• 한국항행학회논문지
• /
• 제27권1호
• /
• pp.111-118
• /
• 2023

스마트팩토리의 구축을 위해서는 제조환경에서 여러 센서 및 기기 등을 연결하여 데이터를 수집하고, 데이터 분석을 통해 생산설비 등의 장애를 진단하거나 예측하여야 한다. 본 논문에서는 공작기계에서 제품을 가공하기 위해 사용되는 절삭용 인서트의 잔여 유효 수명을 예측하기 위해 진동 신호를 기반으로 한 가중화 k-최근접이웃(Weighted k-NN) 알고리즘, 의사결정나무(Decision Tree), 서포트벡터회귀(SVM), XGBoost, 랜덤포레스트(Random forest), 1차원 합성곱신경망(1D-CNN), 그리고 진동 신호를 FFT한 주파수 스펙트럼에 대해 알아보았다. 연구결과, 주파수 스펙트럼으로는 잔여 유효수명의 정확한 예측에 대해서는 신빙성있는 기준을 제공하지 못한다는 것을 알수 있었고, 예측 모델 중 가중화 k-최근접이웃 알고리즘이 MAE가 0.0013, MSE가 0.004, RMSE가 0.0192로 가장 우수한 성능을 나타내었다. 이는 가중화 k-최근접이웃 알고리즘에 의해 예측되는 인서트의 잔여 유효 수명의 오차가 0.001초 수준으로 평가되어, 실제 산업현장에 적용이 가능한 수준으로 사료된다.

• 한국산학기술학회논문지
• /
• 제22권4호
• /
• pp.611-620
• /
• 2021

정비 산업은 사후정비, 예방정비를 거쳐, 상태기반 정비를 중심으로 진행되고 있다. 상태기반 정비는 장비의 상태를 파악하여, 최적 시점에서의 정비를 수행한다. 최적의 정비 시점을 찾기 위해서는 장비의 상태, 즉 잔여 유효 수명을 정확하게 파악하는 것이 중요하다. 이에, 본 논문은 시뮬레이션 데이터(C-MAPSS)를 사용한 터보팬 엔진의 잔여 유효수명(RUL, Remaining Useful Life) 예측 모델을 제시한다. 모델링을 위해 C-MAPSS(Commercial Modular Aero-Propulsion System Simulation) 데이터를 전처리, 변환, 예측하는 과정을 거쳤다. RUL 임계값 설정, 이동평균필터 및 표준화를 통해 데이터 전처리를 수행하였고, 주성분 분석(Principal Component Analysis)과 k-NN(k-Nearest Neighbor)을 활용하여 잔여 유효 수명을 예측하였다. 최적의 성능을 도출하기 위해, 5겹 교차검증기법을 통해 최적의 주성분 개수 및 k-NN의 근접 데이터 개수를 결정하였다. 또한, 사전 예측의 유용성, 사후 예측의 부적합성을 고려한 스코어링 함수(Scoring Function)를 통해 예측 결과를 분석하였다. 마지막으로, 현재까지 제시되어온 뉴럴 네트워크 기반의 알고리즘과 예측 성능 비교 및 분석을 통해 k-NN 활용 모델의 유용성을 검증하였다.

• 한국기계가공학회지
• /
• 제19권12호
• /
• pp.21-27
• /
• 2020

Lithium-ion batteries are the heart of energy-storing devices and electric vehicles. Owing to their superior qualities, such as high capacity and energy efficiency, they have become quite popular, resulting in an increased demand for failure/damage prevention and useable life maximization. To prevent failure in Lithium-ion batteries, improve their reliability, and ensure productivity, prognosticative measures such as condition monitoring through sensors, condition assessment for failure detection, and remaining useful life prediction through data-driven prognostics and health management approaches have become important topics for research. In this study, the residual useful life of Lithium-ion batteries was predicted using two efficient artificial recurrent neural networks-ong short-term memory (LSTM) and gated recurrent unit (GRU). The proposed approaches were compared for prognostics accuracy and cost-efficiency. It was determined that LSTM showed slightly higher accuracy, whereas GRUs have a computational advantage.

• 산업경영시스템학회지
• /
• 제45권3호
• /
• pp.18-30
• /
• 2022

Predicting remaining useful life (RUL) becomes significant to implement prognostics and health management of industrial systems. The relevant studies have contributed to creating RUL prediction models and validating their acceptable performance; however, they are confined to drive reasonable preventive maintenance strategies derived from and connected with such predictive models. This paper proposes a data-driven preventive maintenance method that predicts RUL of industrial systems and determines the optimal replacement time intervals to lead to cost minimization in preventive maintenance. The proposed method comprises: (1) generating RUL prediction models through learning historical process data by using machine learning techniques including random forest and extreme gradient boosting, and (2) applying the system failure time derived from the RUL prediction models to the Weibull distribution-based minimum-repair block replacement model for finding the cost-optimal block replacement time. The paper includes a case study to demonstrate the feasibility of the proposed method using an open dataset, wherein sensor data are generated and recorded from turbofan engine systems.

• 전력전자학회논문지
• /
• 제25권4호
• /
• pp.251-260
• /
• 2020

A growing interest has emerged in recycling used automobile batteries into energy storage systems (ESSs) to prevent their harmful effects to the environment from improper disposal and to recycle such resources. To transform used batteries into ESSs, composing battery modules with similar performance by grading them is crucial. Imbalance among battery modules degrades the performance of an entire system. Thus, the selection of modules with similar performance and remaining life is the first prerequisite in the reuse of used batteries. In this study, we develop an instrument to measure the impedance spectrum of a battery module to predict the useful remaining life of the used battery. The developed hardware and software are used to apply the AC perturbation to the used battery module and measure its impedance spectrum. The developed instrument can measure the impedance spectrum of the battery module from 0.1 Hz to 1 kHz and calculate the equivalent circuit parameters through curve fitting. The performance of the developed instrument is verified by comparing the measured impedance spectra with those obtained by a commercial equipment.

• 산업경영시스템학회지
• /
• 제44권4호
• /
• pp.1-11
• /
• 2021

Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.