Summary form only given. Electromagnetic transient simulation programs (EMTP-type programs) are powerful tools for the study of a wide range of power system transient problems. This paper introduces a novel tool in which an EMTP-type program becomes the objective function evaluator for a non-linear optimization algorithm. In this approach, the non-linear optimization program is given control to perform several consecutive runs with a view to minimizing (or maximizing) the desired objective function, which is computed from the results of each simulation run. Since the optimization algorithm strategically selects the parameters for the EMTP run, the overall design process is orders of magnitude faster than that possible from sequential or random (Monte-Carlo type) multiple-runs of the EMTP-type program. The paper discusses the mechanics of the interface as well as design of objective functions. The power of the proposed method is demonstrated through two examples for the design of power electronic converters.
[1]
Hermann W. Dommel,et al.
Digital Computer Solution of Electromagnetic Transients in Single-and Multiphase Networks
,
1969
.
[2]
A. Ravindran,et al.
Engineering Optimization: Methods and Applications
,
2006
.
[3]
John A. Nelder,et al.
A Simplex Method for Function Minimization
,
1965,
Comput. J..
[4]
D. V. Chudnovsky,et al.
Solving the optimal PWM problem for single-phase inverters
,
2002
.
[5]
Robert K. Brayton,et al.
The Sparse Tableau Approach to Network Analysis and Design
,
1971
.