• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.125-130
• /
• 2020

At the end of 1997, the volatility of the exchange rate intensified as the nation's exchange rate system was converted into a free-floating exchange rate system. As a result, managing the exchange rate is becoming a very important task, and the need for forecasting the exchange rate is growing. The exchange rate prediction model using the existing exchange rate prediction method, statistical technique, cannot find a nonlinear pattern of the time series variable, and it is difficult to analyze the time series with the variability cluster phenomenon. And as the number of variables to be analyzed increases, the number of parameters to be estimated increases, and it is not easy to interpret the meaning of the estimated coefficients. Accordingly, the exchange rate prediction model using artificial neural network, rather than statistical technique, is presented. Using DNN, which is the basis of deep learning among artificial neural networks, and LSTM, a recurrent neural network model, the number of hidden layers, neurons, and activation function changes of each model found the optimal exchange rate prediction model. The study found that although there were model differences, LSTM models performed better than DNN models and performed best when the activation function was Tanh.

• Animal Bioscience
• /
• v.34 no.2
• /
• pp.306-311
• /
• 2021

Objective: The objectives were to develop prediction equations for digestible energy (DE) and metabolizable energy (ME) of feed ingredients and diets for pigs based on chemical composition and to evaluate the accuracy of the equations using in vivo data. Methods: A total of 734 data points from 81 experiments were employed to develop prediction equations for DE and ME in feed ingredients and diets. The CORR procedure of SAS was used to determine correlation coefficients between chemical components and energy concentrations and the REG procedure was used to generate prediction equations. Developed equations were tested for the accuracy according to the regression analysis using in vivo data. Results: The DE and ME in feed ingredients and diets were most negatively correlated with acid detergent fiber or neutral detergent fiber (NDF; r = -0.46 to r = -0.67; p<0.05). Three prediction equations for feed ingredients reflected in vivo data well as follows: DE = 728+0.76×gross energy (GE)-25.18×NDF (R2 = 0.64); ME = 965+0.66×GE-24.62×NDF (R2 = 0.60); ME = 1,133+0.65×GE-29.05×ash-23.17×NDF (R2 = 0.67). Conclusion: In conclusion, the equations suggested in the current study would predict energy concentration in feed ingredients and diets.

• Journal of Wind Energy
• /
• v.15 no.2
• /
• pp.10-22
• /
• 2024

This research is a comprehensive analysis of wind power prediction sensitivity using a Long Short-Term Memory (LSTM) deep learning neural network model, accounting for the inherent uncertainties in wind speed estimation. Utilizing a year's worth of operational data from an operational wind farm, the study forecasts the power output of both individual wind turbines and the farm collectively. Predictions were made daily at intervals of 10 minutes and 1 hour over a span of three months. The model's forecast accuracy was evaluated by comparing the root mean square error (RMSE), normalized RMSE (NRMSE), and correlation coefficients with actual power output data. Moreover, the research investigated how inaccuracies in wind speed inputs affect the power prediction sensitivity of the model. By simulating wind speed errors within a normal distribution range of 1% to 15%, the study analyzed their influence on the accuracy of power predictions. This investigation provided insights into the required wind speed prediction error rate to achieve an 8% power prediction error threshold, meeting the incentive standards for forecasting systems in renewable energy generation.

• Smart Structures and Systems
• /
• v.33 no.5
• /
• pp.325-332
• /
• 2024

This study proposes a novel long short-term memory (LSTM)-based approach for predicting carbonation depth, with the aim of enhancing the durability evaluation of concrete structures. Conventional carbonation depth prediction relies on statistical methodologies using carbonation influencing factors and in-situ carbonation depth data. However, applying in-situ data for predictive modeling faces challenges due to the lack of time-series data. To address this limitation, an LSTM-based carbonation depth prediction technique is proposed. First, training data are generated through random sampling from the distribution of carbonation velocity coefficients, which are calculated from in-situ carbonation depth data. Subsequently, a Bayesian theorem is applied to tailor the training data for each target bridge, which are depending on surrounding environmental conditions. Ultimately, the LSTM model predicts the time-dependent carbonation depth data for the target bridge. To examine the feasibility of this technique, a carbonation depth dataset from 3,960 in-situ bridges was used for training, and untrained time-series data from the Miho River bridge in the Republic of Korea were used for experimental validation. The results of the experimental validation demonstrate a significant reduction in prediction error from 8.19% to 1.75% compared with the conventional statistical method. Furthermore, the LSTM prediction result can be enhanced by sequentially updating the LSTM model using actual time-series measurement data.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.13 no.1
• /
• pp.39-49
• /
• 2013

