Evaluation study for the integration of electromagnetic transients simulator and transient stability simulator

Abstract

This paper is devoted to the development of a hybrid simulator which integrates the electromagnetic transients (EMT) simulator and transient stability (TS) simulator. It discusses how two very different simulators can be coordinated to work together and produce accurate solution. A new selection approach is presented to identify the interaction location for interfacing the EMT and TS simulators. Pseudo-transient caused by frequency mismatch during the interaction is discussed with remedy presented. Based on the interaction protocol proposed in the paper, case studies were performed on two test systems, 9-bus systems and 39-bus systems, to assess the performance of the integration on both EMT and TS aspects.

Introduction

The aim of this work presented in this paper is to develop a new broadband digital power system simulator (hybrid simulator), which will provide simulation of large systems, including detailed modeling of parts of the system, while employing minimum processing resources. Currently, electromagnetic transients (EMT) simulators provide high time-resolution simulation of power systems, including device-level models of power electronic devices. However they are computationally very demanding and are not ideal for simulation of very large systems. Transient stability (TS) simulators, on the other hand, have a very fast simulation speed and can handle very large networks, but with compromised time resolution. Consequently power electronic devices and control systems can only be represented as modified steady-state models, and are not modeled at the device level.

A hybrid simulation takes advantage of the computational inexpensive dynamic representation of the main network in a stability program with the accurate dynamic modeling of the remaining components using EMT modeling. The underlying idea of the hybrid simulation is to partition a network into two parts, one for the TS program and the other for the EMT program. Of course, in order to exploit the speed advantage of the TS program, the part simulated by the TS program should include most components of the system, while the part simulated by the EMT program would contain the components that require detailed simulation, as well as parts of the network close to where they are connected. Thus, the slow dynamics of machines are adequately modeled by the stability program while the fast dynamic responses of selected devices are accurately represented by EMT simulation models.

In the early eighties, Heffernan et al. [1] first proposed to interface two distinct simulators for solving HVAC–HVDC systems. They modeled an HVDC link in detail within a stability based ac system framework, thus exploiting the advantages of both EMT program and TS program. They achieved this by running the TS program and EMT program concurrently with periodic coordination of the results. Reeve and Adapa [2] proposed that the location of interface should be extended into the ac network further for taking into consideration of the effect of harmonics generated by power electronics on the ac network. Anderson et al. [3] presented another approach to take the harmonics into account. In the EMT program, the network part simulated by the TS program is represented by frequency-dependent equivalent, instead of a simple fundamental frequency equivalent circuit used by Heffernan and Reeve.

The paper presented by Sultan et al. [4] basically adopts the approaches described above, i.e. extending the interface location into the ac network to some extent, and at the same time representing the network simulated by the TS program with frequency-dependent equivalent. Also, Kasztenny and Kezunovic [5] have discussed a general method for linking different modeling techniques such as waveform-type, phasor-type, and algebraic-type simulation techniques into one complete model.

This paper is devoted to the theoretical analysis and performance evaluation of the hybrid simulation. Issues on the establishment of proper interaction between the EMT and TS simulators are discussed. A new approach to selecting the interface location, techniques to overcome the waveform distortion caused by frequency mismatch during the transient and interaction protocol are presented. Full assessments of the hybrid simulation, from the points of view of TS and EMT, on two different sized power networks are also given.