Sharif Digital Repository

Performance-based optimal seismic design of structures requires repetitive and massive dynamic analyses. In the Endurance Time (ET) method, structures are subjected to specially designed intensifying acceleration functions and their response at various excitation levels is estimated by each single response-history analysis, thus, significantly reducing the required computational demand. In this paper, the ET method is utilized to acquire optimal placement of viscous dampers in short steel frames with the target of obtaining the desirable performance in two hazard levels simultaneously. In the optimization procedure, the damping coefficients of the added dampers are treated as design... 

A practical optimization method is applied to design nonlinear truss-like structures subjected to seismic excitation. To achieve minimum weight design, inefficient material is gradually shifted from strong parts to weak parts of a structure until a state of uniform deformation prevails. By considering different truss structures, effects of seismic excitation, target ductility and buckling of the compression members on optimum topology are investigated. It is shown that the proposed method could lead to 60% less structural weight compared to optimization methods based on elastic behavior and equivalent static loads, and is efficient at controlling performance parameters under a design... 

Theoretical study of nitrogen monoxide adsorption on small Six (x = 3-5) clusters has been carried out using the advanced hybrid meta-density functional method of Truhlar (MPW1B95). MG3 semi-diffuse basis sets were employed to improve the results. The geometry, adsorption energy, natural bond orbital charge, natural population analysis (NPA)-derived spin density and vibrational frequency of NO adsorption on all optimized nanoclusters were investigated. Also using the NPA, we have investigated the change of bond orders through adsorption. It has been found that NO is capable of making n-centre bonds (n = 1-4) from the nitrogen side but bonds to one site from the oxygen end. In the later case... 

The problem of clustering with side information has received much recent attention and metric learning has been considered as a powerful approach to this problem. Until now, various metric learning methods have been proposed for semi-supervised clustering. Although some of the existing methods can use both positive (must-link) and negative (cannot-link) constraints, they are usually limited to learning a linear transformation (i.e., finding a global Mahalanobis metric). In this paper, we propose a framework for learning linear and non-linear transformations efficiently. We use both positive and negative constraints and also the intrinsic topological structure of data. We formulate our metric... 

It has been well documented that multivariate systems will benefit from application of advanced control strategies. The application of these methods has thus far been hindered for a variety of reasons which range from complexity of the design methodologies to the cost associated with implementation of these solutions. This paper shows that there exist alternative means by which existing control strategies of a multivariate system could be extended, thereby leading to increased efficiency, without the need to remove and implement a new control system installation. A realistic model of an entire pulping mill is used in this paper to show that by careful selection of only a few additional... 

Distance metric plays an important role in many machine learning algorithms. Recently, there has been growing interest in distance metric learning for semi-supervised setting. In the last few years, many methods have been proposed for metric learning when pairwise similarity (must-link) and/or dissimilarity (cannot-link) constraints are available along with unlabeled data. Most of these methods learn a global Mahalanobis metric (or equivalently, a linear transformation). Although some recently introduced methods have devised nonlinear extensions of linear metric learning methods, they usually allow only limited forms of distance metrics and also can use only similarity constraints. In this... 

There is no method to determine the optimal topology for multi-layer neural networks for a given problem. Usually the designer selects a topology for the network and then trains it. Since determination of the optimal topology of neural networks belongs to class of NP-hard problems, most of the existing algorithms for determination of the topology are approximate. These algorithms could be classified into four main groups: pruning algorithms, constructive algorithms, hybrid algorithms and evolutionary algorithms. These algorithms can produce near optimal solutions. Most of these algorithms use hill-climbing method and may be stuck at local minima. In this article, we first introduce a... 

Recently, many meta-heuristic algorithms are proposed for optimization of various problems. Some of them originally are presented for continuous optimization problems and some others are just applicable for discrete ones. In the literature, sizing optimization of truss structures is one of the discrete optimization problems which is solved by many meta-heuristic algorithms. In this paper, in order to discover an efficient and reliable algorithm for optimization of truss structures, a discrete optimizer, entitled Soccer League Competition (SLC) algorithm and ten popular and powerful solvers are examined and statistical analysis is carried out for them. The fundamental idea of SLC algorithm is... 

The optimal design of a casting feeding system is considered. The riser topology is systematically modified to minimize the riser volume, while simultaneously ensuring that no defect appears in the product. In this approach, we combine finite-difference analysis of the solidification process with evolutionary topology optimization to systematically improve the feeding system design. The outstanding features of the presented method are: its efficiency, ease of implementation and simplifying definition of the initial design. The efficiency and capability of the presented method are supported by illustrative examples. © 2008 Springer-Verlag  

The purpose of this study is to apply managerial economics and methods of decision analysis to study the optimal pattern of innovation activities for development of new energy technologies in developing countries. For this purpose, a model of energy research and development (R&D) planning is developed and it is then linked to a bottom-up energy-systems model. The set of interlinked models provide a comprehensive analytical tool for assessment of energy technologies and innovation planning taking into account the specific conditions of developing countries. An energy-system model is used as a tool for the assessment and prioritization of new energy technologies. Based on the results of the... 

