• 한국수산과학회지
• /
• 제43권6호
• /
• pp.693-704
• /
• 2010

Seasonal variation in species composition around Dokdo, East Sea of Korea, was investigated using trammel-net catches, from 2006 to 2009. A total of 53 fish species belonging to 23 families in 12 orders were captured; the orders Perciformes (12 families, 22 species) and Scorpaeniformes (four families, 22 species) were dominant. Between 2008 and 2009, 43 species were collected by trammel net. The number of species was highest in August 2009 (25 species) and lowest in February 2009 (11 species). The number of individuals and total biomass peaked in November 2009. Diversity indices for fish catches were highest in August 2008 (2.4368) and lowest in November 2009 (0.4253). The dominant species were Thamnaconus modestus and Sebastes schlegeli. Hierarchical clustering analysis showed five fish groups, with frequency and number of individuals similar to results of correspondence analysis (CA), which showed a closer relationship to the year term than the season term. CA showed that temperature was an important factor influencing fish species richness and abundance. Three main fish assemblage types coexisted around Dokdo: an East Sea coastal fish assemblage, a subtropical fish assemblage, and a cold water fish assemblage.

• 사물인터넷융복합논문지
• /
• 제10권2호
• /
• pp.17-23
• /
• 2024

국내 상수도관의 파열, 결함 등으로 인한 누수율로 인한 손실이 매우 크고, 이런 누수를 예방을 위한 방지 대책이 필요한 상황이다. 본 논문에서는 진동 센서를 활용한 누수 탐지 센서를 개발하고 인공지능 기술을 활용한 최적의 누수 탐지 알고리듬을 제시하고자 한다. 상수도 배관에서 취득한 진동음은 FFT(Fast Fourier Transform)를 이용한 전처리 과정을 거친 뒤, 최적화된 트리 기반 부스팅 알고리듬을 적용하여 누수 분류를 하였다. 다양한 실증 환경에서 취득한 약 26만여 개의 실험 데이터에 적용한 결과 기존의 SVM(Support Vector Machine) 방법에 비해약 4%가 향상된 97%의 정확도를 얻었고, 연산 처리속도는 약 1,362배가 향상되어 엣지 디바이스 적용에도 적합함을 확인하였다.