• The Transactions of the Korean Institute of Power Electronics
• /
• v.27 no.1
• /
• pp.48-55
• /
• 2022

This paper proposes a battery remaining useful life (RUL) prediction method using a deep learning-based EMD-CNN-LSTM hybrid method. The proposed method pre-processes capacity data by applying empirical mode decomposition (EMD) and predicts the remaining useful life using CNN-LSTM. CNN-LSTM is a hybrid method that combines convolution neural network (CNN), which analyzes spatial features, and long short term memory (LSTM), which is a deep learning technique that processes time series data analysis. The performance of the proposed remaining useful life prediction method is verified using the battery aging experiment data provided by the NASA Ames Prognostics Center of Excellence and shows higher accuracy than does the conventional method.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.224-229
• /
• 2009

Digital image recognition has been used for steel bridge surface assessment since late 1990s. However, the non-uniform illumination problems such as shades, shadows, and highlights are still challenges in image processing to date. Therefore, this paper develops a new approach to tackle the non-uniform illumination problem for rust image adjustment. The inhomogeneous illumination problem is divided into shades/shadows and highlights in this paper. The proposed BEMD-morphology approach (BMA) utilizes the bidimensional empirical mode decomposition to mitigate the shade/shadow effect, and the morphological processing to detect and replace the highlight area. Finally, the rust image processed with the BMA will be segmented by the K-Means algorithm, one of the most popular and effective methods, to show the effectiveness of illumination adjustment.

• International Journal of Precision Engineering and Manufacturing
• /
• v.2 no.3
• /
• pp.11-18
• /
• 2001

A simple and convenient method of analysis for obtaining the individual stress intensity factors in a three-dimensional mixed mode crack is proposed. The procedures presented here are based on the path independence of J integral and mutual or two-state conservation integral, which involves two elastic fields. The problem is reduced to the determination of mixed mode stress intensity factor solutions in terms of conservation integrals involving known auxiliary solutions. Some numerical examples are presented to investigate the effectiveness and applicability of the method for a three-dimensional penny-shaped crack problem under mixed mode. This procedure is applicable to a three-dimensional mixed mode curved crack.

• Journal of the Korean Society of Combustion
• /
• v.22 no.2
• /
• pp.1-8
• /
• 2017

To evaluate the combustion instability of a gas turbine combustor, the DMD technique was applied. The mode frequency results for each fuel composition were compared with FFT(Fast Fourier Transform) results. The damping coefficient, which is a quantitative parameter for combustion instability, was evaluated for 5 experimental cases. The flame transfer function (FTF) was calculated in the most unstable test case. In deriving the FTF, gain and phase were calculated using DMD technique. As a result of the analysis of the OH radical perturbation of the DMD, the heat release fluctuation was the highest at 100 Hz, at which the highest value of gain is observed. The frequency of FFT and FTF were different. In order to clarify the reason for this, FTF for various resonance frequencies was performed and it shows that the pattern of gain was similar to FFT.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.6
• /
• pp.493-498
• /
• 2015

Complex acoustic signals can be formed in a water-filled acoustic pulse tube under some exciting conditions. It makes difficult to measure plane-wave reflection coefficient with the pulse tube for low frequency bands. In this study, using COMSOL Multiphysics we show that the tube wall excitation generates complex acoustic field of nonplanar mode as well as planar one. From such field incident or reflected planar mode can be decomposed respectively with a modal decomposition method, two-dimensional Fourier filtering. It makes possible to more accurately determine the plane-wave reflection coefficient of acoustic specimen with time gating.

• Journal of Power Electronics
• /
• v.16 no.3
• /
• pp.936-945
• /
• 2016

Large magnitude common-mode voltage (CMV) and its variation dv/dt have an adverse effect on motor drives that leads to early winding failure and bearing deterioration. For matrix converters, the switch states that connect each output line to a different input phase result in the lowest CMV among all of the valid switch states. To reduce the output CMV for matrix converters, this paper presents a new space vector modulation (SVM) strategy by utilizing these switch states. By this mean, the peak value and the root mean square of the CMV are dramatically decreased. In comparison with the conventional SVM methods this strategy has a similar computation overhead. Experiment results are shown to validate the effectiveness of the proposed modulation method.

