• International journal of advanced smart convergence
• /
• v.9 no.2
• /
• pp.20-27
• /
• 2020

Sound event detection in real-world environments suffers from the interference of non-stationary and time-varying noise. This paper presents an adaptive noise reduction method for sound event detection based on non-negative matrix factorization (NMF). In this paper, we proposed a deep learning model that integrates Convolution Neural Network (CNN) with Non-Negative Matrix Factorization (NMF). To improve the separation quality of the NMF, it includes noise update technique that learns and adapts the characteristics of the current noise in real time. The noise update technique analyzes the sparsity and activity of the noise bias at the present time and decides the update training based on the noise candidate group obtained every frame in the previous noise reduction stage. Noise bias ranks selected as candidates for update training are updated in real time with discrimination NMF training. This NMF was applied to CNN and Hidden Markov Model(HMM) to achieve improvement for performance of sound event detection. Since CNN has a more obvious performance improvement effect, it can be widely used in sound source based CNN algorithm.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.8 no.2
• /
• pp.9-15
• /
• 2002

Ship operators use anchor dredging for the collision avoidance or safety of ship handling in a harbour or narrow channel. This paper clarifies the technique of the anchor dredging known as a common sense for. the seafarers A mathematical model at low speed range is established for the estimation of ship motion under the assumed environment, simulate the advance speed , and turning ability under the anchor dredging or not. The results shows good agreement with the conventional seamanship and their experiences as follows. Ahead speed used the anchor dredging is slower(speed reduction ratio:40%) than the normal ahead speed and the stopping distance is shorter (distance reduction ratio:40%)than the normal ahead distance without the anchor dredging.. Turning speed used anchor dredging is slower(speed reduction ratio:72%)than the normal ahead speed and the tactical diameter is shorter(distance reduction ratio:24%)than the diameter by the normal turning without the anchor dredging.

• Proceedings of KOSOMES biannual meeting
• /
• 2002.10a
• /
• pp.127-134
• /
• 2002

Ship operators are used to dredge anchor for the collision avoidance or safety of ship handling in a harbour or narrow channel. This paper clarifies the technique using tile anchor dredging known as a common sense for the seafarers. A mathematical model at low speed range has been established for the estimation of ship motion under the assumed environment , simulate the advance speed , and turning ability under the anchor dredging or not. The results shows good agreement with the conventional seamanship and their experiences as follows. Ahead speed used the anchor dredging is slower(speed reduction ratio:40%) than the normal ahead speed and the stopping distance is shorter (distance reduction ratio:40%)than the normal ahead distance without the anchor dredging. Turning speed used anchor dredging is slower(speed reduction ratio:72%)than the normal ahead speed and the tactical diameter is shorter(distance reduction ratio:24%)than the diameter by the normal turning without the anchor dredging.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.21 no.4
• /
• pp.100-106
• /
• 2022

Topology optimization is applied for the optimal design of various products to ensure weight reduction and productivity improvement. Reducing the weight of the mold while maintaining its rigidity can ensure shortening of the production cycle, stabilization of the mold temperature, and reduction of mold material costs. In this study, a topology optimization technique was applied to the optimal design of the injection mold, and a topology-optimized model of the mold was obtained. First, the injection mold for the square specimens was modeled. Subsequently, a structural analysis was performed by implementing a load condition generated during the injection molding process. Topology optimization was performed based on the structural analysis results, and the models of the initial and topology-optimized designs were manufactured at 1/4 magnification using a 3D printer. Consequently, compared with the existing model, the weight of the topology-optimized model decreased by 9.8%, and the manufacturing time decreased by 7.61%.

• Structural Engineering and Mechanics
• /
• v.85 no.1
• /
• pp.89-104
• /
• 2023

This work presents a proposal for employing reduced numerical integration in the formulation of the 4-node quadrilateral solid finite element. The use of these low-order integration rules leads to numerical instabilities such as those producing the hourglass effect. The proposed procedure allows evaluating a given constitutive model only in one integration point, achieving an attractive computational cost reduction and, also, successfully controls the hourglass effect. A validation of the proposal is included and discussed throughout the paper. To show the efficiency of the proposal, several application examples of masonry structures are studied and discussed. To represent the non-linear mechanical behaviour of masonry a plastic-damage model is implemented within the application of this sub-integration scheme. Also, in order to have a full and computationally efficient strategy to determine the behaviour of masonry structures, involving its evolution to collapse, a homogenization technique with a macro-modeling approach is used. The methodology discussed throughout this paper demonstrates a substantial computational cost reduction and an improved approximation of the non-linear problem evidenced by a reduction of up to 85% of the computational time for some cases.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.154.2-154
• /
• 2001

This paper presented a adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when ...

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.99.1-99
• /
• 2001

This paper presented an adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when only one adaptation utterance is used.

• Proceedings of the IEEK Conference
• /
• 2001.06b
• /
• pp.133-136
• /
• 2001

A new package design method to reduce resonance effect due to an IC package is represented. Frequency-variant circuit model of the power/ground plane was developed to accurately reflect the resonance. The circuit model is benchmarked with a full wave simulation, thereby verifying its accuracy. Then it was shown that the proposed technique can efficiently reduce the resonance due to the IC package.

• Journal of the Korea Society for Simulation
• /
• v.12 no.4
• /
• pp.83-94
• /
• 2003

Managing a call center is a complex and diverse challenge. Call center becomes a very important contact point and a data ware house for successful CRM. Improving performance of call center is critical and valuable for providing better service. In this study we applied forecasting technique to estimate incoming calls and ProModel based simulation model to enhance performance of a mobile telecommunication company's call center. The simulation study shows reduction in managing cost and better customer's satisfaction.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.4
• /
• pp.687-690
• /
• 1994

This paper presents a method of obtaining transmission network equivalents from the network's response to the pulse excitation signal. Proposed method is base on Prony signal analysis and jtransfer function identification technique. As a result Thevenin-type of discrete-time filter model can be generated. It can reproduce the driving point impedance characteristic of the network.