• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.592-597
• /
• 2003

In this paper, some differential geometric conditions for the observer using a recurrent neural network are provided in terms of a stochastic nonlinear system control. In the stochastic nonlinear system, it is necessary to make an additional condition for observation of stochastic nonlinear system, called perfect filtering condition. In addition, we provide a observer using a recurrent neural network for the observation of a stochastic nonlinear system with the proposed observation conditions. Computer simulation shows that the control performance of the stochastic nonlinear system with a observer using a recurrent neural network satisfying the proposed conditions is more efficient than the conventional observer as Kalman filter

• Structural Engineering and Mechanics
• /
• v.32 no.1
• /
• pp.71-94
• /
• 2009

In this paper a procedure for Monte Carlo simulation of univariate stationary stochastic processes with the aid of neural networks is presented. Neural networks operate model-free and, thus, circumvent the need of specifying a priori statistical properties of the process, as needed traditionally. This is particularly advantageous when only limited data are available. A neural network can capture the "pattern" of a short observed time series. Afterwards, it can directly generate stochastic process realizations which capture the properties of the underlying data. In the present study a simple feed-forward network with focused time-memory is utilized. The proposed procedure is demonstrated by examples of Monte Carlo simulation, by synthesis of future values of an initially short single process record.

• Journal of the Korea Society for Simulation
• /
• v.9 no.2
• /
• pp.89-102
• /
• 2000

In this paper, we consider an optimal allocation of constant service efforts in queueing network to maximize the system throughput. For this purpose, using the perturbation analysis, we apply a stochastic optimization algorithm to two types of queueing systems. Our simulation results indicate that the estimates obtained from a stochastic optimization algorithm for a two-tandem queuing network are very accurate, and those for closed loop manufacturing system are a little apart from the known optimal allocation. We find that as simulation time increases for obtaining a new gradient (performance measure with respect to decision variables) by perturbation algorithm, the estimates tend to be more stable. Thus, we consider that it would be more desirable to have more accurate sensitivity of performance measure by enlarging simulation time rather than more searching steps with less accurate sensitivity. We realize that more experiments on various types of systems are needed to identify such a relationship with regards to stopping rule, the size of moving step, and updating period for sensitivity.

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.18-21
• /
• 2001

In this paper, some geometric condition for a stochastic nonlinear system and an adaptive control method for minimum-phase stochastic nonlinear system using neural network me provided. The state feedback linearization is widely used technique for excluding nonlinear terms in nonlinear system. However, in the stochastic environment, even if the minimum phase linear system derived by the feedback linearization is not sufficient to be controlled robustly. In the viewpoint of that, it is necessary to make an additional condition for observation of nonlinear stochastic system, called perfect filtering condition. In addition, on the above stochastic nonlinear observation condition, I propose an adaptive control law using neural network. Computer simulation shoo's that the stochastic nonlinear system satisfying perfect filtering condition is controllable and the proposed neural adaptive controller is more efficient than the conventional adaptive controller.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2000.11a
• /
• pp.16-21
• /
• 2000

본 연구는 대형 프로젝트의 위험분석을 위한 확률적 Network 시뮬레이션모델의 연구로서 Simulation방법으로 프로젝트의 성공 및 실패확률을 산정 하였다. 프로젝트의 주요 불확실성 요소(Uncertainty Factors)인 프로젝트의 수행기간(Time), 비용(Cost) 및 성과(Performance) 등의 계획은 실패 없이 추진되어야 하는 것이 중요하다. 연구 개발 및 신기술개발과 같이 대형 프로젝트의 경우, 그 성과 달성의 위험(Risk)성은 매우 크며 이러한 위험 예측 및 분석이 프로젝트의 성공적인 수행을 위하여 매우 중요 시 된다. 본 연구에서는 이를 위한 위험분석(Risk Analysis)의 방법으로 일반적으로 쉽게 사용할 수 있는 위험요인법(Risk Factor Analysis)과 확률적 Network 시뮬레이션모델을 제시하였으며 또한 이를 위한 Simulation프로그램을 개발하였으며 이를 신 기술개발 프로젝트에 응용하는 과정을 보였다. 본 연구에서 개발된 관련 프로그램을 보완 할 경우 대형 프로젝트의 각종 의사결정 시에 매우 유용하게 활용될 수 있으리라 생각된다.

