• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.316-319
• /
• 2014

People's interests in coffee are increasing with the expansion of coffee market. In this trend, people's taste becomes more luxurious and coffee bean's quality is considered to be very important. Currently, bean defects are mainly detected by experienced specialists. In this paper, a detection system of bean defects using machine learning is presented. This system concentrates on detecting two main defect types : bean's shape and insect damage. Convolutional Neural Networks are used for machine learning. The neural networks are comprised of two neural networks. The first neural network detects defects in the bean's shape, and the second one detects the bean's insect damage. The development of this system could be a starting point for automated coffee bean defects detection. Later, further research is needed to detect other bean defect types.

• Smart Structures and Systems
• /
• v.15 no.1
• /
• pp.227-244
• /
• 2015

This study develops a two stage procedure to identify the structural damage based on the optimized artificial neural networks. Initially, the modal strain energy index (MSEI) is established to extract the damaged elements and to reduce the computational time. Then the genetic algorithm (GA) and artificial neural networks (ANNs) are combined to detect the damage severity. The input of the network is modal strain energy index and the output is the flexural stiffness of the beam elements. The principal component analysis (PCA) is utilized to reduce the input variants of the neural network. By using the genetic algorithm to optimize the parameters, the ANNs can significantly improve the accuracy and convergence of the damage identification. The influence of noise on damage identification results is also studied. The simulation and experiment on beam structures shows that the adaptive parameter selection neural network can identify the damage location and severity of beam structures with high accuracy.

• Earthquakes and Structures
• /
• v.11 no.5
• /
• pp.841-859
• /
• 2016

Bridge bearings are important connection elements between bridge superstructures and substructures, whose health states directly affect the performance of the bridges. This paper systematacially presents a new method to identify the bridge bearing damage based on the neural network theory. Firstly, based on the analysis of different damage types, a description of the bearing damage is introduced, and a uniform description for all the damage types is given. Then, the feasibility and sensitivity of identifying the bearing damage with bridge vibration modes are investigated. After that, a Radial Basis Function Neural Network (RBFNN) is built, whose input and output are the beam modal information and the damage information, respectively. Finally, trained by plenty of data samples formed by the numerical method, the network is employed to identify the bearing damage. Results show that the bridge bearing damage can be clearly reflected by the modal information of the bridge beam, which validates the effectiveness of the proposed method.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.3
• /
• pp.272-278
• /
• 2010

A method for damage detection in a plate structure is presented based on strain waves that are generated by impact or damage in the structure. Strain responses from continuous sensors, which are long ribbon-like sensors made from piezoceramic fibers or other materials, were used with a neural network technique to estimate the damage location. The continuous sensor uses only a small number of channels of data acquisition and can cover large areas of the structure. A grid type structural neural system composed of the continuous sensors was developed for effective damage localization in a plate structure. The ratios of maximum strains and arrival times of the maximum strains obtained from the continuous sensors were used as input data to a neural network. Simulated damage localizations on a plate were carried out and the identified damage locations agreed reasonably well with the exact damage locations.

• Smart Structures and Systems
• /
• v.18 no.6
• /
• pp.1233-1250
• /
• 2016

Cracks in plate-like structures are some of the main reasons for destruction of the entire structure. In this study, a novel two-stage methodology is proposed for damage detection of flexural plates using an optimized artificial neural network. In the first stage, location of damages in plates is investigated using curvature-moment and curvature-moment derivative concepts. After detecting the damaged areas, the equations for damage severity detection are solved via Bat Algorithm (BA). In the second stage, in order to efficiently reduce the computational cost of model updating during the optimization process of damage severity detection, multiple damage location assurance criterion index based on the frequency change vector of structures are evaluated using properly trained cascade feed-forward neural network (CFNN) as a surrogate model. In order to achieve the most generalized neural network as a surrogate model, its structure is optimized using binary version of BA. To validate this proposed solution method, two examples are presented. The results indicate that after determining the damage location based on curvature-moment derivative concept, the proposed solution method for damage severity detection leads to significant reduction of computational time compared with direct finite element method. Furthermore, integrating BA with the efficient approximation mechanism of finite element model, maintains the acceptable accuracy of damage severity detection.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.982-986
• /
• 1997

