• Geophysics and Geophysical Exploration
• /
• v.24 no.2
• /
• pp.53-66
• /
• 2021

Velocity model building is an essential procedure in seismic data processing. Conventional techniques, such as traveltime tomography or velocity analysis take longer computational time to predict a single velocity model and the quality of the inversion results is highly dependent on human expertise. Full-waveform inversions also depend on an accurate initial model. Recently, deep neural network techniques are gaining widespread acceptance due to an increase in their integration to solving complex and nonlinear problems. This study investigated cases of seismic velocity model building using deep neural network techniques by classifying items according to the neural networks used in each study. We also included cases of generating training synthetic velocity models. Deep neural networks automatically optimize model parameters by training neural networks from large amounts of data. Thus, less human interaction is involved in the quality of the inversion results compared to that of conventional techniques and the computational cost of predicting a single velocity model after training is negligible. Additionally, unlike full-waveform inversions, the initial velocity model is not required. Several studies have demonstrated that deep neural network techniques achieve outstanding performance not only in computational cost but also in inversion results. Based on the research results, we analyzed and discussed the characteristics of deep neural network techniques for building velocity models.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.1-6
• /
• 1990

Neural network is a computational model of ft biological nervous system developed ID exploit its intelligence and parallelism. Applying neural networks so robots creates many advantages over conventional control methods such as learning, real-time control, and continuous performance improvement through training and adaptation. In this paper, dynamic control of a six-link robot will be presented using neural networks. The neural network model used in this paper is the backpropagation network. Simulated control of the PUMA 560 am shows that it can move a high speed as well as adapt to unforseen load changes and sensor noise. The results are compared with the conventional PD control scheme.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.500-503
• /
• 1990

Neural network is a computational model of the biological nervous system developed to exploit its intelligence and parallelism. Applying neural networks to robots creates many advantages over conventional control methods such as learning, real-time control, and continuous performance improvement through training and adaptation. In this paper, dynamic control of a six-link robot will be presented using neural networks. The neural network model used in this paper is the backpropagation network. Simulated control of the PUMA 560 arm shows that it can move at high speed as well as adapt to unforseen load changes. The results are compared with the conventional PD control scheme.

• Journal of the Korean Society of Mechanical Technology
• /
• v.20 no.6
• /
• pp.872-878
• /
• 2018

Steam tables including superheated, saturated and compressed region were simultaneously modeled using the neural networks. Pressure and temperature were used as two inputs for superheated and compressed region. On the other hand Pressure and dryness fraction were two inputs for saturated region. The outputs were specific volume, specific enthalpy and specific entropy. The neural network model were compared with the linear interpolation model in terms of the percentage relative errors. The criterion of judgement was selected with the percentage relative error of 1%. In conclusion the neural networks showed better results than the interpolation method for all data of superheated and compressed region and specific volume of saturated region, but similar for specific enthalpy and entropy of saturated region.

• Journal of Korean Society of Steel Construction
• /
• v.11 no.1 s.38
• /
• pp.79-88
• /
• 1999

Damages of a steel box girder bridge are detected using neural networks. Damage detection using neural networks has increasing momentum in structural engineering. It is a new effort to overcome the limitations of the conventional analytical approaches and applied to the damage detection of a steel box-girder bridge. Data sets for training neural networks are obtained from the acceleration response of the bridge under moving load. Finite element model is first defined and damages of 5, 10, 15 and 20% are assumed in the model. Not only the trained damages but untrained damages are detected in the assessment stage. The untrained damages can be detected with acceptable errors. Because the number of damaged locations are limited to a few parts, more researches are needed to put this technique into practice.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2699-2701
• /
• 2001

The study is concerned with an approach to the design of a new category of fuzzy neural networks. The proposed Fuzzy Polynomial Neural Networks(FPNN) with hybrid multi-layer inference architecture is based on fuzzy neural networks(FNN) and polynomial neural networks(PNN) for model identification of complex and nonlinear systems. The one and the other are considered as premise and consequence part of FPNN respectively. We introduce two kinds of FPNN architectures, namely the generic and advanced types depending on the connection points (nodes) of the layer of FNN. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process and to get output performance with superb predictive ability. The availability and feasibility of the FPNN is discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed FPNN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1034-1038
• /
• 2020

In order to effectively process target recognition using radar, accurate signal information for the target is required. However, such a target signal is usually mixed with noise, and this part of the study is continuously carried out. Especially, image processing, target signal processing and target recognition for the target are examples. Since the field of target recognition is important from a military point of view, this paper carried out research on target recognition of aircraft using a tree-structured fuzzy neural networks. Fuzzy neural networks are learned by using reflected signal data for an aircraft to optimize the model, and then test data for the target are used for the optimized model to perform an experiment on target recognition. The effectiveness of the proposed method is verified by the simulation results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3C
• /
• pp.257-267
• /
• 2009

This paper presents a training method for neural networks and the employment of MSE (mean scare error) values as the basis of a decision regarding the identity claim of a speaker in a recurrent neural networks based speaker verification system. Recurrent neural networks (RNNs) are employed to capture temporally dynamic characteristics of speech signal. In the process of supervised learning for RNNs, target outputs are automatically generated and the generated target outputs are made to represent the temporal variation of input speech sounds. To increase the capability of discriminating between the true speaker and an impostor, a discriminative training method for RNNs is presented. This paper shows the use and the effectiveness of the MSE value, which is obtained from the Euclidean distance between the target outputs and the outputs of networks for test speech sounds of a speaker, as the basis of speaker verification. In terms of equal error rates, results of experiments, which have been performed using the Korean speech database, show that the proposed speaker verification system exhibits better performance than a conventional hidden Markov model based speaker verification system.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.11
• /
• pp.57-63
• /
• 1996

Monitoring of grinding troble occurring during the process is classified into the quantitative data which depends upon a sensor and the qualitative knowledge which relies upon an empirical knowledge. Since grinding operation is highly related with a large amount of functional parameters, it is actually deficulty in copying wiht the grinding troubles through the process. To cope with grinding trouble, it is an effective monitoring systems when occurring the grinding process. The use of neural networks is an effective method of detection and/or monitroing on the grinding trouble. In this paper, four parameters which are derived from the AE(Acoustic Emission) signatures are identified, and grinding monitoring system utilized a back propagation learning algorithm of PDP neural networks is presented.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.8
• /
• pp.307-324
• /
• 2018

Recently, recurrent neural networks have been attracting attention in solving prediction problem of sequential data through structure considering time dependency. However, as the time step of sequential data increases, the problem of the gradient vanishing is occurred. Long short-term memory models have been proposed to solve this problem, but there is a limit to storing a lot of data and preserving it for a long time. Therefore, research on memory-augmented neural network (MANN), which is a learning model using recurrent neural networks and memory elements, has been actively conducted. In this paper, we describe the structure and characteristics of MANN models that emerged as a hot topic in deep learning field and present the latest techniques and future research that utilize MANN.