In recent automated manufacturing systems, compressed air-based pneumatic cylinders have been widely used for basic perpetration including picking up and moving a target object. They are relatively categorized as small machines, but many linear or rotary cylinders play an important role in discrete manufacturing systems. Therefore, sudden operation stop or interruption due to a fault occurrence in pneumatic cylinders leads to a decrease in repair costs and production and even threatens the safety of workers. In this regard, this study proposed a fault detection technique by developing a time-variant deep learning model from multivariate sensor data analysis for estimating a current health state as four levels. In addition, it aims to establish a real-time fault detection system that allows workers to immediately identify and manage the cylinder’s status in either an actual shop floor or a remote management situation. To validate and verify the performance of the proposed system, we collected multivariate sensor signals from a rotary cylinder and it was successful in detecting the health state of the pneumatic cylinder with four severity levels. Furthermore, the optimal sensor location and signal type were analyzed through statistical inferences.
PURPOSES : This study develops a model that can estimate travel speed of each movement flow using deep-learning-based probe vehicles at urban intersections. METHODS : Current technologies cannot determine average travel speeds for all vehicles passing through a specific real-world area under obseravation. A virtual simulation environment was established to collect information on all vehicles. A model estimate turning speeds was developed by deep learning using probe vehicles sampled during information processing time. The speed estimation model was divided into straight and left-turn models, developed as fully-offset, non-offset, and integrated models. RESULTS : For fully-offset models, speed estimation for both straight and left-turn models achieved MAPE within 10%. For non-offset models, straight models using data drawn from four or more probe vehicles achieved a MAPE of less than 15%. The MAPE for left turns was approximately 20%. CONCLUSIONS : Using probe-vehicle data(PVD), a deep learning model was developed to estimate speeds each movement flow. This, confirmed the viability of real-time signal control information processing using a small number of probe vehicles.
기후변화 영향으로 이상고수온, 태풍, 홍수, 가뭄 등 재난 및 안전 관리기술은 지속적으로 고도화를 요구받고 있으며, 특히 해 수면 온도는 한반도 주변에서 발생되는 여름철 적조 발생과 동해안 냉수대 출현, 소멸 등에 영향을 신속하게 분석할 수 있는 중요한 인자 이다. 따라서, 본 연구에서는 해수면 온도 자료를 해양 이상현상 및 연구에 적극 활용되기 위해 통계적 방법과 딥러닝 알고리즘을 적용하 여 예측성능을 평가하였다. 예측에 사용된 해수면 수온자료는 흑산도 조위관측소의 2018년부터 2022년까지 자료이며, 기존 통계적 ARIMA 방법과 Long Short-Term Memory(LSTM), Gated Recurrent Unit(GRU)을 사용하였고, LSTM의 성능을 더욱 향상할 수 있는 Sequence-to-Sequence(s2s) 구조에 Attention 기법을 추가한 Attention Long Short-Term Memory (LSTM)기법을 사용하여 예측 성능 평가를 진행하 였다. 평가 결과 Attention LSTM 모델이 타 모델과 비교하여 더 좋은 성능을 보였으며, Hyper parameter 튜닝을 통해 해수면 수온 성능을 개 선할 수 있었다.
On pig farms, the highest mortality rate is observed among nursing piglets. To reduce this mortality rate, farmers need to carefully observe the piglets to prevent accidents such as being crushed and to maintain a proper body temperature. However, observing a large number of pigs individually can be challenging for farmers. Therefore, our aim was to detect the behavior of piglets and sows in real-time using deep learning models, such as YOLOv4-CSP and YOLOv7-E6E, that allow for real-time object detection. YOLOv4-CSP reduces computational cost by partitioning feature maps and utilizing Cross-stage Hierarchy to remove redundant gradient calculation. YOLOv7-E6E analyzes and controls gradient paths such that the weights of each layer learn diverse features. We detected standing, sitting, and lying behaviors in sows and lactating and starving behaviors in piglets, which indicate nursing behavior and movement to colder areas away from the group. We optimized the model parameters for the best object detection and improved reliability by acquiring data through experts. We conducted object detection for the five different behaviors. The YOLOv4-CSP model achieved an accuracy of 0.63 and mAP of 0.662, whereas the YOLOv7-E6E model showed an accuracy of 0.65 and mAP of 0.637. Therefore, based on mAP, which includes both class and localization performance, YOLOv4-CSP showed the superior performance. Such research is anticipated to be effectively utilized for the behavioral analysis of fattening pigs and in preventing piglet crushing in the future.
Elevators are the main means of transport in buildings. A malfunction of an elevator in operation may cause in convenience to users. Furthermore, fatal accidents, such as injuries and death, may occur to the passengers also. Therefore, it is important to prevent failure before accidents happen. In related studies, preventive measures are proposed through analyzing failures, and the lifespan of elevator components. However, these methods are limited to existing an elevator model and its surroundings, including operating conditions and installed environments. Vibration occurs when the elevator is operated. Experts have classified types of faults, which are symptoms for malfunctions (failures), via analyzing vibration. This study proposes an artificial intelligent model for classifying faults automatically with deep learning algorithms through elevator vibration data, hereby preventing failures before they occur. In this study, the vibration data of six elevators are collected. The proposed methodology in this paper removes "the measurement error data" with incorrect measurements and extracts operating sections from the input datasets for proceeding deep learning models. As a result of comparing the performance of training five deep learning models, the maximum performance indicates Accuracy 97% and F1 Score 97%, respectively. This paper presents an artificial intelligent model for detecting elevator fault automatically. The users’ safety and convenience may increase by detecting fault prior to the fatal malfunctions. In addition, it is possible to reduce manpower and time by assisting experts who have previously classified faults.