This paper presents a method of improving the performance of a day-ahead 24-h load curve and peak load forecasting. The next-day load curve is forecasted using radial basis function (RBF) neural network models built using the best design parameters. To improve the forecasting accuracy, the load curve forecasted using the RBF network models is corrected by the weighted sum of both the error of the current prediction and the change in the errors between the current and the previous prediction. The optimal weights (called "gains" in the error correction) are identified by differential evolution. The peak load forecasted by the RBF network models is also corrected by combining the load curve outputs of the RBF models by linear addition with 24 coefficients. The optimal coefficients for reducing both the forecasting mean absolute percent error (MAPE) and the sum of errors are also identified using differential evolution. The proposed models are trained and tested using four years of hourly load data obtained from the Korea Power Exchange. Simulation results reveal satisfactory forecasts: 1.230% MAPE for daily peak load and 1.128% MAPE for daily load curve.

• Design & Manufacturing
• /
• v.7 no.1
• /
• pp.28-33
• /
• 2013

During the cold forming, due to high working pressure acting on the die surface, failure mechanics must be considered before die design. One of the main reasons of die failure in industrial application of metal forming technologies is wear. The mechanisms of wear are consisted of adhesion, abrasion, erosion and so on. Die wear affects the tolerances of formed parts, metal flow, and costs of process. The only way to control these failures is to develop a prediction method on die wear suitable in the design state in order to optimize the process. The wear system is used to analyse 'operating variables' and 'system structure'. In this study, with AISI D2, AISI 1020, AISI 304SS materials, a series of the wear experiments of pin-on-disk type to obtain the wear coefficients from Archard's wear model and the upsetting processes are carried out to observe the wear phenomenon during the cold forming process. The analysis of upsetting processes are performed by the rigid-plastic finite element method. The result of the analysis is used to investigate the die wear the processes, and the analysis simulated die wear profiles are compared with the experimental measured die wear profiles.

• Environmental Analysis Health and Toxicology
• /
• v.16 no.4
• /
• pp.189-196
• /
• 2001

Octanol/water partition coefficients of 52 chemicals were calculated using RP-HPLC estimation method and predicted by computer program, PCHEM. The result showed relationship between literature values and RP-HPLC observed values (relative coefficient r$^2$＝0.916), but the relationship of PCHEM values with literature values was lower than RP-HPLC value (relative coefficient r$^2$＝0.795). The average difference in partition coefficient between the RP-HPLC method and flask-shaking method was log Kow＝0.54, while the average difference between the values predicted form the computer program and flask- shaking method was log Kow = 0.36 Compared to octanol/water partition coefficients by 3 methods (Flask-shaking, RP-HPLC, computer prediction), the octanol/water partition coefficient values based on the flask-shaking method were very similar to the literature values, while the octanol/water partition coefficient values by RP-HPLC method without to consider the dead time, and computer prediction values did not significantly differ with the literature values.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.7
• /
• pp.873-879
• /
• 2010

It is common to predict structural dynamic design parameters due to the change of design parameter, but to predict the amount of changed design parameter where the mass and stiffness are being modified are rarely found in previous literature. In this study, the changed design parameter in a proportional damping system is predicted by using sensitivity coefficients and an iterative method. The sensitivity coefficients are determined from the changes in eigenvectors; these changes are due to modification. This method is applied to a three-story shear structure. To validate the prediction of the changed design parameter, the results are compared to the reanalysis results; both results are in good agreement.

• Journal of the Society of Naval Architects of Korea
• /
• v.52 no.5
• /
• pp.407-417
• /
• 2015

For the more accurate prediction on manoeuvring performance of a ship at initial design phase, bare hull manoeuvring coefficients were estimated by RANS(Reynolds Averaged Navier-Stokes) based virtual captive model tests. Hydrodynamic forces and moment acting on the hull during static drift and harmonic oscillatory motions were computed with a commercial RANS code STAR-CCM+. Automatic and consistent mesh generation could be implemented by using macro functions of the code and user dependency could be greatly reduced. Computed forces and moments on KCS and KVLCC 1&2 were compared with the corresponding measurements from PMM(Planar Motion Mechanism) tests. Quite good agreement can be observed between the CFD and EFD results. Manoeuvring coefficients and IMO standard manoeuvres estimated from the computed data also showed reasonable agreement with those from the experimental data. Based on these results, we could confirm that the developed virtual captive manoeuvring model test process could be applied to evaluate manoeuvrability of a ship at the initial hull design phase.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4
• /
• pp.217-223
• /
• 2005

In this paper, we propose a predictive block constrained trellis coded quantization (BC-TCQ) to quantize cepstral coefficients for the distributed speech recognition. For Prediction of the cepstral coefficients. the 1st order auto-regressive (AR) predictor is used. To quantize the prediction error signal effectively. we use a BC-TCQ. The performance is compared to the split vector quantizers used in the ETSI standard, demonstrating reduction in the cepstral distance and computational complexity.