바로가기메뉴

본문 바로가기 주메뉴 바로가기

ACOMS+ 및 학술지 리포지터리 설명회

  • 한국과학기술정보연구원(KISTI) 서울분원 대회의실(별관 3층)
  • 2024년 07월 03일(수) 13:30
 

AI기반 상수도시설 개량 의사결정 모델 분석

Model Analysis of AI-Based Water Pipeline Improved Decision

한국사물인터넷학회논문지 / Journal of The Korea Internet of Things Society, (P)2466-0078;
2022, v.8 no.5, pp.11-16
https://doi.org/https://doi.org/10.20465/kiots.2022.8.5.011
김기태 (목원대)
민병원 (목원대학교)
오용선 (목원대학교)
  • 다운로드 수
  • 조회수

초록

상수도분야 인공지능 기술개발 관심도가 증가함에 따라 상수도 관로에 대해서 노후관 상태평가 데이터 결과를활용하여 반복적인 학습으로 개량 의사결정 등급을 예측할 수 있는 인공신경망 알고리즘을 개발하고 검증과정을 통하여가장 신뢰성 있는 예측 모델을 제시하고자 한다. 2020년 한강유역의 노후관로 정비 기본계획에 의한 간접평가 데이터12개 항목을 기반으로 데이터 전처리 하고 인공신경망 알고리즘을 적용하여 반복학습과 검증을 통해 계산된 결과값과직접평가 결과값의 일치율이 90% 이상이 되도록 역전파 과정을 통해 가중치를 업데이트 하면서 최적화하여 관로 등급을 예측하는 알고리즘을 개발하였다. 알고리즘 정확도 검증결과 모든 관종 데이터가 고르게 분포되어 있고 학습 데이터가 많아야 예측평가 정확도가 높아지는 것을 확인할 수 있었다. 향후 전국의 다양한 데이터가 확보되면 인공신경망을이용한 관로등급 예측의 신뢰도가 좀 더 향상되어 객관화된 노후관 상태평가 의사결정 지원 역할을 수행할 수 있을것으로 기대된다.

keywords
Artificial Intelligence Technology, Data Analysis, Aging Pipe Condition Evaluation, Algorithm Model, Indirect Evaluation, Direct Evaluation, 인공지능, 데이터 분석, 노후관 상태평가, 알고리즘 모델, 간접평가, 직접평가

Abstract

As an interest in the development of artificial intelligence(AI) technology in the water supply sector increases, we have developed an AI algorithm that can predict improvement decision-making ratings through repetitive learning using the data of pipe condition evaluation results, and present the most reliable prediction model through a verification process. We have developed the algorithm that can predict pipe ratings by pre-processing 12 indirect evaluation items based on the 2020 Han River Basin's basic plan and applying the AI algorithm to update weighting factors through backpropagation. This method ensured that the concordance rate between the direct evaluation result value and the calculated result value through repetitive learning and verification was more than 90%. As a result of the algorithm accuracy verification process, it was confirmed that all water pipe type data were evenly distributed, and the more learning data, the higher prediction accuracy. If data from all across the country is collected, the reliability of the prediction technique for pipe ratings using AI algorithm will be improved, and therefore, it is expected that the AI algorithm will play a role in supporting decision-making in the objective evaluation of the condition of aging pipes.

keywords
Artificial Intelligence Technology, Data Analysis, Aging Pipe Condition Evaluation, Algorithm Model, Indirect Evaluation, Direct Evaluation, 인공지능, 데이터 분석, 노후관 상태평가, 알고리즘 모델, 간접평가, 직접평가

참고문헌

1.

“Waterworks Standards”, Korea Water and Wastewater Association, p2, 2010.

2.

S M Lee and D S Kang, “Water pipe deterioration assessment using ANN-Clustering”, Journal of the Korean Water Resources Association, pp959-960, 2018

3.

Y R Choi and K H Kim, “Artificial Intelligence Overview and Application Cases”, ie Magazine , p23, 2016.

4.

S B Cho and J H Kim, “Computational Complexity of Artificial Neural Networks”, Korean Institute of Information Scientists and Engineers, pp.315-318, 1989.

5.

Y I Cho1, M J Oh, Y S Seok, S J Lee and M I Roh, “Resistance Estimation of a Ship in the Initial Hull Design Using Deep Learning”, Korean Journal of Computational Design and Engineering, pp.203-210, 2019.

6.

Cybenko, George. "Approximation by Superpositions of a Sigmoidal Function." Mathematics of Control, Signals, and Systems, pp.303-314, 1989.

7.

S R Jo, H N Sung and B H Ahn, “A Comparative Study on the Performance of SVM and an Artificial Neural Network in Intrusion Detection”, Journal of the Korea Academia-Industrial cooperation Society, pp.703-711, 2016.

8.

C W Lee, S J Hong, J S Park and D G Yoo, “Assessment of Water Supply Pipes Deterioration Based on Deep Learning”, Journal of the Korean Society of Civil Engineers, pp.62-63, 2020.

9.

“Water Pipeline Precision Investigation Manual(Draft)”, Ministry of Environment, pp.10-24, 2020.

10.

K H Han, Y J Ryu, T S Kim and J H Heo, “Input Variables Selection of Artificial Neural Network Using Mutual Information”, Korea Water Resources Assocition, pp.85-87, 2010.

11.

H K Park, J S Lee, W H Song and S S Park, “Technology Forecasting using Patent Outlier Remova”, Journal of the Korean Institute of Intelligent Systems, pp.177-184, 2018.

12.

Y M Go, S W Go, “Alleviation of Vanishing Gradient Problem Using Parametric Activation Functions”Journal of the Korea Information Processing Society, pp.407-420, 2021.

13.

“A Report on the Schematic Plan for the Improvement of the Aged Water Pipeline”, Korea Water Resources Corporation, 2021.

14.

J Y Seo, “Deep Learning TensorFlow Textbook”, Gilbut, p.25, 2021.

15.

S M An, “Deep learning architectures and applications”Journal of the Korea Intelligent Information Systems Society, pp.127-142, 2016.

한국사물인터넷학회논문지