ANN 알고리즘을 통한 절단 가공의 실시간 표면거칠기 예측에 관한 연구 = A Study on the Real-Time Prediction of the Surface Roughness during the Cutting Process Using an ANN Algorithm

다국어 초록 (Multilingual Abstract)

Cutting is the first step and key processing factor during the manufacturing of shipbuilding equipment. Recently, various environmental problems have emerged in cutting workshops owing to the increase in the importance of safety in shipyards during shipbuilding, which has created the need for high-quality and high-efficiency processing methods in the industry. In this study, a surface roughness prediction model was developed using real-time data obtained during the plasma cutting process of a A106 B pipe. Subsequently, a surface roughness prediction model was developed based on cutting process variables such as current and cutting speed using an ANN. The mean square error (MSE) of the ANN prediction model was determined based on the error and prediction performance. Additionally, the accuracy of the model was evaluated through an analysis of the surface roughness prediction model using the experimental value and predictive ability of the model (PAM), where a high prediction performance of 98.69% was obtained.

국문 초록 (Abstract)

조선 기자재 제작 공정의 첫 단계이자 핵심 가공 요인은 절단이라 볼 수 있다. 최근 조선소의 선박건조에 있어서 안전의 중요성이 고조되면서 절단 작업장의 환경문제가 크게 대두되고 고품질 및 고능률 가공법이 산업계를 중심으로 요구되고 있다. 본 연구에서는 A106 B 재질의 pipe를 사용하여 플라즈마 절단공정시 획득한 실시간(real-time) 데이터를 활용하여 표면 거칠기 예측모델 개발을 목표로 하였다. 절단실험의 절단 공정변수인 전류 및 절단 속도를 선정하여 ANN을 통하여 절단 공정변수에 따른 표면거칠기 예측 모델을 개발하였다. ANN 예측모델을 오차 및 예측 성능을 기반으로 평균 제곱 오차(MSE)를 비교하였다. 또한 표면 거칠기 예측모델의 분석을 통해 실험값과 PAM(predictive ability of model)을 이용하여 정확도를 평가하여 98.69%로 높은 예측성능을 확인하였다.

참고문헌 (Reference)

1 이충우 ; 윤태종 ; 오원빈 ; 이보람 ; 김영수 ; 박민우 ; 박동환 ; 김일수, "플라즈마 절단공정에서 ELM을 이용한 품질 예측 모델에 관한 연구" 대한기계학회 45 (45): 961-967, 2021

2 박종경 ; 장태우, "스마트제조 기술 국내연구 고찰" 한국전자거래학회 23 (23): 123-133, 2018

3 김성일, "선박용 Al 합금의 레이저 절단시 표면 특성에 관한 연구" 한국생산제조학회 18 (18): 529-535, 2009

4 Nemchinsky, V. A, "What We Know and What We Do Not Know about Plasma Arc Cutting" 39 (39): 2006

5 Lee, G, "Tool Condition Monitoring Based on Wavelet Transform" Sogang University 2003

6 Naik, D. K, "Temperature Analysis by Moving Heat Source during Plasma Arc Cutting Process: An Analytical Approach" 16 (16): 569-572, 2019

7 Agarwal, P. H, "Optimizing Plasma Arc Cutting Parameters for Structural Steel Using grey Relational Analysis" 8 (8): 27-30, 2019

8 Lida, K, "Observation System of Molten Steel Begavior on Plasma Arc Cutting Surface" 15 (15): 796-801, 2020

9 Lasi, H, "Industry 4.0" 6 (6): 239-242, 2014

10 Xu, X, "Industrial Big Data Analysis in Smart Factory: Current Status and Research Strategies" 5 : 17543-17551, 2017

11 Mitra, M. S, "Improved Plasma Arc Cutting Systems, Consumables and Operational Methods, Patent No. WO20170-24149"

12 Nemchinsky, V, "Handbook of Thermal Science and Engineering" 1-62, 2017

13 Maeng, M. J, "Fracture Detection of Milling Cutter Using Cutting Force and Acoustic Emission Signals" 3 (3): 28-37, 2004

14 이충우 ; 윤태종 ; 오원빈 ; 이보람 ; 김영수 ; 김일수, "Elman 신경망을 이용한 절단면 표면거칠기 예측에 관한 연구" 대한기계학회 45 (45): 567-572, 2021

15 Manburu, P. K, "Electrode for a Plasma Arc Cutting Torch, Patent No. EP0772957B1"

16 Wan, J, "Artificial Intelligence for Cloud-Assisted Smart Factory" 6 : 55419-55430, 2018

17 Seo, J. H, "An Experimental Study on the Tool Failure Detection in the Machining by Face Milling" 12 (12): 92-100, 1995

18 Lee, M. H, "Aerodymanic Diameter Distribution of Aerosols from Plasma Arc Cutting for Steels at Different Cutting Power Levels" 323 (323): 613-624, 2020

19 Um, K. W, "A Study on the Plasma Arc Cutting Phenomena of Plate Materials" 9 (9): 69-74, 1991

20 Nho, T. J, "A Study on Development of Plasma-Arc Cutting System with Computer-Numerical Control" 8 (8): 60-69, 1990

21 Jeong, C. H, "A Research on Fault Diagnosis of Pipeline Using Acoustic Emission Signal Data and Machine Learning Algorithm" University of Ulsan 2020

22 Lee, E. M, "4th Industrial Revolution and Changes in Industrial Structure" 28 (28): 1-22, 2016