Wyniki wyszukiwania - BazTech

1

Object library of algorithms for dynamic optimization problems; Benchmarking SQP and nonlinear interior point methods

Błaszczyk J., Karbowski A., Malinowski K.

International Journal of Applied Mathematics and Computer Science

|

2007

|

Vol. 17, no 4

515-537

EN

The main purpose of this paper is to describe the design, implementation and possibilities of our object-oriented library of algorithms for dynamic optimization problems. We briefly present library classes for the formulation and manipulation of dynamic optimization problems, and give a general survey of solver classes for unconstrained and constrained optimization. We also demonstrate methods of derivative evaluation that we used, in particular automatic differentiation. Further, we briefly formulate and characterize the class of problems solved by our optimization classes. The solution of dynamic optimization problems with general constraints is performed by transformation into structured large-scale nonlinear programming problems and applying methods for nonlinear optimization. Two main algorithms of solvers for constrained dynamic optimization are presented in detail: the sequential quadratic programming (SQP) exploring the multistage structure of the dynamic optimization problem during the solution of a sequence of quadratic subproblems, and the nonlinear interior-point method implemented in a general-purpose large-scale optimizer IPOPT. At the end, we include a typical numerical example of the application of the constrained solvers to a large-scale discrete-time optimal control problem and we use the performance profiles methodology to compare the efficiency and robustness of different solvers or different options of the same solver. In conclusions, we summarize our experience gathered during the library development.

2

Optimisation of welding parameters for weld penetration in FCAWN

Mostafa N. B., Khajavi M. N.

Journal of Achievements in Materials and Manufacturing Engineering

|

2006

|

Vol. 16, nr 1-2

132--138

EN

Purpose: This paper describes prediction of weld penetration as influenced by FCAW process parameters of welding current, arc voltage, nozzle-to-plate distance, electrode-to-work angle and welding speed. Optimization of these parameters to maximize weld penetration is also investigated. Design/methodology/approach: It deals with the statistical technique of central composite rotatable design to develop a mathematical model for predicting weld penetration as a function of welding process parameters. The constrained optimization method is then applied to this model to optimize process parameters for maximizing weld penetration. Findings: The result obtained from the developed model indicates that the model predicts the weld penetration adequately. The optimization result also shows that weld penetration attains its maximum value when welding current, arc voltage, nozzle-to-plate distance and electrode-to-work angle are maximum and welding speed is minimum. Research limitations/implications: The statistical technique of developing a model for prediction of weld penetration is valid only within the specified limits of welding process parameters and hence maximization of penetration is also valid within these limits. This technique can be modified to include other parameters such as plate thickness affecting penetration. Practical implications: The statistical method for modeling and the optimization method used have found increasing applications in a variety of fields for investigation because through these methods research becomes economical, fast and versatile. Originality/value: The methods described in this paper for weld penetration prediction and optimization can eliminate the need for performing experiments on the basis of the conventional trial and error method which is time consuming and economically not justifiable.

3

Porównanie efektywności algorytmu deterministycznego i algorytmu ewolucyjnego w zastosowaniu do estymacji parametrów transformatora

Kowol M., Młot A.

Zeszyty Naukowe. Elektryka / Politechnika Opolska

|

2004

|

Vol. 297, z. 54

43-53

PL

W pracy przedstawiono wyniki estymacji parametrów modelu matematycznego transformatora jednofazowego otrzymane przy użyciu dwu różnych metod minimalizacji funkcji celu. W obliczeniach zastosowano algorytm deterministyczny bazujący na metodzie sekwencyjnego programowania kwadratowego - SQP (ang. Sequential Quadratic Programming) oraz algorytm ewolucyjny - AE (ang. Evolutionary Algorithm) z kodowaniem rzeczywistym. Parametrów poszukiwano w stanach nieustalonych na podstawie modelu obwodowego analizowanego układu. Dokonano porównania wyników obliczeń oraz kosztów obliczeniowych. Przedstawiono problemy związane z zastosowaniem obydwu metod w estymacji parametrów analizowanego układu.

EN

The paper presents results of estimation of non-linear parameters of one-phase, low-power transformer with ferromagnetic core. The estimation is performed by using three different methods to minimize an objective function. In order to calculate parameter estimates the evolutionary algorithm, deterministic algorithm and hybrid algorithm are used. The deterministic algorithm is based on Sequental Quadratic Programming (SQP). Estimation of parameters is performed in the current transient state on the basis of equvalent reluctance network transformer model.