• Smart Structures and Systems
• /
• v.2 no.2
• /
• pp.141-154
• /
• 2006

Recently, the Hilbert-Huang transform (HHT) has gained considerable attention as a novel technique of signal processing, which shows promise for the system identification and damage detection of structures. This study investigates the effectiveness and accuracy of the HHT method for the system identification and damage detection of structures through a series of experiments. A multi-degree-of-freedom (MDOF) structural model has been constructed with modular members, and the columns of the model can be replaced or removed to simulate damages at different locations with different severities. The measured response data of the structure due to an impulse loading is first decomposed into modal responses using the empirical mode decomposition (EMD) approach with a band-pass filter technique. Then, the Hilbert transform is subsequently applied to each modal response to obtain the instantaneous amplitude and phase angle time histories. A linear least-square fit procedure is used to identify the natural frequencies and damping ratios from the instantaneous amplitude and phase angle for each modal response. When the responses at all degrees of freedom are measured, the mode shape and the physical mass, damping and stiffness matrices of the structure can be determined. Based on a comparison of the stiffness of each story unit prior to and after the damage, the damage locations and severities can be identified. Experimental results demonstrate that the HHT method yields quite accurate results for engineering applications, providing a promising tool for structural health monitoring.

• Journal of The Korean Astronomical Society
• /
• v.53 no.6
• /
• pp.139-147
• /
• 2020

The sunspot area is a critical physical quantity for assessing the solar activity level; forecasts of the sunspot area are of great importance for studies of the solar activity and space weather. We developed an innovative hybrid model prediction method by integrating the complementary ensemble empirical mode decomposition (CEEMD) and extreme learning machine (ELM). The time series is first decomposed into intrinsic mode functions (IMFs) with different frequencies by CEEMD; these IMFs can be divided into three groups, a high-frequency group, a low-frequency group, and a trend group. The ELM forecasting models are established to forecast the three groups separately. The final forecast results are obtained by summing up the forecast values of each group. The proposed hybrid model is applied to the smoothed monthly mean sunspot area archived at NASA's Marshall Space Flight Center (MSFC). We find a mean absolute percentage error (MAPE) and a root mean square error (RMSE) of 1.80% and 9.75, respectively, which indicates that: (1) for the CEEMD-ELM model, the predicted sunspot area is in good agreement with the observed one; (2) the proposed model outperforms previous approaches in terms of prediction accuracy and operational efficiency.

• Asia-Pacific Journal of Atmospheric Sciences
• /
• v.54 no.4
• /
• pp.611-622
• /
• 2018

Observed climate data are processed under the assumption that their time series are stationary, as in multi-step temperature and precipitation prediction, which usually leads to low prediction accuracy. If a climate system model is based on a single prediction model, the prediction results contain significant uncertainty. In order to overcome this drawback, this study uses a method that integrates ensemble prediction and a stepwise regression model based on a mean-valued generation function. In addition, it utilizes empirical mode decomposition (EMD), which is a new method of handling time series. First, a non-stationary time series is decomposed into a series of intrinsic mode functions (IMFs), which are stationary and multi-scale. Then, a different prediction model is constructed for each component of the IMF using numerical ensemble prediction combined with stepwise regression analysis. Finally, the results are fit to a linear regression model, and a short-term climate prediction system is established using the Visual Studio development platform. The model is validated using temperature data from February 1957 to 2005 from 88 weather stations in Guangxi, China. The results show that compared to single-model prediction methods, the EMD and ensemble prediction model is more effective for forecasting climate change and abrupt climate shifts when using historical data for multi-step prediction.

• Structural Monitoring and Maintenance
• /
• v.8 no.4
• /
• pp.379-402
• /
• 2021

Traditional approaches for structural health monitoring (SHM) seldom take ambient uncertainty (temperature, humidity, ambient vibration) into consideration, while their impacts on structural responses are substantial, leading to a possibility of raising false alarms. A few predictors model-based approaches deal with these uncertainties through complex numerical models running online, rendering the SHM approach to be compute-intensive, slow, and sometimes not practical. Also, with model-based approaches, the imperative need for a precise understanding of the structure often poses a problem for not so well understood complex systems. The present study employs a data-based approach coupled with Empirical mode decomposition (EMD) to correlate recorded response time histories under varying temperature conditions to corresponding damage scenarios. EMD decomposes the response signal into a finite set of intrinsic mode functions (IMFs). A two-dimensional Convolutional Neural Network (2DCNN) is further trained to associate these IMFs to the respective damage cases. The use of IMFs in place of raw signals helps to reduce the impact of sensor noise while preserving the essential spatio-temporal information less-sensitive to thermal effects and thereby stands as a better damage-sensitive feature than the raw signal itself. The proposed algorithm is numerically tested on a single span bridge under varying temperature conditions for different damage severities. The dynamic strain is recorded as the response since they are frame-invariant and cheaper to install. The proposed algorithm has been observed to be damage sensitive as well as sufficiently robust against measurement noise.