• Smart Structures and Systems
• /
• v.28 no.6
• /
• pp.799-810
• /
• 2021

When monitoring the structural integrity of a bridge using data collected through accelerometers, identifying the profile of the load exerted on the bridge from the vehicles passing over it becomes a crucial task. In this study, the speed and location of vehicles on the deck of a bridge is reconfigured using real-time video to implicitly associate the load applied to the bridge with the response from the bridge sensors to develop an image-based deep learning network model. Instead of directly measuring the load that a moving vehicle exerts on the bridge, the intention in the proposed method is to replace the correlation between the movement of vehicles from CCTV images and the corresponding response by the bridge with a neural network model. Given the framework of an input-output-based system identification, CCTV images secured from the bridge and the acceleration measurements from a cantilevered beam are combined during the process of training the neural network model. Since in reality, structural damage cannot be induced in a bridge, the focus of the study is on identifying local changes in parameters by adding mass to a cantilevered beam in the laboratory. The study successfully identified the change in the material parameters in the beam by using the deep-learning neural network model. Also, the method correctly predicted the acceleration response of the beam. The proposed approach can be extended to the structural health monitoring of actual bridges, and its sensitivity to damage can also be improved through optimization of the network training.

• Smart Structures and Systems
• /
• v.32 no.3
• /
• pp.179-193
• /
• 2023

This study explores an alternative to the existing centralized process for data anomaly detection in modern Internet of Things (IoT)-based structural health monitoring (SHM) systems. An edge intelligence framework is proposed for the early detection and classification of various data anomalies facilitating quality enhancement of acquired data before transmitting to a central system. State-of-the-art deep neural network pruning techniques are investigated and compared aiming to significantly reduce the network size so that it can run efficiently on resource-constrained edge devices such as wireless smart sensors. Further, depthwise separable convolution (DSC) is invoked, the integration of which with advanced structural pruning methods exhibited superior compression capability. Last but not least, quantization-aware training (QAT) is adopted for faster processing and lower memory and power consumption. The proposed edge intelligence framework will eventually lead to reduced network overload and latency. This will enable intelligent self-adaptation strategies to be employed to timely deal with a faulty sensor, minimizing the wasteful use of power, memory, and other resources in wireless smart sensors, increasing efficiency, and reducing maintenance costs for modern smart SHM systems. This study presents a theoretical foundation for the proposed framework, the validation of which through actual field trials is a scope for future work.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.49-56
• /
• 2024

There have been emerging many use-cases applying recommendation systems especially in online platform. Although the performance of recommendation systems is affected by a variety of factors, selecting appropriate features is difficult since most of recommendation systems have sparse data. Conventional matrix factorization (MF) method is a basic way to handle with problems in the recommendation systems. However, the MF based scheme cannot reflect non-linearity characteristics well. As deep learning technology has been attracted widely, a deep neural network (DNN) framework based collaborative filtering (CF) was introduced to complement the non-linearity issue. However, there is still a problem related to feature embedding for use as input to the DNN. In this paper, we propose an effective method using singular value decomposition (SVD) based feature embedding for improving the DNN performance of recommendation algorithms. We evaluate the performance of recommendation systems using MovieLens dataset and show the proposed scheme outperforms the existing methods. Moreover, we analyze the performance according to the number of latent features in the proposed algorithm. We expect that the proposed scheme can be applied to the generalized recommendation systems.

• Proceedings of the IEEK Conference
• /
• 2003.07d
• /
• pp.1251-1254
• /
• 2003

In this paper, Experience Sensitive Cumulative Neural Network (ESCNN) is introduced, which can cumulate the same or similar experiences. As the same or similar training patterns are cumulated in the network, the system recognize more important information in the training patterns. The functions of forgetting less important informations and attending more important informations resided in the training patterns are surveyed and implemented by simulations. The system behaves well under the noisy circumstances due to its forgetting and/or attending properties, even in 50 percents noisy environments. This paper also describes the creation of the generalized patterns for the input training patterns.

• Tunnel and Underground Space
• /
• v.6 no.4
• /
• pp.316-325
• /
• 1996

A new accelerated neural network adopting modified sigmoid function was developed and applied to estimate engineering properties of rock from insufficient geological data. Developed network was tested on the well-known XOR and character recognition problems to verify the validity of the algorithms. Both learning speed and recognition rate were improved. Test learn on the Lee and Sterling's problems showed that learning time was reduced from tens of hours to a few minutes, while the output pattern was almost the same as other studies. Application to the various case studies showed exact coincidence with original data or measured results.

• Proceedings of the IEEK Conference
• /
• 1998.06a
• /
• pp.909-912
• /
• 1998

In this paper, a new connection admission controller using neural network is presented. The controller measures traffic flow, cell loss rate, and cell delay periodically. Using those measured information, it learns the distributions of traffics of each traffic. Also the proposed controller is able to measure and manage the delays that source traffics experience through the network by using DWRR multiplexer with buffers dedicated to each traffic source. Experimental result show that the heterogeneous traffic sources with various QoS requirement.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.679-680
• /
• 2008

When we apply the neural networks to applications, we need to select proper architecture of the network and the activation function of the network is one of most important characteristics. In this research, we propose a method to make a network using multiple activation functions. The performance of the proposed method is investigated through the computer simulations on various regression problems.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.124-126
• /
• 1989

Inverse kinematic problem is a crucial point for robot manipulator control. In this paper, to implement the Jacobian control technique we used the Hopfield(Tank)'s neural network. The states of neurons represent joint veocities, and the connection weights are determined from the current value of the Jacobian matrix. The network energy function is constructed so that its minimum corresponds to the minimum least square error. At each sampling time, connection weights and neuron states are updated according to current joint position. Inverse kinematic solution to the planar redundant manipulator is solved by computer simulation.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.04a
• /
• pp.403-413
• /
• 1996

Two-phase self-organizing neuro-modeling (SONM). the global SONM and local SONM, is designed for tracking non-stationary manufacturing processes. Radial basis function (RBF) neural network is employed, and self-tuning estimator is also developed for the determination of RBF network parameters on-line. A pattern recognition approach is presented for identifying a correct RBF neural network, which is used for identifying current manufacturing processes. Experimental results showed that the proposed approach is suitable for tracking non-stationary processes.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2290-2292
• /
• 1998

This paper introduces Two-stage neural network that is capable of recognizing spatio-temporal patterns. First stage takes a spatio-temporal pattern as input and compress it into sparse spatio-temporal pattern. Second stage is for temporal pattern recognition with nonuniform inhibitory connections and different cell sizes. These are basic properties for detecting a embeded pattern in a larger pattern. The network is evaluated by computer simulation.