DOI QR코드

DOI QR Code

커널회귀 모델기반 가스터빈 축진동 신호이상 분석

Kernel Regression Model based Gas Turbine Rotor Vibration Signal Abnormal State Analysis

  • Kim, Yeonwhan (KEPCO Research Institute, Korea Electric Power Corporation) ;
  • Kim, Donghwan (KEPCO Research Institute, Korea Electric Power Corporation) ;
  • Park, SunHwi (KEPCO Research Institute, Korea Electric Power Corporation)
  • 투고 : 2018.08.28
  • 심사 : 2018.11.19
  • 발행 : 2018.12.30

초록

본 논문에서는 가스 터빈 축 진동 신호 비정상 상태 분석의 사례 연구를 위해 커널 회귀 모델을 적용한다. 원격으로 전송되는 발전소 가스터빈의 진동데이터에 커널 회귀 모델을 적용하여 설비를 실시간으로 감시 및 분석 외에도, 축진동 신호의 비정상 상태를 분석하기 위하여 활용될 수 있다. 정상운전 중에 측정한 가스터빈의 정상적인 축진동 데이터 기반의 훈련데이터를 사용하여 생성한 자동연관커널회귀의 경험적 모델을 생성하고 적용할 수 있다. 이 데이터 기반 모델의 예측치를 실시간 데이터와 비교하여 신호의 상태를 분석하고 잔차를 감시하여 이상상태에 대한 분석 정보를 제공할 수 있다. 이상상태에서 발생하는 잔차는 비정상적으로 변화됨으로서 비정상 상태를 분석 할 수 있다. 본 논문에서 커널회귀모델은 축진동 센서의 신호 이상의 원인 분석 사례에서 고장을 구분할 수 있는 정보를 제공한다.

In this paper, the kernel regression model is applied for the case study of gas turbine abnormal state analysis. In addition to vibration analysis at the remote site, the kernel regression model technique can is useful for analyzing abnormal state of rotor vibration signals of gas turbine in power plant. In monitoring based on data-driven techniques correlated measurements, the fault free training data of shaft vibration obtained during normal operations of gas turbine are used to develop a empirical model based on auto-associative kernel regression. This data-driven model can be used to predict virtual measurements, which are compared with real-time data, generating residuals. Any faults in the system may cause statistically abnormal changes in these residuals and could be detected. As the result, the kernel regression model provides information that can distinguish anomalies such as sensor failure in a shaft vibration signal.

키워드

JROGG5_2018_v4n2_101_f0001.png 이미지

Fig.1. Architecture of remote data transmission network.

JROGG5_2018_v4n2_101_f0001.png 이미지

Fig.1. Architecture of remote data transmission network.

JROGG5_2018_v4n2_101_f0002.png 이미지

Fig 2. Trend of gas turbine rotor vibration to load. (a) 1B~4B rotor vibration trends, (b) power(=load) trend.

JROGG5_2018_v4n2_101_f0002.png 이미지

Fig 2. Trend of gas turbine rotor vibration to load. (a) 1B~4B rotor vibration trends, (b) power(=load) trend.

JROGG5_2018_v4n2_101_f0003.png 이미지

Fig.3. Data‐sets selected by grouping component variables of 2B_1X vibration for learning.

JROGG5_2018_v4n2_101_f0003.png 이미지

Fig.3. Data‐sets selected by grouping component variables of 2B_1X vibration for learning.

JROGG5_2018_v4n2_101_f0004.png 이미지

Fig.4. Validation of regression prediction models of 2B_1X B vibration.

JROGG5_2018_v4n2_101_f0004.png 이미지

Fig.4. Validation of regression prediction models of 2B_1X B vibration.

JROGG5_2018_v4n2_101_f0005.png 이미지

Fig.5. Plots of result demonstrating 2B_1X vibration prediction models.

JROGG5_2018_v4n2_101_f0005.png 이미지

Fig.5. Plots of result demonstrating 2B_1X vibration prediction models.

JROGG5_2018_v4n2_101_f0006.png 이미지

Fig.6. Trends of 1B~4B vibration to power of 12.01.2016~12.16.2016. (a) signals of 1B~4B vibration sensors, (b) trend of power.

JROGG5_2018_v4n2_101_f0006.png 이미지

Fig.6. Trends of 1B~4B vibration to power of 12.01.2016~12.16.2016. (a) signals of 1B~4B vibration sensors, (b) trend of power.

JROGG5_2018_v4n2_101_f0007.png 이미지

Fig.7. 2B vibration prediction to power. (a) “0” area's prediction result to power, (b) “2” area's prediction result to power.

JROGG5_2018_v4n2_101_f0007.png 이미지

Fig.7. 2B vibration prediction to power. (a) “0” area's prediction result to power, (b) “2” area's prediction result to power.

Table 1. Correlation analysis result

JROGG5_2018_v4n2_101_t0001.png 이미지

Table 1. Correlation analysis result

JROGG5_2018_v4n2_101_t0001.png 이미지

참고문헌

  1. In‐Tae Lee, Ji‐Hoon Lee, Suk‐Won Jang, Yeon‐Whan Kim, In‐Chel Lee,Won‐Hae Suh, 2015, Design and Implementation of the Cyber Power Generation Operation Monitoring System, IPEG 2015.
  2. J.W. Hines, D. Garvey, R. Seibert, and A. Usynin, 2005, Technical Review of On‐line Monitoring Techniques for Performance Assessment, Nuclear Regulatory Research U.S. Nuclear Regulatory Commission, NUREG‐6895, Vol. 3 pp.10-24.
  3. JAMIE GARVEY, DUSTIN GARVEY, REBECCA SEIBERT and J. WESLEY HINES, 2007, VALIDATION OF ON‐LINE MONITORING TECHNIQUES TO NUCLEAR PLANT DATA, Nuclear Engineering and Technology, Vol.39 NO.2 pp.149-158. https://doi.org/10.5516/NET.2007.39.2.149
  4. Jae‐Young Jung, Byoung‐Oh Lee, Hyoung‐Kyun Kim and Dae‐Woong Kim, 2016, Performance Analysis on Early Detection of Fault Symptom of a Pump with Abnormal Signals, Journal of the Korean Society for Power System Engineering, Vol. 20, No. 2, pp. 66-72.