http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
풍력발전기 상태 감시를 위한 SaaS 클라우드 인프라 내 데이터 처리 알고리즘 개선 연구
이광세(Gwang-Se Lee),최정철(Jungchul Choi),강민상(Minsang Kang),박사일(Sail Park),이진재(JinJae Lee) 한국신재생에너지학회 2020 신재생에너지 Vol.16 No.1
In this study, an SW for the analysis of the wind-turbine vibration characteristics was developed as an application of SaaS cloud infrastructure. A measurement system for power-performance, mechanical load, and gearbox vibration as type-test class was installed at a target MW-class wind turbine, and structural meta and raw data were then acquired into the cloud. Data processing algorithms were developed to provide cloud data to the SW. To operate the SW continuously, raw data was downloaded consistently based on the algorithms. During the SW test, an intermittent long time-delay occurred due to the communication load associated with frequent access to the cloud. To solve this, a compression service for the target raw data was developed in the cloud and more stable data processing was confirmed. Using the compression service, stable big data processing of wind turbines, including gearbox vibration analysis, is expected.
풍력발전기 증속기 상태를 감시하기 위한 SaaS 클라우드 인프라 개발
이광세(Gwang-Se Lee),최정철(Jungchul Choi),강승진(Seung-Jin Kang),박사일(Sail Park),이진재(Jin-jae Lee) 한국산학기술학회 2018 한국산학기술학회논문지 Vol.19 No.9
본 논문에서, 풍력발전기 운영관리 및 유지보수 비용을 저감하기 위해, 분산되는 전산자원을 통합하고 인적 자원을 효율적으로 운영 할 목적으로 SaaS 클라우드 방식의 상태 감시 인프라를 설계 및 개발하였다. 개발한 인프라에서 관련 업무 및 서비스에 따라 각 데이터들을 계층화 하였다. 인프라 상에서 상태 감시를 수행 할 경우에 필요한 기본적인 SW를 개발하였다. 측정 시스템에 대응하는 데이터베이스 설계 SW, 현장 측정 SW, 데이터 전송 SW, 모니터링 SW로 구성되어 있다. 기존의 SCADA 데이터 뿐 아니라 추가적인 센서를 설치하여 풍력발전기의 상태 관측이 가능하다. 상태감시 알고리즘 내단계 별 지연 시간을 모델링하여, 현장 측정에서 최종 모니터링 단계 까지 소요되는 총 지연 시간을 정의 하였다. 진동 데이터는 고해상도 측정에 의해 취득되기 때문에, 지연 시간은 불가피하고 유지보수에 관한 프로그램 운영 시 지연시간 분석은 필수적이다. 모니터링 대상은 MW 용량의 풍력발전기의 증속기이며, 해당 풍력발전기는 운전 한 지 10년이 넘는 모델로서, 이는 앞으로 해당 풍력발전기의 효율적인 유지보수를 위해 정확한 상태 감시가 필수적임을 뜻한다. 본 인프라는 연간 50 TB 용량의 고해상도 풍력발전기 증속기 상태를 처리 할 수 있도록 운영 중이다. In this paper, to integrate distributed IT resources and manage human resource efficiently as purpose of cost reduction, infrastructure of wind turbine monitoring system have been designed and developed on the basis of SaaS cloud. This infrastructure hierarchize data according to related task and services. Softwares to monitor conditions via the infrastructure are also developed. Softwares are made up of DB design, field measurement, data transmission and monitoring programs. The infrastructure is able to monitor conditions from SCADA data and additional sensors. Total time delay from field measurement to monitoring is defined by modeling of step-wise time delay in condition monitoring algorithms. Since vibration data are acquired by measurements of high resolution, the delay is unavoidable and it is essential information for application of O&M program. Monitoring target is gearbox in wind turbine of MW-class and it is operating for 10 years, which means that accurate monitoring is essential for its efficient O&M in the future. The infrastructure is in operation to deal with the gearbox conditions with high resolution of 50 TB data capacity, annually.
1.5 MW 풍력 터빈 소음 측정 및 저주파 소음 특성 분석
손은국(Eunkuk Son),이광세(Gwang-Se Lee),이진재(Jinjae Lee),강승진(Seungjin Kang),황성목(Sungmok Hwang),박사일(Sail Park),김석우(Seokwoo Kim) 한국신재생에너지학회 2018 신재생에너지 Vol.14 No.4
The noise from a 1.5 MW wind turbine was measured and the apparent sound power level, tonal audibility, and spectrum balance were analyzed. The apparent sound power level and tonal audibility were analyzed according to IEC 61400-11: ed3. (2012-11). The measured noise data at the turbine site was mainly in the north-west (NW) and north-north-west (NNW) directions, and approximately 500 and 250 samples of total noise and background noise data were obtained. Three tone components were observed in the low frequency region below 100 Hz. The tones in the low frequency critical band with a center frequency of 78.1 Hz were found to have a higher level than the hearing threshold. The possibility of wind turbine noise annoyance was also analyzed through the spectrum balance analysis.
최정철(Choi Jungchul),손은국(Son Eunkuk),이광세(Lee Gwangse),강민상(Kang Minsang),이진재(Lee Jinjae),황성목(Hwang Sungmok),박사일(Park Sail) 한국태양에너지학회 2021 한국태양에너지학회 논문집 Vol.41 No.4
Continuous fatigue information is essential for the structural health monitoring (SHM) of wind turbines. Faults, such as sensor failure, data loss, and cable disconnection, can result in a total loss of SHM. To avoid such a malfunction, machine learning algorithms and polynomial curve fitting are suggested to predict the missing fatigue data from the otherwise known measurement data. Artificial neural networks showed the best prediction performance. Decision trees and regularized linear regression are also powerful alternatives.