Ilhan Kocar

Multiphase Load-Flow Solution for Large-Scale Distribution Systems Using MANA

The unbalanced nature of distribution systems requires a multiphase load-flow solution capable of handling arbitrary network topologies and providing accurate results. The need for detailed analysis of secondary grid systems found in dense urban areas and the modeling of distribution networks including the subtransmission level requires using highly efficient and large-scale system-capable methods. In this paper, three different load-flow solution algorithms are presented using the modified-augmented-nodal-analysis formulation. The load-flow solution algorithms are compared for the IEEE 8500-node distribution test feeder using a proposed regulator tap control strategy.

A Wideband Line/Cable Model for Real-Time Simulations of Power System Transients

This paper describes the implementation of a full frequency-dependent model for transmission lines and cables in a state-space-based solver for electromagnetic transients. This implementation is for real-time simulation of power system switching transients. It is based on the wideband universal line model with modifications being incorporated to meet the computational speed requirements of real-time applications. The real-time performance of the implementation is demonstrated through application examples.

Improvement of Numerical Stability for the Computation of Transients in Lines and Cables

This paper discusses numerical stability problems of a frequency-dependent transmission-line and cable modeling approach used for electromagnetic transient analysis. Time-domain numerical errors due to the discrete computation of convolution integrals can be estimated in terms of transfer function parameters for a given line or cable model. Based on this estimation, a methodology for the improvement of numerical stability is presented. The numerical advantages of the new method are supported by demonstrations and comparisons with existing models. The method presented in this paper is applicable to power cables and transmission lines.

Weighting Method for Transient Analysis of Underground Cables

Accurate computation of electromagnetic transients for underground cables and overhead transmission lines requires the frequency domain characterization of two matrix functions: propagation and characteristic admittance. The propagation function constitutes a high order wideband system with inherent time delays. In this paper a weighted fitting technique is presented and advocated for the identification of the propagation function as a rational transfer function. The paper also contributes to theoretical clarifications on vector fitting and orthonormal vector fitting. The application of orthonormal vector fitting is demonstrated together with weighted vector fitting. Although the presentation is also applicable to transmission lines, the numerical examples are focused on underground cables where the frequency dependency problems are more complex.

A Unified Distribution System State Estimator Using the Concept of Augmented Matrices

Distribution system state estimation (DSSE) is becoming an essential element of many distribution management systems (DMS), serving as the basis of numerous smart grid applications. In this paper, a new DSSE approach is proposed based on the modified augmented nodal analysis formulation, which is known to have advantageous numerical characteristics. Special consideration is given to the estimation of transformer and regulator tap positions, as they are seldom telemetered on the distribution level. The proposed approach can be easily derived from existing power flow and short-circuit calculation algorithms, thus unifying the major steady-state analysis tools for distribution systems. The proposed DSSE algorithm is tested on the IEEE 8500-Node Test Feeder, and is shown to have good accuracy and efficiency. The proposed DSSE approach is also compared to existing algorithms and shown to have better convergence while yielding a more accurate solution.

General and simplified computation of fault flow and contribution of distributed sources in unbalanced distribution networks

In this paper, the application of a comprehensive augmented nodal matrix formulation to the multiphase fault analysis of unbalanced distribution systems is demonstrated. The solution technique results in a sparse matrix which can be rapidly solved by employing sparse LU factorization algorithms. The augmented matrix fully respects the actual circuit components of distribution systems and eliminates the necessity of pre- and post-processing in the implementation of the algorithm. For example, protective devices that exist in the network are not eliminated but they are represented with benign switches in the matrix. Network devices including transmission lines, cables and transformers can all be modeled in phase domain with their respective physical quantities. The fault conditions including short circuit and open conductor faults are represented with switches by adding the switch equations to the bottom of the augmented matrix which has a dynamic allocation structure. Although the presented solution is of general nature and can encompass networks from transmission level to meshed secondary distribution grids, the examples are rather concentrated on distribution systems where multiphase analysis is more common and desirable due to unbalanced network components.

An Efficient Voltage-Behind-Reactance Formulation-Based Synchronous Machine Model for Electromagnetic Transients

This paper proposes a new synchronous machine model based on the voltage-behind-reactance (VBR) formulation. The proposed approach maintains the accuracy of the VBR model and eliminates its computational inefficiencies. The new model also includes an iterative solution option for achieving higher accuracy. The proposed model is tested and compared to other modeling approaches using practical power system test cases.

Accelerated Computation of Multiphase Short Circuit Summary for Unbalanced Distribution Systems Using the Concept of Selected Inversion

In this paper, an ultra-fast algorithm for the computation of multiphase short-circuit currents at all buses is proposed by addressing the concept of selected inversion. The short circuit summary is achieved through the computation of multiphase Thevenin equivalent at each bus by using the augmented nodal matrix. This typically requires the sequential solution of linear system of equations, i.e., direct inversion which can be demanding in terms of CPU time. In this paper, it is shown that dramatic gains in computational time can be obtained when the selected inversion (SelInv) technique is performed where only a subset of entries of the inverse of augmented nodal matrix is computed. These entries include the phase domain Thevenin impedances seen from network buses which are used in the computation of short-circuit currents with the predefined boundary conditions specific to the fault type. The performance of the method is demonstrated with very large scale and realistic distribution systems.

Development of Data Translators for Interfacing Power-Flow Programs With EMTP-Type Programs: Challenges and Lessons Learned

This paper describes the challenges and lessons learned when developing industrial-grade data translators aimed for the interfacing of power-flow programs with Electromagnetic Transients Program-type programs. It has been found that the greatest challenges to overcome include: the lack, in the databases used in power-flow programs, of vital pieces of information necessary to perform transient studies; inconsistency in the format of data files; the presence of data entry mistakes in very large databases; the validation of the translated data; and the analysis of the large amount of data that transient simulations provide. Several examples are presented to show the implemented solutions. Finally, recommendations based on experience are made to help future developers of interfacing tools.

Performance Analysis of DC Primary Power Protection in Railway Cars Using a Transient Analysis Tool

This paper presents a detailed simulation model developed in an EMT-type (Electromagnetic Transients-type) tool to study transient short- circuit performance of DC primary power protective devices in railway cars. Detailed models are needed since the fault level depends on the location of the moving train, the primary power architecture, and the vehicle operating conditions. Traditional AC RMS time-current curves (TCCs) and AC peak let-through curves are limited practices for protection analysis of current- limiting devices in DC systems. Theoretical background, modeling techniques and typical case studies are presented to demonstrate the functionality and the advantages of using a transient analysis tool for advanced protection studies.