Sharif Digital Repository

In this paper, a framework for distributed and decentralized state estimation in high-pressure and long-distance gas transmission networks (GTNs) is proposed. The non-isothermal model of the plant including mass, momentum and energy balance equations are used to simulate the dynamic behavior. Due to several disadvantages of implementing a centralized Kalman filter for large-scale systems, the continuous/discrete form of extended Kalman filter for distributed and decentralized estimation (DDE) has been extended for these systems. Accordingly, the global model is decomposed into several subsystems, called local models. Some heuristic rules are suggested for system decomposition in gas pipeline... 

In this paper, a non-isothermal model of natural gas in pipelines including mass, momentum and energy balance equations are used as model equations for modeling and state estimation in gas pipeline systems. It is shown that differential equations describing the dynamic behavior of a high-pressure and long-distance gas transmission network (GTN) can be solved efficiently using the orthogonal collocation method. The issues corresponding to the presence of discontinuities in the dynamic model is substantially discussed and studied. The non-isothermal model of a GTN can experience discontinuities during transient operations, which causes challenges in simulation and state estimation in this... 

In natural gas transmission networks, the efficiency of daily operation is strongly dependent on our knowledge about the customer future demands. Unavailability of accurate demand forecasts makes it more important to be able to characterize the loads and quantify the corresponding uncertainty. A planning strategy for natural gas transmission networks under future demand uncertainty is addressed, which involves coupling of a gas transmission network dynamic simulator with the stochastic optimization framework. Loads from a gas-fired power plant are studied where the loads are characterized by a number of uncertain parameters, and unscented transform is utilized for uncertainty propagation.... 

The activity coefficient at infinite dilution is a representative of the limiting non-ideality of a solute in a mixture. Various methods for the prediction of infinite dilution activity coefficients (IDACs) have been developed. Artificial neural networks are powerful mapping tools for nonlinear function approximations. Accordingly, an artificial neural network model is proposed for the prediction of the IDACs of binary systems where the properties of the individual components are used as inputs to the network. The input parameters of the neural network are the mixture temperature, critical temperature, critical pressure, critical volume, molecular weight, dipole moment and the acentric... 

This paper presents the results of an experimental and analytical study on high-strength, reinforced-concrete (RC) columns with different levels of initial ductility enhanced with carbon fiber reinforced polymer (CFRP) wraps. Six square columns 260 mm wide and 1650 mm long were tested under constant axial load and reversed cyclic lateral load. The test specimens were divided into three groups. Groups I and II were designed and detailed according to the requirements of American Concrete Institute ACI318-02 for intermediate-moment frames. Group I was wrapped with CFRP, group II was control specimens, and group III was designed and detailed according to the requirements of ACI318-02 for... 

In the present work, specimens were cut out from St-37 plates with 19 mm thickness. The thickness of plates was reduced to 12.5 mm by milling and grinding operations. Then a standard V-shaped fillet was made on one edge of the plates. Two plates were butt-welded by standard metal arc gas (MAG) welding process. Residual stresses induced by welding were measured on 20 specimens by centre hole drilling. Load controlled axial fatigue tests were carried out to determine the fatigue life of specimens. Crack growth rates were obtained from experiment. Fractography of specimens was performed. Genetic Algorithm (GA) was employed for prediction of residual stress value in weldments using the crack... 

The qualitative effects of residual stress parameters on fatigue crack growth are investigated using Monte Carlo probabilistic analysis technique. These parameters include the maximum residual stress,σ0, half-width of the tensile residual stress region, b, and the redistribution of residual stress parameter, β. A normal distribution is assumed for all parameters considered in the investigation and fatigue life is determined using Forman equation. The results indicate that: (i) fatigue life decreases nonlinearly if the mean of random maximum residual stress and the half-width of tensile residual stress region increases; (ii) fatigue life increases nonlinearly if the rate of residual stress... 

In this paper residual stresses induced by butt metal arc gas welding (MAGW), in accordance with ASTM E837, are determined by hole drilling. The results are analyzed by probabilistic methods. Also, the effect of fatigue life on relaxation of welding residual stress is investigated. The specimens are tested to measure fatigue life and crack growth rates in accordance with ASTM E466 for force controlled amplitude axial fatigue testing. © 2008 Verlag Stahleisen GmbH, Düsseldorf  

Tracking by detection is the topic of recent research that has received considerable attention in computer vision community. Mainly off-line classification methods have been used, however, they perform weakly in the case of appearance changes. Training theclassifier incrementally and in an online manner solves this problem, but nevertheless, raises drifting due to soft or hard labeling in the online adaptation. In this paper a novel semi-supervised online tracking algorithm based on manifold assumption is proposed. Target object and background patches lie near low-dimensional manifolds. This motivates us to make use of the intrinsic structure of data in classification, and benefit from the... 

During recent years, active learning has evolved into a popular paradigm for utilizing user's feedback to improve accuracy of learning algorithms. Active learning works by selecting the most informative sample among unlabeled data and querying the label of that point from user. Many different methods such as uncertainty sampling and minimum risk sampling have been utilized to select the most informative sample in active learning. Although many active learning algorithms have been proposed so far, most of them work with binary or multi-class classification problems and therefore can not be applied to problems in which only samples from one class as well as a set of unlabeled data are...