한전, 인공지능 활용한 가스터빈 운영 소프트웨어 개발
‘인공지능 기술을 이용한 가스터빈 압축기 성능저하 예측 기술’
KEPCO develops gas turbine operation software using artificial intelligence
'Gas turbine compressor performance degradation prediction technology using artificial intelligence technology'
[The story of the case / Reporter Park Yeon-pa] = KEPCO (CEO Jong-gap Kim, KEPCO) developed a'gas turbine compressor deterioration prediction technology' through joint research with Siemens, Germany in October 2020.
Gas turbine power generation is attracting attention as eco-friendly energy because it emits only 1/8 of fine dust such as sulfur oxides and nitrogen oxides compared to coal-fired power plants.
A gas turbine is a facility that rotates a turbine and generates electricity by burning compressed air with fuel under high pressure conditions. During long-term operation, the compressor that compresses air is contaminated with fine dust, and the gas turbine performance is degraded. To prevent this, power plants periodically perform cleaning to remove contaminants adhering to the compressor blades.
Compressor contamination is difficult to check with the naked eye due to its complicated structure, and the power plant sets a certain period and performs compressor cleaning, but there is a problem that additional costs are incurred due to unnecessary cleaning.
In order to improve gas turbine power plant efficiency and reduce maintenance costs paid to foreign gas turbine manufacturers, KEPCO started an international joint research with KEPCO-Siemens to improve gas turbine performance in February 2020, and'Gas turbine compressor performance degradation prediction software' Developed.
The software developed this time is a technology that predicts the degradation of gas turbine compressor performance and compressor contamination by inputting the temperature, humidity, and operation status data obtained in real time from the power plant measurement equipment into artificial intelligence technology.
This allows the power plant operator to determine whether the compressor needs cleaning and reduce unnecessary costs.
When this software is used for compressor cleaning, the number of cleanings is expected to decrease three times a year, and it is possible to reduce costs of 4.4 billion won annually when applying the system to 74 gas turbines owned by domestic power generation companies.
In the future, KEPCO will also develop a program to optimize the air filter replacement cycle through its own research.
KEPCO announced that it has laid a foothold for entering the gas turbine overseas market through this software development, and that it will lead the core gas turbine technology by establishing a cooperation system with overseas companies such as Siemens.
<저작권자 ⓒ 사건의내막 무단전재 및 재배포 금지>