Loading Events

« All Events

  • This event has passed.

Thesis Defence: Matrix Exponential Integrator Algorithm for Accurate and Efficient Simulation of Power Converter Systems

June 28 at 1:00 pm - 5:00 pm

Nicole Lofroth, supervised by Dr. Liwei Wang, will defend their thesis titled “Matrix Exponential Integrator Algorithm for Accurate and Efficient Simulation of Power Converter Systems” in partial fulfillment of the requirements for the degree of Master of Applied Science in Electrical Engineering.

An abstract for Nicole Lofroth’s thesis is included below.

Defences are open to all members of the campus community as well as the general public. Please email liwei.wang@ubc.ca to receive the Zoom link for this defence.


ABSTRACT

Electromagnetic transient simulation has become a prominent method for intricate power system design and analyses. Current research emphasizes the challenges that arise in developing a computational algorithm for efficient and accurate simulations of power systems due to the conjoined behaviour of continuous states and discrete events. Common methodologies, known as discrete state event-driven (DSED) approaches, aim to overcome the modelling hindrances through employing a fixed or variable simulation step-size and integration order that reduces the number of computational points required to achieve high accuracy with accelerated transient simulation.

The exponential-integrator-based simulation algorithm proposed is a novel variable-step fixed-order (VSFO) method that performs sequential intermediate integration calculations between the discrete events throughout the simulation. The matrix exponential integrator portion of the algorithm offers a competitive approach for an accurate, numerically efficient, and stable technique to accelerate transient simulation. Furthermore, the algorithm is easily adapted to accommodate stiff or nonstiff circuit systems which complement the desired topology and parameters. This thesis conducts a comprehensive analysis of the foundational aspects of the theoretical model for an exponential integrator algorithm and its practical applications for simulation of power converter systems. In addition, the simulation-based case studies are presented for the validation of the algorithm, and provide insightful perspective and approach to converter control modelling and operating principles. A three-phase two-level AC/DC converter, a three-phase two-level AC/DC/AC converter, and a six-pulse rectifier are simulated by the proposed matrix exponential integrator algorithm. These case studies are typical and commonly used converter typologies with active and passive switching events. They demonstrate the proposed algorithm’s ability to accurately capture the hybrid nature of events within the system before each execution of a simulation cycle. The comparison of CPU times of the proposed exponential integrator algorithm to the models built by Simulink/Simscape Electrical toolbox show the simulation efficiency improvements are 2.3749913 s to 15.8125000 s for Case Study One, 8.5514098 s to 15.6562500 s for Case Study Two, and 0.2655774 s to 2.7656250 s for Case Study Three.

Details

Date:
June 28
Time:
1:00 pm - 5:00 pm

Additional Info

Registration/RSVP Required
Yes (see event description)
Event Type
Thesis Defence
Topic
Research and Innovation, Science, Technology and Engineering
Audiences
Alumni, Community, Faculty, Staff, Families, Partners and Industry, Students, Postdoctoral Fellows and Research Associates