Showing papers on "Metaheuristic published in 1984"

Optimization by Simulated Annealing: Quantitative Studies

Abstract: Simulated annealing is a stochastic optimization procedure which is widely applicable and has been found effective in several problems arising in computeraided circuit design. This paper derives the method in the context of traditional optimization heuristics and presents experimental studies of its computational efficiency when applied to graph partitioning and traveling salesman problems.

A portable machine-independent global optimizer--design and measurements

Abstract: This dissertation addresses the topic of portable and machine-independent program optimization on a standard, well-defined intermediate code. The design of the intermediate code, and features needed to support machine-independent optimization are discussed. A number of new techniques in program optimization are presented. These techniques when applied can substantially reduce both the implementation complexities and running time of optimizers in general, with no sacrifice in the optimizations performed. A register allocation algorithm based on node coloring, suitable for use in the machine-independent context, is introduced. An implementation of these techniques in the machine-independent optimizer UOPT is presented. The optimization performance, efficiency and the relative importance among the different types of optimization transformations are studied according to timing measurements, optimization statistics and by variation in optimization parameters. Finally, the effectiveness of portable machine-independent optimization on a number of target machines that support the intermediate code is discussed, based on optimization performance data in the different machines and comparisons of machine characteristics. The overall evaluation confirms the advantages of using portable machine-independent optimization in a retargetable compiler system.

Dynamic optimization on a non-convex feasible set: Some general results for non-smooth technologies

Abstract: In the theory of optimal intertemporal allocation, the assumption of a convex feasible set has played a dominant role. In recent years, several contributions have focused on the implications for this theory, when the feasible set does not have the convexity property. (See, in particular, Skiba (1978), Majumdar and Mitra (1982, 1983), Dechert and Nishimura (1983), Majumdar and Nermuth (1982), and the much earlier insightful paper by Clark (1971)). These contributions have not only clarified the qualitative differences in the theory in convex and non-convex models, but they have also led to the development of new analytical techniques which have made some issues in the earlier theory in convex models simpler to address (see, for example, Mitra (1983)). However, most of the contributions mentioned above have focused on particular types of non-convex feasible sets; that is, those generated by an S-shaped production function, exhibiting an initial phase of increasing returns, with diminishing returns setting in eventually. Majumdar and Nermuth (1982) work with a more general production function than this, but even there, for the development of the asymptotic stability theory of optimal programs, they have to impose some structure on the type of "non-concavities"

Algorithms for constrained optimization

Abstract: Methods for solving a constrained optimization problem in n variables and m constraints can be divided roughly into four categories that depend on the dimension of the space in which the accompanying algorithm works. Primal methods work in n – m space, penalty methods work in n space, dual and cutting plane methods work in m space, and Lagrangian methods work in n + m space. Each of these approaches are founded on different aspects of NLP theory. Nevertheless, there are strong interconnections between them, both in the final form of implementation and in performance. The rates of convergence of most practical algorithms are determined by the structure of the Hessian of the Lagrangian, much like the structure of the Hessian of the objective function determines the rates of convergence for most unconstrained methods.

Spatial and time decomposition algorithms for dynamic nonlinear network optimization using duality

Abstract: The structural and computational aspects of two decomposition algorithms suitable for dynamic optimization of nonlinear interconnected networks are examined. Both methods arise from a decomposition based on Lagrangian duality theory of the addressed dynamic optimization problem, which is the minimization of energy costs over a given time period, subject to the requirement that the network equations and inequality restrictions are satisfied. The first algorithm uses a spatial decomposition of the state space into subgroups of state variables associated with particular network zones. This leads to a number of lower-dimensional optimization problems which can be solved individually at one level and coordinated at a higher level to account for interactions between these zones. The second algorithm uses time decomposition to solve a sequence of static optimization problems, one for each time increment into which the interval is subdivided, which are then coordinated to take account of dynamic interaction between the time increments. Computational results from an actual network in the United Kingdom are presented for both methods.

Structural optimization using interior penalty function

Abstract: Mathematical programming methods are among the most powerful optimization techniques. They may be classified into direct or indirect methods. In the indirect methods, the constrained design problem is converted into a sequence of unconstrained problems using penalty functions. In this way, the optimal solution of a constrained problem may be obtained using one of the unconstrained search techniques. The interior penalty function appears to be the most reliable uncon strained method while the variable metric method seems to be an extremely powerful algorithm. This paper presents the use of the interior penalty function coupled with the variable metric method for the solution of structural design optimization problems.

A numerical method of shape optimization and the coarse line search strategy

Abstract: The present paper analyses some special features of structural shape optimization and the sizing of elements, such as cross-sectional areas of bars and thicknesses of membranes, and points out that the introduction of a coarse line search strategy is necessary in order to maintain the rigour of mathematical programming and preserve the high efficiency of modern structural size optimization algorithms. Making use of the idea of a coarse line search strategy, we supplement sequential linear programming with the adaptive move limit approach and apply it to minimax problems in shape optimization, i.e. problems of minimizing the maximum stress of a plane stress (plane strain) continuum. Encouraging numerical results are obtained.

On Dialogue Algorithms for Linear and Nonlinear Vector Optimization from the Point of View of Parametric Optimization

Abstract: In treating vector optimization problems, dialogue methods play an ever increasing role (see e.g. R. Dupre et al /2/, A. Wierzbicki /1 1 /, A. Lewandowski et al /8/, S. Zionts et al /13 /).