AI-Based Thermal Efficiency Prediction and Operational Economic Optimization Strategy for Steam Turbines

Abstract

In the power industry, steam turbines are critical equipment; their thermal efficiency and operational economic performance are of significant importance for energy conservation, emission reduction, and lowering power generation costs. Traditional methods for predicting thermal efficiency and optimizing turbine operation suffer from limitations such as low model accuracy and poor adaptability. This study employs artificial intelligence (AI) technologies to develop an AI-based predictive model for steam turbine thermal efficiency. By comprehensively collecting steam turbine operational data and subjecting it to rigorous data preprocessing, this model selects appropriate AI algorithms to construct a neural network architecture; after training and validation, it demonstrates excellent predictive performance. Furthermore, based on the thermal efficiency prediction results, a multi-objective optimization algorithm is utilized to formulate an operational economic optimization strategy, enabling precise adjustment of steam turbine operating parameters. Through a practical case study, this strategy is validated as an effective means of enhancing the operational economic efficiency of steam turbines, providing robust support for the sustainable development of the power industry.