• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.3
• /
• pp.277-288
• /
• 2014

This article reviews various state estimation methods for nonlinear systems, particularly with a perspective of a process control engineer. Nonlinear state estimation methods can be classified into the following two categories: stochastic approaches and deterministic approaches. The current review compares the Bayesian approach, which is mainly a stochastic approach, and the MHE (Moving Horizon Estimation) approach, which is mainly a deterministic approach. Though both methods are reviewed, emphasis is given to the latter as it is particularly well-suited to highly nonlinear systems with slow sampling rates, which are common in chemical process applications. Recent developments in underlying theories and supporting numerical algorithms for MHE are reviewed. Thanks to these developments, applications to large-scale and complex chemical processes are beginning to show up but they are still limited at this point owing to the high numerical complexity of the method.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.5
• /
• pp.839-844
• /
• 2008

A novel state of health estimation method for hybrid electric vehicle lithium battery using sliding mode observer has been presented. A simple R-C circuit method has been used for the lithium battery modeling for the reduced calculation time and system resources due to the simple matrix operations. The modeling errors of simple model are compensated by the sliding mode observer. The design methodology for state of health estimation using dual sliding mode observer has been presented in step by step. The structure of the proposed system is simple and easy to implement, but it shows robust control property against modeling errors and temperature variations. The convergence of proposed observer system has been proved by the Lyapunov inequality equation and the performance of system has been verified by the sequence of urban dynamometer driving schedule test. The test results show the proposed observer system has superior tracking performance with reduced calculation time under the real driving environments.

• Proceedings of the KIEE Conference
• /
• 1996.11a
• /
• pp.182-185
• /
• 1996

The purpose of slate estimation in power system is to estimate the best-fit Slate variables from the measurements contaminated by various kind of noise. But because the majority of state estimation modules in EMS lack the convergence characteristics, sometimes the desirable outputs can't be obtained. So, in this paper, the new algorithm using the load now output as initial values in the state estimation calculation is proposed to guarantee the convergence. And if the load now outputs were used as the initial values in the calculation, the change in each step would be small compared to the original method using the flat start point. And the Inverse Lemma is used in the algorithm to calculate the new stale in each iteration step for reducing the calculation time. The proposed algorithm was tested on the IEEE 14, 30, 118 bus systems. Eventually, we were able to verity that the differences between the results obtained by the original method and proposed method were relatively small, and the effectiveness of the proposed algorithm increased when applied to the bigger systems.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.39 no.6
• /
• pp.598-606
• /
• 1990

Common problems encountered when orthogonal functions are used in system analysis and state estimation are the time consuming process of high order matrix inversion required in finding the Kronecker products and the truncation errors. In this paper, therefore, a method for the analysis of bilinear systems using Walsh, Block pulse, and Haar functions is devised, Then, state estimation of bilinear system is also studied based on single term expansion of orthogonal functions. From the method presented here, when compared to the other conventional methods, we can obtain the results with simpler computation as the number of interval increases, and the results approach the original function faster even at randomly chosen points regardless of the definition of intervals. In addition, this method requires neither the inversion of large matrices on obtaining the expansion coefficients nor the cumbersome procedures in finding Kronecker products. Thus, both the computing time and required memory size can be significantly reduced.

• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.1466-1472
• /
• 2007

State estimation is to estimate the values of the states that minimize the error between the real states and the measured states, which are usually hampered by noise. It exploits the redundant data and the equality constraints achieved from the power systems. In the electric railway feeding systems, especially, the measured states may have significant level of noise in comparison with the commercial power systems. Since the meters - the sources of the data that include vehicles - are distributed in the long distance along the railroad, they are vulnerable to the signal interference. In this paper we have studied the application of state estimation method to the AT feeding systems and shown that this method can increase the reliability of the measured data.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.28 no.11
• /
• pp.77-83
• /
• 2014

The Estimation of State of Charge(SOC) for batteries is an important aspect of a Battery Management System(BMS). A method for estimating the SOC is proposed in order to overcome the individual disadvantages of the current integral and Open Circuit Voltage(OCV) estimation methods by combining them using Extended Kalman filter(EKF). The non-linear characteristics of the Li-Ion RC battery model used in this study is also solved through EKF. The proposed method is simulated in a Matlab environment with a Li-Ion Kokam battery (3.7V, 1,500mAh). Results showed that there is an improvement in the estimation error when using the proposed model compared to the conventional current integral method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.6
• /
• pp.487-493
• /
• 2001

In this paper, a new reduced-order dynamic observer method is presented. Two types of observers are pronounced, namely, the model based reduced-order dynamic observer and the Luenburger type reduced-order dynamic observer. Useful design algorithms are also provided for each structure. The essential features of the proposed observed design methods are addressed to be qualified ad effective observers. The proposed method clarifies the duality between the controller and observer designs.

• Communications for Statistical Applications and Methods
• /
• v.29 no.1
• /
• pp.1-25
• /
• 2022

A modified Bagdonavičius-Nikulin chi-square goodness-of-fit is defined and studied. The lymphoma data is analyzed using the modified goodness-of-fit test statistic. Different non-Bayesian estimation methods under complete samples schemes are considered, discussed and compared such as the maximum likelihood least square estimation method, the Cramer-von Mises estimation method, the weighted least square estimation method, the left tail-Anderson Darling estimation method and the right tail Anderson Darling estimation method. Numerical simulation studies are performed for comparing these estimation methods. The potentiality of the new model is illustrated using three real data sets and compared with many other well-known generalizations.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1469-1472
• /
• 1997

For system state estimation, existing LMS type esimators widely used. For example Kalman filter is one of them. In this paper, a state estimator is derived for the H$_{\infty}$ norm of the estimation error spectrum matrix to be minimized. For the solution of this problem classical NP interpolation problem is used. Also, by deriving the duality between the filter problem and the well-known H$_{\infty}$ control problem, another solution is obtained. The computer simuation results show that H$_{\infty}$ estimator has less estimation error and so this is better than the existing Kalman filter estimator.or.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.566-578
• /
• 2018

In this paper, we design a segment training based individual channel estimation (STICE) scheme for the classical two-way relay network (TWRN) with multi-pair sources (MPS) and amplify-and-forward (AF). We adopt the linear minimum mean square error (LMMSE) channel estimator to minimize the mean square error (MSE) without channel estimation error, where the optimal power allocation strategy from the relay for different sources is obtained. Then the MSE gains are given with different source pairs among the proposed power allocation scheme and the existing power allocation schemes. Numerical results show that the proposed method outperforms the existing ones.