스마트미디어저널 (Smart Media Journal)

  • 제13권4호
  • /
  • Pages.48-56
  • /
  • 2024
  • /
  • 2287-1322(pISSN)
  • /
  • 2288-9671(eISSN)
한국스마트미디어학회 (THE KOREAN INSTITUTE OF SMART MEDIA)
권호 표지
DOI QR코드

DOI QR Code

LSTM 오토인코더를 활용한 축산 환경 시계열 데이터의 이상치 탐지: 경계값 설정에 따른 성능 비교

Anomaly Detection in Livestock Environmental Time Series Data Using LSTM Autoencoders: A Comparison of Performance Based on Threshold Settings

  • 투고 : 2023.11.01
  • 심사 : 2023.12.20
  • 발행 : 2024.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

축산업에서 환경의 이상치 탐지와 데이터 예측은 매우 중요한 과제이다. 대부분 시계열 데이터로 수집되는 축산 환경 데이터의 이상치는 급격한 생육환경의 변화와 예상치 못한 전염병의 징후를 나타낼 수 있으므로 이상치를 빠르게 탐지하는 것이 중요하다. 이상치의 빠른 탐지와 효과적인 대응은 가축의 스트레스를 최소화하고 전염병 발생 환경을 조기에 발견하여 농가의 경제적인 손실을 감소시키는 역할을 할 수 있다. 본 연구에서는 축산환경 데이터의 이상치 탐지 분야에서 이상치를 규정하는 경계값(Threshold) 설정에서 두 가지 설정 방법을 이용하여 실험하고 성능을 비교하였다. Mean Squared Error(MSE)를 활용한 이상치 탐지 방법과 Dynamic Threshold를 이용한 이상치 탐지 방법을 이용하여 이를 통해 주어진 이전 데이터의 평균값과의 변동성을 분석하여 이상 상황을 식별하는 연구를 진행하였다. MSE를 활용한 이상치 탐지 방법은 94.98% 정확도를 보였고 표준편차를 활용한 Dynamic Threshold 방법은 99.66%정확도로 성능이 더 우수함을 확인할 수 있었다.

In the livestock industry, detecting environmental outliers and predicting data are crucial tasks. Outliers in livestock environment data, typically gathered through time-series methods, can signal rapid changes in the environment and potential unexpected epidemics. Prompt detection and response to these outliers are essential to minimize stress in livestock and reduce economic losses for farmers by early detection of epidemic conditions. This study employs two methods to experiment and compare performances in setting thresholds that define outliers in livestock environment data outlier detection. The first method is an outlier detection using Mean Squared Error (MSE), and the second is an outlier detection using a Dynamic Threshold, which analyzes variability against the average value of previous data to identify outliers. The MSE-based method demonstrated a 94.98% accuracy rate, while the Dynamic Threshold method, which uses standard deviation, showed superior performance with 99.66% accuracy.

키워드

과제정보

본 논문은 한국연구재단 대학중점연구소지원사업(NRF-2019R1A6A1A09031717)과 2023년도 정부(농림축산식품부, 과학기술정보통신부, 농촌진흥청 공동)의 재원으로 스마트팜연구개발사업단의 지원을 받아 수행된 연구임(No. 421023-04)

참고문헌

  1. Seung-Jae Kim, Meong-Hun Lee, Gwang-Ho Yang, & Hyun Yoe, "A Study on the Interface Specification between Smart Livestock Big Data Service Provider and Smart Livestock Barn System", The Journal of Korean Institute of Communications and Information Sciences, vol, 46, no. 12, 2021. 
  2. Jeong-Won Lee, Ku-Rye Shon, Han-Sol Jeong, Won-Seok Ko, Chae-Woong Lim, and Bum-Seok Kim, "Mannheimiosis case in Hanwoo caused by heat stress," Korean Journal of Veterinary Service, Vol. 38, No. 3, pp. 195-198, 2015. 
  3. Yu J, Yin P, Liu F, Cheng G, Guo K, Lu A, Zhu X, Luan W, Xu J, "Effect of heat stress on the porcine small intestine: a morphological and gene expression study", Comp Biochem Physiol A Mol Integr Physiol, vol. 156, no. 1, 2010. 
  4. Bong-Hyun Kim, "Implementation of Feeding Management Service Model based on Pig Raising Data," The Society of Digital Policy and Management, vol.19, no.10, pp. 105-110, 2021. 
  5. Kno Younsik, Jeong Kwangpil, Kweon Junoh, Insu Cho, & Yonghan Ju, "Development of Smart Factory Abnormal Detection Method Using Object Detection and AutoEncoder," Global Business Administration Review, vol. 19, no. 5, pp. 76-90, 2022. 
  6. Hwi-Min Choi, Kim joo man, "Anomaly Detection System of Smart Farm ICTDevice," The International Journal of Internet, Broadcasting and Communication, vol. 19, no. 2, pp. 169-174. 
  7. Gun-Ha Kang, Jung-Mo Sohn, & Gun-Wu Sim, "Comparative Analysis of Anomaly Detection Models using AE and Suggestion of Criteria for Determining Outliers," Journal of the Korea Society of Computer and Information , vol. 26, no. 8, pp. 23-30, 2019. 
  8. Kulanuwat, L.; Chantrapornchai, C.; Maleewong, M.; Wongchaisuwat, P.; Wimala, S.; Sarinnapakorn, K.; Boonya-aroonnet, S, "Anomaly Detection Using a Sliding Window Technique and Data Imputation with Machine Learning for Hydrological Time Series," Water, vol. 13, no.13, 2021. 
  9. Medium, "Time series anomaly detection - with Python example," https://medium.com/@krzysztofdrelczuk,(accessed Nov., 27, 2023). 
  10. Jeong-Hwan Hwang , JU HUI DONG 2,Ho-cheol Lee, Joong hyeon Gang, YOE HYUN, Hoon-Meong Lee, "Implementation of Swinery Integrated Management System under Ubiquitous Agricultural Environments," The Journal of Korean Institute of Communications and Information Sciences, Vol. 26 No. 8, pp. 23-30, 2010. 
  11. Mi-aye Yang, Won-Ho Nam, Han-Joong Kim, Taegon Kim, An-Kook Shin, Mun-Sung, Kang, "Anomaly Detection in Reservoir Water Level Data Using the LSTM Model Based on Deep Learning," Korean Society of Hazard Mitigation, vol. 21, no.1, pp. 71-81, 2021. 
  12. Fundamentals of Recurrent Neural Network (RNN)\ and Long Short-Term Memory (LSTM) Network, Alex Sherstinsky, arXiv:1808.03314v10.  https://doi.org/10.1016/j.physd.2019.132306
  13. Jin Ho Kang, & Myung Hwan Na, "Prediction of pig weight using breed, sex, and age with LSTM model", Korean Institute of Electrical Engineers (KIEE) Conference, pp. 2247-2249, 2023. 
  14. Seng-Kyoun Jow, Se-Han Ki, "Digital Twin-Based Smart Livestock House- Case Study," The Journal of Korean Institute of Communications and Information Sciences Vol. 45, No. 8, pp. 1472-1481, Aug. 2020. 
  15. Cheol Won Le ,Su-yong An, Jae-Young Kim Hyeong-tae Ahn, "Sensor anomaly detection system in greenhouse-type smart farm using environmental data," Journal of the Korean Data & Information Science Society, vol/ 32, no. 6, pp.1237-1248, 2021. 
  16. 농어촌공사, 수위계측 중앙값 모델 활용 정확도 높인다(2021). https://www.paxetv.com/news/articleView.html?idxno=114974 (accessed Nov., 02, 2023).