• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• Design & Manufacturing
• /
• v.17 no.4
• /
• pp.72-78
• /
• 2023

Injection molding is a process widely used in various industries because of its high production speed and ease of mass production during the plastic manufacturing process, and the product is molded by injecting molten plastic into the mold at high speed and pressure. Since process conditions such as resin and mold temperature mutually affect the process and the quality of the molded product, it is difficult to accurately predict quality through mathematical or statistical methods. Recently, studies to predict the quality of injection molded products by applying artificial neural networks, which are known to be very useful for analyzing nonlinear types of problems, are actively underway. In this study, structural optimization of neural networks was conducted by applying multi-task learning techniques according to the characteristics of the input and output parameters of the artificial neural network. A structure reflecting the characteristics of each process step was applied to the input parameters, and a structure reflecting the quality characteristics of the injection molded part was applied to the output parameters using multi-tasking learning. Building an artificial neural network to predict the three qualities (mass, diameter, height) of injection-molded product under six process conditions (melt temperature, mold temperature, injection speed, packing pressure, pacing time, cooling time) and comparing its performance with the existing neural network, we observed enhancements in prediction accuracy for mass, diameter, and height by approximately 69.38%, 24.87%, and 39.87%, respectively.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.60 no.1
• /
• pp.87-99
• /
• 2024

This study deals with the application of an artificial neural network (ANN) model to predict power consumption for utilizing seawater source heat pumps of recirculating aquaculture system. An integrated dynamic simulation model was constructed using the TRNSYS program to obtain input and output data for the ANN model to predict the power consumption of the recirculating aquaculture system with a heat pump system. Data obtained from the TRNSYS program were analyzed using linear regression, and converted into optimal data necessary for the ANN model through normalization. To optimize the ANN-based power consumption prediction model, the hyper parameters of ANN were determined using the Bayesian optimization. ANN simulation results showed that ANN models with optimized hyper parameters exhibited acceptably high predictive accuracy conforming to ASHRAE standards.

• Proceedings of the KIEE Conference
• /
• 2006.10d
• /
• pp.47-49
• /
• 2006

This paper is proposed an efficiency optimization control algorithm for a synchronous reluctance motor which minimizes the copper and iron losses. The design of the speed controller based on adaptive learning mechanism-fuzzy neural networks(ALM-FNN) controller that is implemented using adaptive, fuzzy control and neural networks. The control performance of the hybrid artificial intelligent controller is evaluated by analysis for various operating conditions. Analysis results are presented to show the validity of the proposed algorithm.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 2007.05a
• /
• pp.306-310
• /
• 2007

This paper is proposed an efficiency optimization control algorithm for a synchronous reluctance motor which minimizes the copper and iron losses. The design of the speed controller based on adaptive learning mechanism-fuzzy neural networks(ALM-FNN) controller that is implemented using adaptive, fuzzy control and neural networks. The control performance of the hybrid artificial intelligent controller is evaluated by analysis for various operating conditions. Analysis results are presented to show the validity of the proposed algorithm

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1007-1008
• /
• 2007

This paper is proposed an efficiency optimization control algorithm for IPMSM which minimizes the copper and iron losses. The design of the speed controller based on adaptive fuzzy learning control-fuzzy neural networks(AFLC-FNN) controller that is implemented using adaptive, fuzzy control and neural networks. The control performance of the AFLC-FNN controller is evaluated by analysis for various operating conditions. Analysis results are presented to show the validity of the proposed algorithm

• Structural Engineering and Mechanics
• /
• v.50 no.2
• /
• pp.151-161
• /
• 2014

In this paper, fuel tank of the locomotive ER 24 has been studied. Firstly the behavior of fuel and air during the braking time has been investigated by using a two-phase model. Then, the distribution of pressure on the surface of baffles caused by sloshing has been extracted. Also, the fuel tank has been modeled and analyzed using Finite Element Method (FEM) considering loading conditions suggested by the DIN EN 12663 standard and real boundary conditions. In each loading condition, high stressed areas have been identified. By comparing the distribution of pressure caused by sloshing phenomena and suggested loading conditions, optimization of the tank has been taken into consideration. Moreover, internal baffles have been investigated and by modifying their geometric properties, search of the design space has been done to reach the optimal tank. Then, in order to reduce the mass and manufacturing cost of the fuel tank, Non-dominated Sorting Genetic Algorithm (NSGA-II) and Artificial Neural Networks (ANNs) have been employed. It is shown that compared to the primary design, the optimized fuel tank not only provides the safety conditions, but also reduces mass and manufacturing cost by %39 and %73, respectively.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• pp.807-817
• /
• 2019

Airborne Gamma Ray Spectrometry (AGRS) with its important applications such as gathering radiation information of ground surface, geochemistry measuring of the abundance of Potassium, Thorium and Uranium in outer earth layer, environmental and nuclear site surveillance has a key role in the field of nuclear science and human life. The Broyden-Fletcher-Goldfarb-Shanno (BFGS), with its advanced numerical unconstrained nonlinear optimization in collaboration with Artificial Neural Networks (ANNs) provides a noteworthy opportunity for modern AGRS. In this study a new AGRS system empowered by ANN-BFGS has been proposed and evaluated on available empirical AGRS data. To that effect different architectures of adaptive ANN-BFGS were implemented for a sort of published experimental AGRS outputs. The selected approach among of various training methods, with its low iteration cost and nondiagonal scaling allocation is a new powerful algorithm for AGRS data due to its inherent stochastic properties. Experiments were performed by different architectures and trainings, the selected scheme achieved the smallest number of epochs, the minimum Mean Square Error (MSE) and the maximum performance in compare with different types of optimization strategies and algorithms. The proposed method is capable to be implemented on a cost effective and minimum electronic equipment to present its real-time process, which will let it to be used on board a light Unmanned Aerial Vehicle (UAV). The advanced adaptation properties and models of neural network, the training of stochastic process and its implementation on DSP outstands an affordable, reliable and low cost AGRS design. The main outcome of the study shows this method increases the quality of curvature information of AGRS data while cost of the algorithm is reduced in each iteration so the proposed ANN-BFGS is a trustworthy appropriate model for Gamma-ray data reconstruction and analysis based on advanced novel artificial intelligence systems.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.200-206
• /
• 2021

Fraud in e-commerce transaction increased in the last decade especially with the increasing number of online stores and the lockdown that forced more people to pay for services and groceries online using their credit card. Several machine learning methods were proposed to detect fraudulent transaction. Neural networks showed promising results, but it has some few drawbacks that can be overcome using optimization methods. There are two categories of learning optimization methods, first-order methods which utilizes gradient information to construct the next training iteration whereas, and second-order methods which derivatives use Hessian to calculate the iteration based on the optimization trajectory. There also some training refinements procedures that aims to potentially enhance the original accuracy while possibly reduce the model size. This paper investigate the performance of several NN models in detecting fraud in e-commerce transaction. The backpropagation model which is classified as first learning algorithm achieved the best accuracy 96% among all the models.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.757-759
• /
• 1996

This paper proposes an application of artificial neural networks to the bus-bar separation in a substation for radial network operation. For the effective bus-bar operation, the insecurity index of transmission line load is introduced. For the radial network operation. the constraints of bus-bar switch is formulated in the performance function with the insecurity index. The determination of bus-bar switching is to find the states of 0 or 1 in the circuit breakers. In this paper, it is tested that the bus-bar separation of binary optimization problem can be solved by Hopfield networks with adequate manipulations.