This project involved the design, simulation, and analysis of a single-phase isolation transformer using COMSOL Multiphysics. The primary focus was to understand the behaviour of the transformer with both a solid core and a laminated core, and to evaluate the equivalent circuit parameters under different conditions. The project also explored methods to reduce computational cost through domain symmetry.
Download Project Report (PDF)
Solid Core Model: The initial simulation model was based on a solid core made of electrical steel. The core exhibited skin effects, leading to non-uniform current density and magnetic flux concentration, particularly near the core edges. This model highlighted inefficiencies due to elevated eddy current losses and core saturation at high magnetic flux densities.
Laminated Core Model: To improve transformer performance, the core was laminated, reducing eddy current losses and increasing overall efficiency. The laminated core model was created by adjusting material properties in COMSOL, simulating the effects without altering the model’s geometry. This approach allowed for more uniform magnetic flux distribution and reduced current density concentrations.
The simulation results demonstrated significant improvements with the laminated core model. By partitioning the core into laminated layers, the resistance to eddy currents increased, leading to reduced losses and better overall transformer performance. The laminated core also showed a lower magnetic flux density, staying within optimal ranges, while the current density was more evenly distributed, reducing the impact of the skin effect.
Solid Core: The solid core model showed that the skin effect caused magnetic flux density to reach values above 2 Tesla, leading to inefficiencies and higher core losses.
Laminated Core: The laminated core model significantly reduced magnetic flux density, improving performance by keeping it below 1.6 Tesla and limiting eddy current losses.
Open Circuit Test: This test was used to calculate the core loss resistance (Rm) and magnetising reactance (Xm) by setting a low load resistance on the secondary winding. The results indicated that the laminated core had better efficiency, with reduced losses compared to the solid core.
Short Circuit Test: By setting a high load resistance on the secondary winding, this test allowed for the calculation of winding resistance (R) and leakage reactance (X). The laminated core performed better under short-circuit conditions, showing reduced resistive and reactive losses.
One of the key challenges in this project was accurately simulating high-frequency effects and temperature impacts on the transformer’s performance. Future improvements could include studying the effect of non-linear magnetic materials, exploring the transformer’s behaviour over a wider frequency range, and investigating the thermal effects on transformer efficiency. These enhancements would provide a more comprehensive understanding of the transformer’s real-world performance.
