• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.12 s.177
• /
• pp.149-155
• /
• 2005

The rotating components such as turbine rotors in service are generally subjected to multiaxial cyclic loading conditions. The prediction of fatigue lift for turbine rotor components under complex multiaxial loading conditions is very important to prevent the fatigue failures in service. In this paper, axial and torsional low cycle fatigue tests were preformed for 3.5NiCrMo steels serviced low pressure turbine rotor of nuclear power plant. Several methods to predict biaxial fatigue life such as Tresca, von Mises and Brown & Miller's critical plane approach were evaluated to correlate the experimental results for serviced NiCrMoV steel. The fracture mode and fatigue characteristics of NiCrMoV steel were discussed based on the results of fatigue tests performed under the axial and torsional test conditions. In particular, the Brown and Miller's critical plane approach was found to best correlate the experimental data with predictions being within a factor of 2.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.46 no.8
• /
• pp.671-678
• /
• 2018

Helicopter noise prediction is an essential process for developing low noise helicopter technology. In this paper, the noise prediction method is developed using the helicopter integrated performance analysis program CAMRAD-II and in-house noise analysis code. In addition, the analytical technique was verified by analyzing blade-vortex interaction noise, which is the biggest cause of helicopter noise. In order to predict the actual helicopter noise, the noise analysis was performed for the flyover and approach condition, which is the standard measurement condition of the International Civil Aviation Organization (ICAO). Finally, we confirmed the suitability of the analytical method through comparison and analysis with the flight test results.

• Journal of Intelligence and Information Systems
• /
• v.7 no.2
• /
• pp.19-33
• /
• 2001

Link weight analysis approach is suggested as a heuristic for selection of input nodes in artificial neural network for bankruptcy prediction. That is to analyze each input node\\\\`s link weight-absolute value of link weight between an input node and a hidden node in a well-trained neural network model. Prediction accuracy of three methods in this approach, -weak-linked-neurons elimination method, strong-linked-neurons selection method and integrated link weight model-is compared with that of decision tree and multivariate discrimination analysis. In result, the methods suggested in this study show higher accuracy than decision tree and multivariate discrimination analysis. Especially an integrated model has much higher accuracy than any individual models.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.41 no.4
• /
• pp.13-20
• /
• 2004

In this paper, we developed a method for predicting the tracking performance of the multiple hypothesis tracking (MHT) algorithm. The MHT algerithm is known to be a measurement-oriented optimal Bayesian approach and is superior to any other tracking filters because it takes into account the events that the measurements can be originated from new targets and false alarms as well as interesting targets. In the MHT algorithm, a number of candidate hypotheses are generated and evaluated later as more data are received. The probability of each candidate hypotheses is approximately evaluated by using the hybrid conditional average approach (HYCA). We performed numerical experiments to show the validity of our performance prediction.

• The KIPS Transactions:PartB
• /
• v.12B no.2 s.98
• /
• pp.209-214
• /
• 2005

The use of protein interaction data is highly reliable for predicting functions to proteins without function in proteomics study. The computational studies on protein function prediction are mostly based on the concept of guilt-by-association and utilize large-scale interaction map from revealed protein-protein interaction data. This study compares graph-based approaches such as neighbor-counting and $\chi^2-statistics$ methods using protein-protein interaction data and proposes an approach that is effective in analyzing large-scale protein interaction data. The proposed approach is also based protein interaction map but sequence similarity and heuristic knowledge to make prediction results more reliable. The test result of the proposed approach is given for KDD Cup 2001 competition data along with those of neighbor-counting and $\chi^2-statistics$ methods.

• Advances in nano research
• /
• v.13 no.5
• /
• pp.499-512
• /
• 2022

This study investigates the efficiency of ensemble machine learning for predicting the lightweight-aggregate concrete (LWC) characteristics. A stacking ensemble (STEN) approach was proposed to estimate the dry density (DD) and 28 days compressive strength (Fc-28) of LWC using two meta-models called random forest regressor (RFR) and extra tree regressor (ETR), and two novel ensemble models called STEN-RFR and STEN-ETR, were constructed. Four standalone machine learning models including artificial neural network, gradient boosting regression, K neighbor regression, and support vector regression were used to compare the performance of the proposed models. For this purpose, a sum of 140 LWC mixtures with 21 influencing parameters for producing LWC with a density less than 1000 kg/m³, were used. Based on the experimental results with multiple performance criteria, it can be concluded that the proposed STEN-ETR model can be used to estimate the DD and Fc-28 of LWC. Moreover, the STEN-ETR approach was found to be a significant technique in prediction DD and Fc-28 of LWC with minimal prediction error. In the validation phase, the accuracy of the proposed STEN-ETR model in predicting DD and Fc-28 was found to be 96.79% and 81.50%, respectively. In addition, the significance of cement, water-cement ratio, silica fume, and aggregate with expanded glass variables is efficient in modeling DD and Fc-28 of LWC.

