine efficiency to reduce fuel consumption, operational costs, and greenhouse gas emissions, thereby promoting environmentally sustainable power generation in Pakistan. The study specifically evaluates the influence of ambient temperature, humidity, and dew point on turbine output and operational efficiency. Real-time operational data were collected from three gas turbines operating at full load over a one-year period. Both conventional statistical methods and artificial intelligence (AI)-based approaches were employed for performance evaluation and prediction. Conventional techniques included descriptive statistics, correlation analysis, and regression modeling, while AI-based methods incorporated support vector machine (SVM) and artificial neural network (ANN) models. The findings indicate that ambient temperature has a significant negative impact on turbine performance, showing a strong correlation with output power and efficiency (r = -0.959, R2 = 92.1%). Humidity and dew point also affected turbine behavior, although their relationships with performance were comparatively less pronounced. A regression-based predictive model was developed and validated using operational plant data, achieving high prediction accuracy with a minimum error of -0.01088 for Gas Turbine-I. Comparative analysis demonstrated that ANN-based models outperformed conventional techniques in prediction capability. Overall, the study provides a reliable framework for predicting gas turbine performance and improving operational efficiency under varying climatic conditions in Pakistan.