• Transactions on Control, Automation and Systems Engineering
• /
• v.2 no.1
• /
• pp.40-46
• /
• 2000

In this paper, a transient identification based on a Hidden Markov Model (HMM) has been suggested and evaluated experimentally for the classification of transients in the dynamic process. The transient can be identified by its unique time dependent patterns related to the principal variables. The HMM, a double stochastic process, can be applied to transient identification which is a spatial and temporal classification problem under a statistical pattern recognition framework. The HMM is created for each transient from a set of training data by the maximum-likelihood estimation method. The transient identification is determined by calculating which model has the highest probability for the given test data. Several experimental tests have been performed with normalization methods, clustering algorithms, and a number of states in HMM. Several experimental tests have been performed including superimposing random noise, adding systematic error, and untrained transients. The proposed real-time transient identification system has many advantages, however, there are still a lot of problems that should be solved to apply to a real dynamic process. Further efforts are being made to improve the system performance and robustness to demonstrate reliability and accuracy to the required level.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.16 no.4 s.109
• /
• pp.394-403
• /
• 2006

A new equivalent linearization technique is proposed for a friction damper-brace system (FDBS) idealized as a elastoplastic system. The equivalent linearization technique utilizes secant stiffness and dissipated energy defined by the probability distribution of the extremal displacement of the FDBS. In addition, a conversion scheme is proposed so that an equivalent linear system is designed first and converted to the FDBS. For comparative study, an existing model update technique based on system identification is modified in a form appropriate to update single element. For the purpose of verification, shaking table tests of a small scale three-story shear building model, in which a rotational FDBS is installed, are conducted and equivalent linear systems are obtained using the proposed technique and the model update technique. Complex eigenvalue analysis is conducted for those equivalent linear systems, and the natural frequencies and modal damping ratios are compared with those obtained from system identification. Additionally, RMS and peak responses obtained from time history analysis of the equivalent linear systems are compared with measured ones.

• 전기의세계
• /
• v.22 no.3
• /
• pp.49-62
• /
• 1973

An algorithm for the state estimation and identification of multivariable nonlinear systems with noisy nonlinear observation has been investigated on the basis of the multidimensional Hermitian expansion for the a posteriori probability densities of the predicted observation, the predicted state and the observation conditioned by the state. A new approach for construction of this sequential nonlinear estimator, retaining up to the second order term of the observation error, has been developed, along with the approximation of nonlinear system functions, truncating at the second term. The estimation of the unknown parameters has been established by extending the state estimation technique, regarding the parameters as another state variables. The results of investigation indicate the feasibility of the schemes presented in this paper.

• Smart Structures and Systems
• /
• v.15 no.3
• /
• pp.735-749
• /
• 2015

The major difficulty of using Bayesian probabilistic inference for system identification is to obtain the posterior probability density of parameters conditioned by the measured response. The posterior density of structural parameters indicates how plausible each model is when considering the uncertainty of prediction errors. The Markov chain Monte Carlo (MCMC) method is a widespread medium for posterior inference but its convergence is often slow. The differential evolution adaptive Metropolis-Hasting (DREAM) algorithm boasts a population-based mechanism, which nms multiple different Markov chains simultaneously, and a global optimum exploration ability. This paper proposes an improved differential evolution adaptive Metropolis-Hasting algorithm (IDREAM) strategy to estimate the posterior density of structural parameters. The main benefit of IDREAM is its efficient MCMC simulation through its use of the adaptive Metropolis (AM) method with a mutation strategy for ensuring quick convergence and robust solutions. Its effectiveness was demonstrated in simulations on identifying the structural parameters with limited output data and noise polluted measurements.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.12A
• /
• pp.1177-1184
• /
• 2008