• Advances in nano research
• /
• v.17 no.2
• /
• pp.109-124
• /
• 2024

In the era of smart manufacturing, precise prediction of springback-a common issue in ultra-thin sheet metal forming- and forming limits are critical for ensuring high-quality production and minimizing waste. This paper presents a novel approach that leverages the Internet of Things (IoT) and Artificial Neural Networks (ANN) to enhance springback and forming limits prediction accuracy. By integrating IoT-enabled sensors and devices, real-time data on material properties, forming conditions, and environmental factors are collected and transmitted to a central processing unit. This data serves as the input for an ANN model, which is trained with crystal plasticity simulations and experimental data to predict springback with high precision. Our proposed system not only provides continuous monitoring and adaptive learning capabilities but also facilitates real-time decision-making in manufacturing processes. Experimental results demonstrate significant improvements in prediction accuracy compared to traditional methods, highlighting the potential of IoT and ANN integration in advancing smart manufacturing. This approach promises to revolutionize quality control and operational efficiency in the industry, paving the way for more intelligent and responsive manufacturing systems.

• Journal of applied mathematics & informatics
• /
• v.28 no.1_2
• /
• pp.439-450
• /
• 2010

In this paper, we propose a new idea to evaluate the prediction accuracy of user's preference generated by memory-based collaborative filtering algorithm before prediction process in the recommender system. Our analysis results show the possibility of a pre-evaluation before the prediction process of users' preference of item's transaction on the web. Classification functions proposed in this study generate a user's rating pattern under certain conditions. In this research, we test whether classification functions select users who have lower prediction or higher prediction performance under collaborative filtering recommendation approach. The statistical test results will be based on the differences of the prediction accuracy of each user group which are classified by classification functions using the generative probability of specific rating. The characteristics of rating patterns of classified users will also be presented.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.118-126
• /
• 2024

In human conversations, listeners often utilize brief backchannels such as "uh-huh" or "yeah." Timely backchannels are crucial to understanding and increasing trust among conversational partners. In human-machine conversation systems, users can engage in natural conversations when a conversational agent generates backchannels like a human listener. We propose a method that simultaneously predicts backchannels and recognizes speech in real time. We use a streaming transformer and adopt multitask learning for concurrent backchannel prediction and speech recognition. The experimental results demonstrate the superior performance of our method compared with previous works while maintaining a similar single-task speech recognition performance. Owing to the extremely imbalanced training data distribution, the single-task backchannel prediction model fails to predict any of the backchannel categories, and the proposed multitask approach substantially enhances the backchannel prediction performance. Notably, in the streaming prediction scenario, the performance of backchannel prediction improves by up to 18.7% compared with existing methods.

• Asian-Australasian Journal of Animal Sciences
• /
• v.17 no.2
• /
• pp.228-236
• /
• 2004

Two varieties of alfalfa (Medicago sativa L cv. Pioneer and Beaver) and timothy (Phleum pratense L cv. Climax and Joliette), grown at different locations in Saskatchewan (Canada), were cut at three stages [1=one week before commercial cut (early bud for alfalfa; joint for timothy); 2=at commercial cut (late bud for alfalfa; pre-bloom head for timothy); 3=one week after commercial cut (early bloom for alfalfa; full head for timothy)]. The energy values of forages were determined using three approaches, including chemical (NRC 2001 formula) and biological approaches (standard in vitro and in situ assay). The objectives of this study were to determine the effects of forage variety and stage of maturity on energy values under the climate conditions of western Canada, and to investigate relationship between chemical (NRC 2001 formula) approach and biological approaches (in vitro and in situ assay) on prediction of energy values. The results showed that, in general, forage species (alfalfa vs. timothy) and cutting stage had profound impacts, but the varieties within each species (Pioneer vs. Beaver in alfalfa; Climax vs. Joliette in timothy) had minimal effects on energy values. As forage maturity increased, the energy contents behaved in a quadratic fashion, increasing at stage 2 and then significantly decreasing at stage 3. However, the prediction methods-chemical approach (NRC 2001 formula) and biological approaches (in vitro and in situ assay) had great influences on energy values. The highest predicted energy values were found by using the in situ approach, the lowest prediction value by using the NRC 2001 formula, and the intermediate values by the in vitro approach. The in situ results may be most accurate because it is closest to simulate animal condition. The energy values measured by biological approaches are not predictable by the chemical approach in this study, indicating that a refinement is needed in accurately predicting energy values.