| description abstract | To achieve higher thermal efficiency, gas turbines operate at increasingly higher turbine inlet temperatures, leading to the need for advanced cooling methods such as film cooling, impingement cooling, and passage cooling in modern high-pressure turbines (HPTs). However, accurately predicting the nonuniform temperature distribution at the HPT inlet in real-time has been a challenge, resulting in oversized cooling flow rates of up to 25% of the core engine flow. To address this issue, this paper proposes a novel approach based on discrete measurements acquired at the HPT exit to predict the turbine inlet temperature and nonuniformity magnitude. The proposed method utilizes a multiwavelet approximation technique to extract the hot-streaks-related component at the HPT exit, which is then used to approximate the HPT inlet temperature distribution based on the hot-streaks' decay rate across the HPT. This paper provides an overview of the proposed method and its potential applications in the gas turbine industry. Additionally, the impact of burner-to-burner asymmetry on the accuracy of the proposed method is analyzed, and recommendations for improving its effectiveness in such scenarios are presented. | |