Non-line-of-sight (NLOS) propagation can severely weaken the accuracy of ranging and localization in wireless location systems. NLOS bias mitigation techniques have recently been proposed to relieve the NLOS effects, but positively rely on the capability to accurately distinguish between LOS and NLOS propagation scenarios. This paper proposes an energy-capture-based NLOS identification method for LDR-UWB systems, based on the analysis of the characteristics of the channel impulse response (CIR). With this proposed energy capture method, the probability of successfully identifying NLOS is much improved than the existing methods, such as the kurtosis method, the strongest path compare method, etc. This NLOS identification method can be employed in adaptive modulation scheme to decrease bit error ratio (BER) level for certain signal-to-noise ratio (SNR). The BER performance with the adaptive modulation can be significantly enhanced by selecting proper modulation method with the knowledge of channel information from the proposed NLOS identification method.

• MALSORI
• /
• no.57
• /
• pp.165-174
• /
• 2006

In this paper we use a training algorithm, MCE (Minimum Classification Error), to improve the performance of a text-independent speaker identification system. The MCE training scheme takes account of possible competing speaker hypotheses and tries to reduce the probability of incorrect hypotheses. Experiments performed on a small set speaker identification task show that the discriminant training method using MCE can reduce identification errors by up to 54% over a baseline system trained using Bayesian adaptation to derive GMM (Gaussian Mixture Models) speaker models from a UBM (Universal Background Model).

• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.749-757
• /
• 2007

Recently, with increasing interested in improvement of operational reliability and the systematic maintenance activities, the RCM analysis has been applied and tried to lots of applicable industries. This study covers applying the probability of failure cause to FMEA, and proposes an analytical method for this. Also, the measures of quantitative classification for the result of failure cause probability are addressed. Based on the field data, this thesis presents an identification for causes and characteristics of failure, and reviews them periodically from the above methodologies. As using FMEA applied the probability of failure cause, we in the future can look forward to improvement of efficiency for failure diagnosis & inspection, and reliability.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.23 no.4
• /
• pp.16-41
• /
• 2020

Geospatial input setting to represent the reality of spatial distribution or quantitative property within model has become a major interest in earth system simulation. Many studies showed the variation of grid resolution could lead to drastic changes of spatial model results because of insufficient surface property estimations. Hence, in this paper, the authors proposed Monte Carlo Integration (MCI) to apply spatial probability (SP) in a spatial-sampling framework using a three-dimensional point cloud (3DPC) to keep the optimized spatial distribution and area/volume property of buildings in urban area. Three different decision rule based building identification results were compared : SP threshold, cell size, and 3DPC density. Results shows the identified building area property tend to increase according to the spatial sampling grid area enlargement. Hence, areal building property manipulation in the sampling frameworks by using decision rules is strongly recommended to increase reliability of geospatial modeling and analysis results. Proposed method will support the modeling needs to keep quantitative building properties in both finer and coarser grids.

• Journal of Environmental Science International
• /
• v.19 no.9
• /
• pp.1153-1159
• /
• 2010

Identification of probability distribution for water quality constituents from specific land use is important to achieve successful implementation of TMDL program. In this 3-year study, distribution of discharge and BOD(Biological Oxygen Demand) concentration from paddy rice fields were monitored. Four probability distributions, normal, log-normal, Gamma and Weibull were fitted and the goodness-of-fit was assessed using probability plots and Kolmogorov-Smirnov test. $EMC_s$ of BOD in runoff from paddy field ranged 0.37 to $7.99\;mgL^{-1}$, and all four probability distributions were acceptable. But the normal distribution would be preferred for BOD from paddy fields considering nature of straight forward application.

• Journal of the Korean Statistical Society
• /
• v.28 no.4
• /
• pp.503-513
• /
• 1999

The Bayes factor for the identification of stationary ARM(p,q) models is exactly computed using the Monte Carlo method. As priors are used the uniform prior for (\ulcorner,\ulcorner) in its stationarity-invertibility region, the Jefferys prior and the reference prior that are noninformative improper for ($\mu$,$\sigma$\ulcorner).