A method using cutting force signal and neural network for detection tool damage is proposed. Cutting force signal is gained by tool dynamometer and the signal is prepocessed to normalize. Cutting force signal is changed by tool state. When tool damage is occurred, cutting force signal goes up in comparison with that in normal state. However,the signal goes down in case of catastrophic fracture. These features are memorized in neural network through nomalizing couse. A new nomalizing method is introduced in this paper. Fist, cutting forces are sumed up except data smaller than threshold value, which is the cutting force during non-cutting action. After then, the average value is found by dividing by the number of data. With backpropagation training process, the neural network memorizes the feature difference of cutting force signal between with and without tool damage. As a result, the cutting force can be used in monitoring the condition of cutting tool and neural network can be used to classify the cutting force signal with and without tool damage.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.10a
• /
• pp.321-329
• /
• 2002

Artificial neural network has been used for damage assessment by many researchers, but there are still some barriers that must be overcome to improve its accuracy and efficiency. The major problems with the conventional neural network are the necessity of many training patterns for neural network teaming process and ambiguity in the relationship of neural network structure to the convergence of solution. In this paper, the PNN is used as a pattern classifier to detect the damages of the railway bridge using dynamic response. The comparison between the mode shape and the natural frequency of structure as training pattern is investigated for approriate selection of the training pattern in the damage detection of railway bridge using the PNN.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2005.04a
• /
• pp.367-374
• /
• 2005

The use of system identification approaches for damage detection has been expanded in recent years. Soft computing techniques such as neural networks have been utilized increasingly. Damage assessment using neural networks is presented in this study. Data set for training neural networks are acceleration response of simple beam under the various damage states ,which are the inputs. The outputs are the damage locations and extents. Not only the trained damages but also untrained damages are. detected accuratelyintheassessmentstage.

• Journal of Wind Energy
• /
• v.9 no.4
• /
• pp.16-23
• /
• 2018

A damage detection method for the tripod support structure of offshore wind turbines is presented for structural health monitoring. A finite element model of a prototype tripod support structure is established and the modal properties are calculated. The degree and location of the damage are estimated based on the neural network technique using the changes of natural frequencies and mode shape due to the damage. The stress distribution occurring in the support structure is obtained by a dynamic analysis for the wind turbine system to select the output data of the neural network. The natural frequencies and mode shapes for 36 possible damage scenarios were used for the input data of the learned neural network for damage assessment. The estimated damages agreed reasonably well with the accurate ones. The presented method could be effectively applied for damage detection and structural health monitoring of various types of support structures of offshore wind turbines.

• Smart Structures and Systems
• /
• v.12 no.3_4
• /
• pp.345-361
• /
• 2013

Locating and assessing the severity of damage in large or complex structures is one of the most challenging problems in the field of civil engineering. Considering that the wavelet packet transform (WPT) has the ability to clearly reflect the damage characteristics of structural response signals and the artificial neural network (ANN) is capable of learning in an unsupervised manner and of forming new classes when the structural exhibits change, this paper investigates a multi-stage structural damage diagnosis method by using the WPT and ANN based on "energy-damage" theory, in which, the wavelet packet component energies are first extracted to be damage sensitive feature and then adopted as input into an improved back propagation (BP) neural network model for damage diagnosis in a step by step mode. To validate the efficacy of the presented approach of the damage diagnosis, the benchmark structure of the American Society of Civil Engineers (ASCE) is employed in the case study. The results of damage diagnosis indicate that the method herein is computationally efficient and is able to detect the existence of different damage patterns in the simulated experiment where minor, moderate and severe damages corresponds to involving in the loss of stiffness on braces or the removal bracing in various combinations.