The optimal feeder design in gravity casting processes is considered. The design problem is formulated as an additive, growth-like, evolutionary topology optimization problem. The initial design of the presented method is a casting part which is placed in a suitable environmental mould box. The method is composed of three stages: determining the feeder-neck connection point on the casting surface, constructing the feeder-neck and the feeder growth. During the growth stage, the feeder topology is changed gradually until satisfying some predefined criteria. The design of multiple feeders is also considered in this work. In the presented method, feeders are designed in a descending order of... 

New penalty functions, which have better convergence properties, as compared to the commonly used exterior and interior penalty functions, have been proposed in this paper. The convergence behavior and accuracy of ordinary penalty functions depend on the selection of appropriate penalty parameters. The optimization of ordinary penalty functions is accomplished after several rounds of optimization where, at each round a different but fixed value of penalty parameter is used. While some useful hints and rules for the selection of suitable penalty parameter values have been provided by different authors, the objective of this paper has been to improve this procedure by including the penalty... 

Microchannels are at the fore front of today's cooling technologies. They are widely being considered for cooling of electronic devices and in micro heat exchanger systems due to their ease of manufacture. One issue which arises in the use of microchannels is related to the small length scale of the channel or channel cross-section. In this work, the maximum heat transfer and the optimum geometry for a given pressure loss have been calculated for forced convective heat transfer in microchannels of various cross-section having finite volume for laminar flow conditions. Solutions are presented for 10 different channel cross sections, namely parallel plate channel, circular duct, rectangular... 

The optimal design of a casting feeding system is considered. The problem is formulated as the volume constrained topology optimization and is solved with the finite element analysis, explicit design sensitivity, and numerical optimization. In contrast to the traditional topology optimization where the objective function is defined on the design space, in the presented method, the design space is a subset of the complement of the objective function space. To accelerate optimization procedure, the nonlinear unsteady heat transfer equation is approximated with a Poisson-like equation. The feasibility of the presented method is supported with illustrative examples. © 2007 Springer-Verlag  

Considering the stress cycles in joints and members due to wave induced forces on offshore platforms, fatigue analysis is therefore one of the most important analyses in offshore platforms design. Most of steel jacket type platforms located in areas with relatively high ratios of operational sea-states to maximum design environmental events; fall in acceptable safety margin in inplace and seismic analyses. But in fatigue analyses they will face critical condition. Therefore it seems that utilizing control mechanisms with the aim of increasing fatigue life in such platforms will be more preferable to merely deck displacement control. Investigation of tuned mass dampers adjustable parameters... 

In this letter we consider optimum code structure for positive optical code division multiple-access (optical CDMA) systems. Positive systems are a class of systems that operate with positive real numbers only. We consider the effect of multipleaccess interference in our model and show that code design for both On-Off Keying (OOK) and Binary PPM optical CDMA systems results in the same solutions. Furthermore, we show that a class of codes known as optical orthogonal codes (OOCs) are the best possible positive codes. In obtaining the results we define normalized divergence based on signal-to-multipleaccess interference ratio (SIR) for a multiple-access system in a useful manner and use it as... 

Electric power deregulation has drastically affected the engineering aspects of planning. In addition need flexible electric systems, changing regulatory and economic scenarios, energy savings and environmental impact are providing impetus to the development of Distributed Generation (DG), which is predicted to play an increasing role in the electric power system of the future. This opens the venue for distribution company's (Disco) aiming to minimize their investment risks by developing optimum new planning strategies to meet the load growth and satisfy the system performance at minimum cost different electricity structures. This paper proposes a framework for solving the distribution... 

An efficient algorithm for the design optimization of the compressible fluid flow problem through a flexible structure is presented. The methodology has three essential parts: first the behavior of compressible flow in a supersonic diffuser was studied numerically in quasi-one-dimensional form using a flux splitting method. Second, a fully coupled sequential iterative procedure was used to solve the steady state aeroelastic problem of a flexible wall diffuser. Finally, a robust Genetic Algorithm was implemented and used to calculate the optimum shape of the flexible wall diffuser for a prescribed pressure distribution. © 2006 Elsevier Ltd. All rights reserved  

In this paper the problem of k out of n detection, between sequence of M random bits is addressed. The main application of the result is in a radar system, when we want to detect a target (with unknown time of arrival (TOA)), using binary integration. Another application is in ESM systems, when the system wants to detect the existence of a swept jammer or a gated noise jammer. But as any other mathematical problem, it may have some other applications. In this paper some simple equations for P fa and P d calculation of a detector which detects the existence of sequence of n ones between M random bits, is derived. These equations are then used to find the optimal detector structure in some... 

This paper proposes a multi-objective structure for the optimization of an engine control unit mapping. In this way, first an integrated engine-vehicle model is developed. The objective functions for the optimization problem can be defined from this model. After describing the structure of the optimization problem, two different multi-objective genetic algorithms, namely Distance-based Pareto Genetic Algorithm and Non-Dominated Sorting Genetic Algorithm (together with Entropy-based Multi-Objective Genetic Algorithm), are proposed and implemented. The results demonstrate the superiority of this computerized structure to the manual mapping methods and also more generality of the...