• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.54 no.1
• /
• pp.34-40
• /
• 2005

This paper presents a frequency transfer function synthesis for simplifying a high-order model with time delay to a low-order model. A model reduction is based on minimizing the error function weighted by the numerator polynomial of reduced systems. The proposed method provides better low frequency fit and a computer aided algorithm. And in this paper, we present a design method of PID controller for achieving the desired specifications via the reduced model. The proposed method identifies the parameter vector of PID controller from a linear system that develops from rearranging the two dimensional input matrices and output vectors obtained from the frequency bounds.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.154.2-154
• /
• 2001

This paper presented a adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when ...

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.99.1-99
• /
• 2001

This paper presented an adaptation approach based on maximum model distance (MMD) method. This method shares the same framework as they are used for training speech recognizers with abundant training data. The MMD method could adapt to all the models with or without adaptation data. If large amount of adaptation data is available, these methods could gradually approximate the speaker-dependent ones. The approach is evaluated through the phoneme recognition task on the TIMIT corpus. On the speaker adaptation experiments, up to 65.55% phoneme error reduction is achieved. The MMD could reduce phoneme error by 16.91% even when only one adaptation utterance is used.

• Journal of Environmental Science International
• /
• v.27 no.7
• /
• pp.499-507
• /
• 2018

Agricultural meteorological information is an important resource that affects farmers' income, food security, and agricultural conditions. Thus, such data are used in various fields that are responsible for planning, enforcing, and evaluating agricultural policies. The meteorological information obtained from automatic weather observation systems operated by rural development agencies contains missing values owing to temporary mechanical or communication deficiencies. It is known that missing values lead to reduction in the reliability and validity of the model. In this study, the hierarchical Bayesian spatio-temporal model suggests replacements for missing values because the meteorological information includes spatio-temporal correlation. The prior distribution is very important in the Bayesian approach. However, we found a problem where the spatial decay parameter was not converged through the trace plot. A suitable spatial decay parameter, estimated on the bias of root-mean-square error (RMSE), which was determined to be the difference between the predicted and observed values. The latitude, longitude, and altitude were considered as covariates. The estimated spatial decay parameters were 0.041 and 0.039, for the spatio-temporal model with latitude and longitude and for latitude, longitude, and altitude, respectively. The posterior distributions were stable after the spatial decay parameter was fixed. root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and bias were calculated for model validation. Finally, the missing values were generated using the independent Gaussian process model.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.3
• /
• pp.184-189
• /
• 2006

This paper deals with a fractional model reduction for T-S fuzzy systems with time varying delayed states. A contractive coprime factorization of time delayed T-S fuzzy systems is defined and obtained by solving linear matrix inequalities. Using generalized controllability and observability gramians of the contractive coprime factor, a balanced state space realization of the system is derived. The reduced model will be obtained by truncating states in the balanced realization and an upper bound of model approximation error is also presented. In order to demonstrate efficacy of the suggested method, a numerical example is performed.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.41 no.2
• /
• pp.29-36
• /
• 2004

This paper deals with a fractional model reduction for linear systems with time varying delayed states. A contractive coprime factorization of linear time delayed systems is defined and obtained by solving linear matrix inequalities. Using generalize controllability and observability gramians of tile contractive coprime factor, a balanced state space realization of the system is derived. The reduced model will be obtained by truncating states in the balanced realization and an upper bound of model approximation error is also presented. In order to demonstrate efficacy of the suggested method, a numerical example is illustrated.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.40 no.5
• /
• pp.326-332
• /
• 2003

This paper deals with a model reduction for linear systems with time varying delayed states. A generalized controllability and observability gramians are defined and obtained by solving linear matrix inequalities. Using the generalized controllability and observability gramians, the balanced state space equation is realized. The reduced model can be obtained by truncating states in the balanced realization and the upper bound of model approximation error is also presented. In order to demonstrate efficacy of the suggested method, a numerical example is performed.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.3
• /
• pp.749-754
• /
• 2015

This article presents Bayesian approach to regression splines with knots on a grid of equally spaced sample quantiles of the independent variables under functional measurement error model.We consider small area model by using penalized splines of non-linear pattern. Specifically, in a basis functions of the regression spline, we use radial basis functions. To fit the model and estimate parameters we suggest a hierarchical Bayesian framework using Markov Chain Monte Carlo methodology. Furthermore, we illustrate the method in an application data. We check the convergence by a potential scale reduction factor and we use the posterior predictive p-value and the mean logarithmic conditional predictive ordinate to compar models.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.436-441
• /
• 2008

One of important factors in designing array-type MEMS resonators is obtaining a desired frequency response function (FRF) within a specific range. In this paper Krylov subspace-based model order reduction using moment-matching with non-zero expansion points is represented to calculate the FRF of array-type resonators. By matching moments at a frequency around a specific range of the array-type resonators, required FRFs can be efficiently calculated with significantly reduced systems regardless of their operating frequencies. In addition, because of the characteristics of moment-matching method, a minimal order of reduced system with a specified accuracy can be determined through an error indicator using successive reduced models, which is very useful to automate the order reduction process and FRF calculation for structural optimization iterations.

• Structural Engineering and Mechanics
• /
• v.43 no.3
• /
• pp.371-394
• /
• 2012

Accurate finite element (FE) models are needed in many applications of Civil Engineering such as health monitoring, damage detection, structural control, structural evaluation and assessment. Model accuracy depends on both the model structure (the form of the equations) and the model parameters (the coefficients of the equations), and can be generally improved through that process of experimental reconciliation known as model updating. However, modelling errors, including (i) errors in the model structure and (ii) errors in parameters excluded from adjustment, may bias the solution, leading to an updated model which replicates measurements but lacks physical meaning. In this paper, an application of ambient-vibration-based model updating to a large-scale benchmark prototype of a building structure is reported in which both types of error are met. The error in the model structure, originating from unmodelled secondary structural elements unexpectedly working as resonant appendages, is faced through a reduction of the experimental modal model. The error in the model parameters, due to the inevitable constraints imposed on parameters to avoid ill-conditioning and under-determinacy, is faced through a multi-model parameterization approach consisting in the generation and solution of a multitude of models, each characterized by a different set of updating parameters. Results show that modelling errors may significantly impair updating even in the case of seemingly simple systems and that multi-model reasoning, supported by physical insight, may effectively improve the accuracy and robustness of calibration.