Search

This paper introduces new techniques for efficient use of electromagnetic transient simulators combined with optimization algorithms to optimize power systems with converter-tied renewable resources. This work is motivated by several challenges that must be overcome for simulation-based optimal design, including high computational burden of simulating large switching systems, repetitive nature of the design cycle, and large number of variables that need to be handled. Two screening methods are proposed in this paper to identify the parameters that do not influence the optimal solution significantly and hence can be ignored. Moreover, hybridization of optimization algorithms and parallel processing techniques are explored to achieve additional computational benefits. Case studies of systems with different complexity and number of variables are used to demonstrate the effectiveness of the proposed techniques. • Several new methods for accelerated simulation-based optimization of power systems are introduced. • Screening methods for removing non-influential parameters from optimization are introduced. • Parallel processing and hybrid, multi-solver algorithms are introduced. • Examples of power systems with renewable resources are shown to quantify the benefits of the proposed methods. [ABSTRACT FROM AUTHOR]

This paper presents an average-value model of a line-commutated converter-based HVDC system using dynamic phasors. The model represents the low-frequency dynamics of the converter and its ac and dc systems, and has lower computational requirements than a conventional electromagnetic-transient (EMT) switching model. The developed dynamic-phasor model is verified against an EMT model of the CIGRE HVDC Benchmark. Simulation results confirm the validity and accuracy of the average value model in predicting the low-frequency dynamics of the ac- and dc-side quantities. Merits and applicability limitations of the average model are highlighted. [ABSTRACT FROM AUTHOR]

This paper extends the concept of dynamic average-value modeling to power-electronic-intensive hybrid-electric vehicular power trains. Hybrid vehicles with plug-in capability are becoming increasingly important in the study of emerging smart power grids, and their simulation using digital programs has gained widespread attention. This paper demonstrates the usefulness of averaging in preserving the dynamic characteristics of the vehicular system while reducing the computational intensity of its simulation on an Electromagnetic Transients Program-type simulator. This paper presents an example of averaging for modeling, simulation, and system-level studies of a power-split gear hybrid drive-train. [ABSTRACT FROM PUBLISHER]

Electric power distribution systems of many commercial and industrial sites often employ variable frequency drives and other loads that internally utilize dc. Such loads are often based on front-end line-commutated rectifiers. The detailed switch-level models of such rectifier systems can be readily implemented using a number of widely available digital programs and transient simulation tools, including the Electromagnetic Transient (EMT)-based programs and Matlab/Simulink. To improve the simulation efficiency for the system-level transient studies with a large number of such subsystems, the so-called dynamic average models have been utilized. This paper presents the average-value modeling methodologies for the conventional three-phase (six-pulse) front-end rectifier loads. We demonstrate the system operation and the dynamic performance of the developed average models in discontinuous and continuous modes, as well as under balanced and unbalanced operation. [ABSTRACT FROM PUBLISHER]