KEPCO works with Simens to predict performance degradation of gas turbine compressors

Lim Chang-won Reporter() | Posted : October 14, 2020, 13:58 | Updated : October 14, 2020, 13:58

[Courtesy of KEPCO ]


SEOUL -- South Korea's state power company has worked with Siemens AG, an industrial manufacturing company in Germany, to develop a new technology that can predict the performance degradation of gas turbine compressors using artificial intelligence.

Gas turbines produce electricity with compressed air burned under high-pressure conditions along with fuel. The component efficiency of compressors and turbines is an important factor affecting thermal efficiency. Because of complicated structure, it is difficult to visually check, and plant operators should periodically remove contaminants attached to compressor blades.

To improve the efficiency of gas turbine power plants and reduce maintenance costs, Korea Electric Power Corporation (KEPCO) and Simens developed software for predicting the poor performance of gas turbine compressors. If the technology is applied to gas turbines in South Korea, KEPCO said that domestic power companies can save 4.4 billion won ($3.8 million) annually.

KEPCO said in a statement that AI analyzes real-time data such as temperature, humidity and operation status to predict the degradation of the gas turbine compressor and the level of contamination, allowing plant operators to reduce unnecessary costs. "With this software development, we have set the stage for entering the overseas gas turbine market," an unnamed KEPCO official was quoted as saying.

A gas turbine breakdown can cause an explosion and other problems. In September 2019, KEPCO unveiled an AI-based gas turbine monitoring system that would help thermal power plant operators to predict the malfunction of parts.
© Aju Business Daily & www.ajunews.com Copyright: All materials on this site may not be reproduced, distributed, transmitted, displayed, published or broadcast without the authorization from the Aju News Corporation.