• Structural Engineering and Mechanics
• /
• v.78 no.5
• /
• pp.529-543
• /
• 2021

Inevitable source-uncertainties in geometry configuration, boundary condition, and material properties may deviate the structural dynamics from its expected responses. This paper aims to examine the influence of these uncertainties on the vibration of functionally graded beams. Finite element procedures are presented for Timoshenko beams and utilized to generate reliable datasets. A prerequisite to the uncertainty quantification of the beam vibration using Monte Carlo simulation is generating large datasets, that require executing the numerical procedure many times leading to high computational cost. Utilizing artificial neural networks to model beam vibration can be a good approach. Initially, the optimal network for each beam configuration can be determined based on numerical performance and probabilistic criteria. Instead of executing thousands of times of the finite element procedure in stochastic analysis, these optimal networks serve as good alternatives to which the convergence of the Monte Carlo simulation, and the sensitivity and probabilistic vibration characteristics of each beam exposed to randomness are investigated. The simple procedure presented here is efficient to quantify the uncertainty of different stochastic behaviors of composite structures.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.18-18
• /
• 2000

In this paper, some geometric condition for a stochastic nonlinear system and an adaptive control method for minimum-phase stochastic nonlinear system using neural network are provided. The state feedback linearization is widely used technique for excluding nonlinear terms in nonlinear system. However, in the stochastic environment, even if the minimum phase linear system derived by the feedback linearization is not sufficient to be controlled robustly. the viewpoint of that, it is necessary to make an additional condition for observation of nonlinear stochastic system, called perfect filtering condition. In addition, on the above stochastic nonlinear observation condition, I propose an adaptive control law using neural network. Computer simulation shows that the stochastic nonlinear system satisfying perfect filtering condition is controllable and the proposed neural adaptive controller is more efficient than the conventional adaptive controller

• Korean Journal of Construction Engineering and Management
• /
• v.6 no.1 s.23
• /
• pp.73-79
• /
• 2005

This paper, introduces a Stochastic Project Scheduling Simulation system (SPSS III) developed by the author to predict a project completion probability in a certain time. The system integrates deterministic CPM, probabilistic PERT, and stochastic Discrete Event Simulation (DES) scheduling methods into one system. It implements automated statistical analysis methods for computing the minimum number of simulation runs, the significance of the difference between independent simulations, and the confidence interval for the mean project duration as well as sensitivity analysis method in What-if analyzer component. The SPSS 111 gives the several benefits to researchers in that it (1) complements PERT and Monte Carlo simulation by using stochastic activity durations via a web based JAVA simulation over the Internet, (2) provides a way to model a project network having different probability distribution functions, (3) implements statistical analyses method which enable to produce a reliable prediction of the probability of completing a project in a specified time, and (4) allows researchers to compare the outcome of CPM, PERT and DES under different variability or skewness in the activity duration data.

• Korea-Australia Rheology Journal
• /
• v.14 no.4
• /
• pp.161-174
• /
• 2002

A new technique for numerical calculation of viscoelastic flow based on the combination of Neural Net-works (NN) and Brownian Dynamics simulation or Stochastic Simulation Technique (SST) is presented in this paper. This method uses a "universal approximator" based on neural network methodology in combination with the kinetic theory of polymeric liquid in which the stress is computed from the molecular configuration rather than from closed form constitutive equations. Thus the new method obviates not only the need for a rheological constitutive equation to describe the fluid (as in the original Calculation Of Non-Newtonian Flows: Finite Elements St Stochastic Simulation Techniques (CONNFFESSIT) idea) but also any kind of finite element-type discretisation of the domain and its boundary for numerical solution of the governing PDE's. As an illustration of the method, the time development of the planar Couette flow is studied for two molecular kinetic models with finite extensibility, namely the Finitely Extensible Nonlinear Elastic (FENE) and FENE-Peterlin (FENE-P) models.P) models.

• ETRI Journal
• /
• v.33 no.5
• /
• pp.704-711
• /
• 2011

This paper proposes hybrid distributed stochastic addressing (HDSA), which combines the advantages of distributed addressing and stochastic addressing, to solve the problems encountered when constructing a network in a ZigBee-based wireless sensor network. HDSA can assign all the addresses for ZigBee beyond the limit of addresses assigned by the existing distributed address assignment mechanism. Thus, it can make the network scalable and can also utilize the advantages of tree routing. The simulation results reveal that HDSA has better addressing performance than distributed addressing and better routing performance than other on-demand routing methods.