• Transactions of Materials Processing
• /
• v.17 no.5
• /
• pp.373-379
• /
• 2008

A fundamental study on warm incremental forming of a magnesium alloy sheet has been carried out. In order to enhance the incremental formability of the magnesium alloy sheet, a local heating device was newly designed and manufactured. Through the incremental forming tests of AZ31 under various forming conditions, the effects of process parameters such as the temperature, feeding depth per cycle, and inclination angle on the incremental formability of AZ31 were investigated. In addition, conventional FLDs at elevated temperatures were constructed experimentally and applied to predict the forming failure.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.477-480
• /
• 2005

메모리 기반 추론 기법에서 기억공간의 효율적 사용과 분류 시간을 줄이기 위한 다양한 방법들이 연구되고 있으며, NGE(Nested Generalized Exemplar) 이론을 예로 들 수 있다. 본 논문에서는 학습 패턴 집합으로부터 대표패턴을 생성하는 RPA(Recursive Partition Averaging) 기법과 점진적으로 대표패턴을 추출하는 IRPA(Incremental RPA) 기법을 제안한다.

• ETRI Journal
• /
• v.32 no.3
• /
• pp.395-405
• /
• 2010

A novel multiclassifier system (MCS) strategy is proposed and applied to a text-dependent speaker verification task. The presented scheme optimizes the linear combination of classifiers on an on-line basis. In contrast to ordinary MCS approaches, neither a priori distributions nor pre-tuned parameters are required. The idea is to improve the most accurate classifier by making use of the incremental information provided by the second classifier. The on-line multiclassifier optimization approach is applicable to any pattern recognition problem. The proposed method needs neither a priori distributions nor pre-estimated weights, and does not make use of any consideration about training/testing matching conditions. Results with Yoho database show that the presented approach can lead to reductions in equal error rate as high as 28%, when compared with the most accurate classifier, and 11% against a standard method for the optimization of linear combination of classifiers.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.2
• /
• pp.111-118
• /
• 2001

In this paper, in order to implement the multi-layered perceptron neural network using pure digital logic circuit model, we propose the new logic neuron structure, the digital canonical multi-layered logic neural network structure, and the multi-stage multi-layered logic neural network structure for pattern recognition applications. And we show that the proposed approach provides an incremental additive learning algorithm, which is very simple and effective.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.3
• /
• pp.429-438
• /
• 1996

An incremental learning algorithm is presented that constructs a multilayer perceptron whose size is optimal for solving a given problem. Unlike conventional algorithms in which a fixed size training set is processed repeat-edly, the method uses an increasing number of critical examples to find a necessary and sufficient number of hidden units for learning the entire data. Experimental results in hand- writtern digit recognition shows that the network size optimization combined with incremental pattern selection generalizes significantly better and converges faster than conventional methods.

• Transactions of Materials Processing
• /
• v.32 no.4
• /
• pp.208-214
• /
• 2023

This paper discusses the deformation characteristics of a scaled-down automotive outer door panel with vacuum-assisted incremental sheet forming. The vacuum condition between the die and Al6052-H32 sheet with a thickness of 1.0 mm is reviewed with the goal of improving the geometrical accuracy of the target product. The material flow according to the forming tool path, including the multi-tool path and conventional contour tool path, is investigated considering the degradation of the pillow effect. To reduce friction between the tool and the sheet during incremental forming, automotive engine oil (5W-30) is used as a lubricant, and the strain field on the surface of the formed product is analyzed using ARGUS. By comparing the geometry and material flow characteristics of products under different test conditions, it is confirmed that the product surface quality can be significantly improved when the vacuum condition is employed in conjunction with a multi-tool path strategy.

• Transactions of Materials Processing
• /
• v.32 no.6
• /
• pp.360-366
• /
• 2023

This study is concerned with the surface quality of products according to the material and coating condition of the forming tool in incremental sheet forming. Three forming tools, SKD11 with and without diamond-like-coating (DLC) and polymer tool tip, were used to form conical and pyramidal geometries to take into account the influence of friction between the forming tool and the sheet on the surface quality including geometric accuracy of deformed samples. Each test was performed using SUS304 with a thickness of 0.4 mm according to different incremental depths per lap of 0.5 mm, 1.0 mm, and 1.5 mm for the contour tool path, considering the increase in normal force which is associated with the frictional behavior during local deformation. The surface quality was then investigated through surface roughness measured with KEYENCE VR-6000 and relative strain distribution including deformed shape analyzed with ARGUS which is a non-contact optical strain measurement system. Differences between 3D CAD surfaces and captured geometry from experiments were evaluated to compare the effect of friction on geometric accuracy. From comparisons of experimental results, it was revealed that the polymer-based tool tip can improve surface quality and geometric accuracy by reducing the undesired material flow due to local friction in the increment sheet forming process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3378-3393
• /
• 2014

We have witnessed the rapid development of information technology in recent years. One of the key phenomena is the fast, near-exponential increase of data. Consequently, most of the traditional data classification methods fail to meet the dynamic and real-time demands of today's data processing and analyzing needs--especially for continuous data streams. This paper proposes an improved incremental learning algorithm for a large-scale data stream, which is based on SVM (Support Vector Machine) and is named DS-IILS. The DS-IILS takes the load condition of the entire system and the node performance into consideration to improve efficiency. The threshold of the distance to the optimal separating hyperplane is given in the DS-IILS algorithm. The samples of the history sample set and the incremental sample set that are within the scope of the threshold are all reserved. These reserved samples are treated as the training sample set. To design a more accurate classifier, the effects of the data volumes of the history sample set and the incremental sample set are handled by weighted processing. Finally, the algorithm is implemented in a cloud computing system and is applied to study user behaviors. The results of the experiment are provided and compared with other incremental learning algorithms. The results show that the DS-IILS can improve training efficiency and guarantee relatively high classification accuracy at the same time, which is consistent with the theoretical analysis.

• The KIPS Transactions:PartD
• /
• v.9D no.4
• /
• pp.561-570
• /
• 2002

In order to evaluate the performance of incremental update algorithms, we perform experimental tests on the time of updating view objects. In this paper, the incremental update algorithms are evaluated on two kinds of materialized methods : materialization by value-copy and materialization by preserving object identifiers (OIDs). The result of performance evaluation shows that there is little difference in the updating time of view objects between two materialization methods. The evaluation of query processing on spatial views shows that materialization by value-copy is much better than materialization by preserving OIDs. As the results of overall performance evaluation, it is more desirable to use the incremental update method based on materialization by value-copy than the incremental update method based on materialization by preserving OIDs.

• The KIPS Transactions:PartA
• /
• v.11A no.3
• /
• pp.149-156
• /
• 2004

As software is large and sophisticate, in order to increase the productivity and efficiency of programs in programming development environments, it is necessary to support the integrated system that offers user interface integrated editing, compiling, debugging, and running steps. The key tool in such environments is an incremental translation. In this paper, in order to increase the productivity and reusability of software, the goal is to construct the integrated incremental interpretation system that supports friendly user interface with editor, debugger, and incremental interpreter. We define the new object-oriented language, IMPLO(IMPerative Language with Object) using EBNF notation, and construct the integrated incremental interpretation system using incremental interpreter of the language. To do so, we extend attribute grammars for specifying static semantics and present new action equations to describe the dynamic semantics. We executed the incremental interpretation by using analyzing the dynamic semantics and then implemented integrated incremental interpretation system with editor and debugger in C, Lex and Yacc using X windows on SUN. We obtain about 50％ speedups in case of incremental execution time for example programs.