1254 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 57, NO. 4, APRIL 2010 A Megawatt-Scale Power Hardware-in-the-Loop Simulation Setup for Motor Drives Michael Steurer, Senior Member, IEEE, Chris S. Edrington, Senior Member, IEEE, Michael Sloderbeck, Member, IEEE, Wei Ren, Senior Member, IEEE, and James Langston, Member, IEEE Abstract—We report on the application of a 5-MW variable voltage source (VVS) amplifier converter for utilization in power hardware-in-the-loop (PHIL) experiments with megawatt-scale motor drives. In particular, a commercial 2.5-MW variable speed motor drive (VSD) with active front end was connected to a virtual power system using the VVS for integrating the drive with a simulated power system. An illustrative example is given, whereby a 4-MW gas turbine generator system, including various loads, is simulated and interfaced with the VSD hardware in the lab through the VVS using current feedback to the simula- tion. Mechanical loading is applied to the motor via an identical 2.5-MW dynamometer connected to the same shaft. This paper first describes the PHIL facility, illustrates the challenges of pow- ering a motor drive from a controlled voltage source converter at the multimegawatt scale, and provides experimental results from dynamic simulations. While certain challenges remain with the accuracy of the interface, it is concluded that PHIL simulations at the megawatt power level are possible and may prove useful for validating models of drive systems in the future. Index Terms—Power hardware-in-the-loop (PHIL), real-time digital signal processing, variable speed drive. I. I NTRODUCTION P OWER hardware-in-the-loop (PHIL) simulation is an emerging method for advanced experimentation, whereby a piece of power hardware, for example, a power electronic drive and motor, is operated from a virtual grid, simulated in real time, and provided to the device under test by means of power amplification. PHIL is an extension of the con- troller hardware-in-the-loop (CHIL) simulation method which is widely applied for investigating the real-time behavior of controller and protection equipment. The different structures of CHIL and PHIL simulations are shown in Fig. 1. Since, in a CHIL simulation, the signals exchanged between the hardware and the simulator are low-power control signals that are usually within the range of ±10 V, 0–100 mA, commercial digital/analog (D/A) or analog/digital (A/D) converters suffice for the interface. Successful application of CHIL simulation Manuscript received March 2, 2009; revised November 2, 2009. First pub- lished November 24, 2009; current version published March 10, 2010. M. Steurer, M. Sloderbeck, and J. Langston are with the Center for Advanced Power Systems, Florida State University, Tallahassee, FL 32310 USA (e-mail: steurer@caps.fsu.edu; sloderbeck@caps.fsu.edu; langston@caps.fsu.edu). C. S. Edrington is with the FAMU-FSU College of Engineering, Tallahassee, FL 32310 USA, and also with the Center for Advanced Power Systems, Florida State University, Tallahassee, FL 32310 USA (e-mail: edrington@caps.fsu.edu). W. Ren is with the GE Global Research Center, Niskayuna, NY 12309 USA (e-mail: w.ren@ge.com). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TIE.2009.2036639 Fig. 1. Structural distinction between CHIL and PHIL simulations. for electric drive design with a variety of different real-time simulation platforms has been reported in [1]–[7]. In contrast, a PHIL simulation of satisfactory accuracy requires a high- precision interface amplifier (AMP in Fig. 1). However, impre- cision present in the amplified signal caused by time delays, nonlinearities, sensor noise, and noise injection caused, for example, by power electronic switching within the amplifier, could result in significant simulation errors and even instabili- ties [8]. This problem is particularly challenging in high-power applications, where high-precision amplifiers are not readily available. Moreover, any specially designed system is typically quite costly. Despite these challenges, PHIL simulation offers significant advantages in different stages of a motor drive development. It allows a portion of the drive system to be tested long before the other parts of the system are ready and therefore expedites the design cycle and reduces the design complexities. It significantly increases the opportunity to identify hidden defects which may not necessarily be design flaws but may stem from the manufacturing process, for example. In other situations, PHIL can be used to thoroughly test, and accept, an existing hardware for an application which was developed after the hardware was designed and built and for which it was not originally intended to be used. Extreme scenarios can be thoroughly scrutinized with minimum risks and costs. Several published papers on using PHIL simulation for mo- tor drive studies at small-to-moderate power levels (less than 10 kW) can be found [10]–[15]. However, no literature could be found on PHIL applications for high-power drives at the megawatt level. Recognizing the need for such capabilities, the author’s research institution established a PHIL facility with a power rating of 5 MW. This facility has already been used successfully to conduct PHIL experiments at the megawatt 0278-0046/$26.00 © 2010 IEEE