DOI QR코드

DOI QR Code

트리 기반 부스팅 알고리듬을 이용한 상수도관 누수 탐지 방법

Leakage Detection Method in Water Pipe using Tree-based Boosting Algorithm

  • 이재흥 (한밭대학교 컴퓨터공학과) ;
  • 오윤성 (한밭대학교 컴퓨터공학과) ;
  • 민준혁 (한밭대학교 컴퓨터공학과)
  • Jae-Heung Lee (Department of Computer Engineering, Hanbat National University) ;
  • Yunsung Oh (Department of Computer Engineering, Hanbat National University) ;
  • Junhyeok Min (Department of Computer Engineering, Hanbat National University)
  • 투고 : 2024.03.04
  • 심사 : 2024.03.29
  • 발행 : 2024.04.30

초록

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.

Losses in domestic water supply due to leaks are very large, such as fractures and defects in pipelines. Therefore, preventive measures to prevent water leakage are necessary. We propose the development of a leakage detection sensor utilizing vibration sensors and present an optimal leakage detection algorithm leveraging artificial intelligence. Vibrational sound data acquired from water pipelines undergo a preprocessing stage using FFT (Fast Fourier Transform), followed by leakage classification using an optimized tree-based boosting algorithm. Applying this method to approximately 260,000 experimental data points from various real-world scenarios resulted in a 97% accuracy, a 4% improvement over existing SVM(Support Vector Machine) methods. The processing speed also increased approximately 80 times, confirming its suitability for edge device applications.

키워드

과제정보

본논문은 산업통상자원부 및 한국산업기술기획평가원(KEIT) 연구비 지원에 의한 연구임.

참고문헌

  1. National Water Supply Information System, National water quantity management, https://www.waternow.go.kr/web/ssdoData/?pMENUID=8&ATTR_1=2011&ATTR_5=4
  2. Seung-Heon Oh, Seung-hee Park, Ho-sung Kim, and Jong-rip Kim, "Implementation of GAN-based Water Pipeline Leakage Monitoring System," Korea Institute of information and Communication Engineering, Vol. 27, No. 1, pp. 32-35, 2023.
  3. Lee, Chan Wook and Yoo, Do Guen, "Development of leakage detection model in water distribution networks applying LSTM-based deep learning algorithm," Journal of Korea Water Resources Association, Vol. 54, No. 8, pp. 599-606, 2021. https://doi.org/10.3741/JKWRA.2021.54.8.599
  4. Jae-Moon Hwang, Ho-Hyun Lee, Gang-Wook Shin and Nam Kim. "Leakage Detection Prediction by Neuro-Fuzzy and WECR in Water Distribution Network.," Journal of Korean Institute of Intelligent Systems, Vol. 27, No. 4, pp. 349-356, 2017. https://doi.org/10.5391/JKIIS.2017.27.4.349
  5. Poulakis, Z., Dimitris Valougeorgis, and Costas Papadimitriou, "Leakage detection in water pipe networks using a Bayesian probabilistic framework." Probabilistic Engineering Mechanics Vol. 18, No. 4, pp. 315-327, 2003. https://doi.org/10.1016/S0266-8920(03)00045-6
  6. Xiaoqin Li, Xiaomei Wu, Mingzhuang Sun, Shengqiao Yang and Weikun Song, "A Novel Intelligent Leakage Monitoring-Warning System for Sustainable Rural Drinking Water Supply," Sustainability Vol. 14, No. 10, 2022.
  7. Wei-Yi Chuang, Yao-Long Tsai, and Li-Hua Wang, "Leak Detection in Water Distribution Pipes Based on CNN with Mel Frequency Cepstral Coefficients," In Proceedings of the 2019 3rd International Conference on Innovation in Artificial Intelligence (ICIAI '19). Association for Computing Machinery, New York, NY, USA, pp. 83-86, 2019
  8. Youngmin Seo, Kwanghyun Choi, Yuseong Lim, Byungjoon Lee and Yunyoung Choi "Application of Machine Learning Models for Water Pipeline Leakage Detection," Crisisonomy, Vol. 19 No. 4, pp. 45-54, 2023. https://doi.org/10.14251/crisisonomy.2023.19.4.45
  9. Jungyu Choi and Sungbin Im, "Leak Detection and Classification of Water Pipeline based on SVM using Leakage Noise Magnitude Spectrum," Journal of the Institute of Electronics and Information Engineers, Vol. 60, No. 2, pp. 6-14, 2023. https://doi.org/10.5573/ieie.2023.60.2.6
  10. Thambirajah Ravichandran, Keyhan Gavahi, Kumaraswamy Ponnambalam, Valentin Burtea and S. Jamshid Mousavi, "Ensemble-based machine learning approach for improved leak detection in water mains," Journal of Hydroinformatics, Vol. 23, No. 2, pp. 307-323, 2021. https://doi.org/10.2166/hydro.2021.093
  11. Quinlan, J. R. "Induction of Decision Trees," Mach Learn, Vol. 1, pp. 81-106, 1985. https://doi.org/10.1007/BF00116251
  12. Tin Kam Ho, "Random decision forests," Proceedings of 3rd International Conference on Document Analysis and Recognition, Montreal, QC, Canada, Vol. 1, pp. 278-282, 1995.
  13. Tianqi Chen, Hang Li, Qiang Yang and Yong Yu "General Functional Matrix Factorization Using Gradient Boosting," Proceedings of the 30th International Conference on Machine Learning, PMLR Vol. 28, No. 1, pp. 436-444, 2013.
  14. Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye and Tie-Yan Liu, "LightGBM: A Highly Efficient Gradient Boosting Decision Tree," In Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS'17), Curran Associates Inc, Red Hook, NY, USA, pp. 3149-3157, 2017.
  15. Tianqi Chen and Carlos Guestrin, "XGBoost: A Scalable Tree Boosting System," In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). Association for Computing Machinery, New York, NY, USA, pp. 785-794, 2016.