해운 시황을 예측하는 것은 중요한 문제이다. 투자 방식의 결정, 선대 편성 방법, 운임 등을 결정하기 위한 판단 근거가 되며 이는 기업의 이익과 생존에 큰 영향을 미치기 때문이다. 이를 위해 본 연구에서는 기계학습 모델인 장단기 메모리 및 간소화된 장단기 메모리 구조의 Gated Recurrent Units를 활용하여 컨테이너선의 해상운임 예측 모델을 제안한다. 운임 예측 대상은 중국 컨테이너 운임지수 (CCFI)이며, 2003년 3월부터 2020년 5월까지의 CCFI 데이터를 학습에 사용하였다. 각 모델에 따라 2020년 6월 이후의 CCFI를 예측한 후 실 제 CCFI와 비교, 분석하였다. 실험 모델은 하이퍼 파라메터의 설정에 따라 총 6개의 모델을 설계하였다. 또한 전통적인 분석 방법과의 성 능을 비교하기 위해 ARIMA 모델도 실험에 추가하였다. 최적 모델은 두 가지 방법에 따라 선정하였다. 첫 번째 방법으로 각 모델을 10회 반복 실험하여 얻은 RMSE의 평균값이 가장 작은 모델을 선정하는 것이다. 두 번째 방법으로는 모든 실험에서 가장 낮은 RMSE를 기록한 모델을 선정하는 것이다. 실험 결과 전통적 시계열 예측모델인 ARIMA 모델과 비교하여 딥러닝 모델의 정확도를 입증하였으며, 정확한 예측모델을 통해 운임 변동의 위험관리 능력을 제고시키는데 기여했다. 반면 코로나19와 같은 외부 효과에 따른 운임의 급격한 변화상황이 발생한 경우, 예측모델의 정확도가 감소하는 한계점을 나타냈다. 제안된 모델 중 GRU1 모델이 두 가지 평가 방법 모두에서 가장 낮은 RMSE(69.55, 49.35)를 기록하며 최적 모델로 선정되었다.
Deep learning, which has recently shown excellent performance, has a problem that the amount of computation and required memory are large. Model compression is very useful because it saves memory and reduces storage size while maintaining model performance. Model compression methods reduce the number of edges by pruning weights that are deemed unnecessary in the calculation. Existing weight pruning methods using ADMM construct an optimization problem by a layer-by-layer addition of pre-defined removal-ratio constraints. Decomposing into two subproblems through the ADMM process, one can solve them through gradient descent and projection. However, the layer-by-layer removal ratios must be structurally specified, causing a sharp increase in training time due to a large number of parameters, and hardly feasible to use for large models that actually require weight pruning. Our proposed method performs weight pruning, producing similar performance, by setting a global removal ratio for the entire model without prior knowledge of structural characteristics in order to solve the shortcomings of the existing ADMM weight-pruning methods. To effectively avoid performance degradation, the method removes a relatively small number of previous layers in charge of feature extraction. Experiments show high-quality performance, not necessarily setting layer-by-layer removal ratios. Additionally, experiments increasing layers yield an insight for feature extraction in pruned layers. The experiment of the proposed method to the LeNet-5 model using MNIST data results in a higher compression ratio of 99.3% outperforming those of other existing algorithms. We also demonstrate the effectiveness of the proposed method in YOLOv4, an object detection model requiring substantial computation.
PURPOSES : This study uses deep learning image classification models and vehicle-mounted cameras to detect types of pavement distress — such as potholes, spalling, punch-outs, and patching damage — which require urgent maintenance.
METHODS : For the automatic detection of pavement distress, the optimal mount location on a vehicle for a regular action camera was first determined. Using the orthogonal projection of obliquely captured surface images, morphological operations, and multi-blob image processing, candidate distressed pavement images were extracted from road surface images of a 16,036 km in-lane distance. Next, the distressed pavement images classified by experts were trained and tested for evaluation by three deep learning convolutional neural network (CNN) models: GoogLeNet, AlexNet, and VGGNet. The CNN models were image classification tools used to identify and extract the combined features of the target images via deep layers. Here, a data augmentation technique was applied to produce big distress data for training. Third, the dimensions of the detected distressed pavement patches were computed to estimate the quantity of repair materials needed.
RESULTS : It was found that installing cameras 1.8 m above the ground on the exterior rear of the vehicle could provide clear pavement surface images with a resolution of 1 cm per pixel. The sensitivity analysis results of the trained GoogLeNet, AlexNet, and VGGNet models were 93 %, 86 %, and 72 %, respectively, compared to 62.7 % for the dimensional computation. Following readjustment of the image categories in the GoogLeNet model, distress detection sensitivity increased to 94.6 %.
CONCLUSIONS : These findings support urgent maintenance by sending the detected distressed pavement images with the dimensions of the distressed patches and GPS coordinates to local maintenance offices in real-time.
Deep learning models, which imitate the function of human brain, have drawn attention from many engineering fields (mechanical, agricultural, and computer engineering etc). The major advantages of deep learning in engineering fields can be summarized by objects detection, classification, and time-series prediction. As well, it has been applied into environmental science and engineering fields. Here, we compiled our previous attempts to apply deep learning models in water-environment field and presented the future opportunities.