Need a good idea : 3D simulation of Inductive wireless power transfer

[Hasan2022]

Hi,

I am willing to simulate a 3D ferrite bar transmitter and receiver where coupling coefficient k and Bt magnetic flux density on the each side uses the finite element method for solving partial differential equations.

The Magnetic Fields module has equation (jωσ − ω2ε0εr)A + ∇ × H = Je,

which enables calculation of magnetic field distribution B = ∇ × A, where ω is the angular frequency, σ is the electrical conductivity, ε0 is the permittivity of vacuum, εr is the relative permittivity, A is the magnetic vector potential, H is the magnetic field intensity, B is the magnetic flux density, and Je is the external current density.

In my research AC/DC Module, Magnetic Fields need to used for simulation of magnetic flux density and the coupling coefficient in 3D numerical models. Coil Current Calculation, and Frequency Domain for each of the transmitter and receiver coil need to use for studies of a problem in Comsol Multiphysics.

Two parameters, the coupling coefficient k and magnetic flux density on transmitter side Bt for stray magnetic fields were used in this paper for measurements. The first parameter k was used to analyze the different ferrite core geometries. The second parameter Bt was used to analyze the optimal geometries of ferrite bars in terms of magnetic shield.

Two models are required in Comsol Multiphysics, where in the first model simulated self-inductance of the transmitter coil L1, mutual inductance M, and magnetic flux density Bt. With the second model it should simulated self-inductance of the receiver coil L2.

Lets ask you some possibilities:

1. In which way you believe I can make a good implementation ?
2. If I change the circular coil to hexagonal, do you think it would be better ?
3. Which theoretical explanation would better fit , lets say I only have simulation result.
4. Take a look on the reference papers,

 

[Paul Colby]

Two models are required in Comsol Multiphysics, where in the first model simulated self-inductance of the transmitter coil L1, mutual inductance M, and magnetic flux density Bt. With the second model it should simulated self-inductance of the receiver coil L2.

Why two models? The transmitter coil is driven with a current which gives rise to a computed B-field, a back emf and, an induced current. I would add a resistive load to the receive coil and compute the power delivered. The back emf is scaled to the desired drive voltage (by adjusting the drive current) and the power received is the received current times the induced current.

 

[Hasan2022]

@Paul Colby thank you a lot here to take part.
This research is focused to realize the mutual inductance and self inductance of the coil. When EM power is transferred then coil parameters determines the energy density factors. It's not a motor or transformer, not sure why you are raising back EMF.
The orientation of core bars plays a significant development but I am not interested about that issue.

What else is in your mind? Did you read the published article?

 

[Paul Colby]

The back emf determines the inductance of the transmitter, something you wish to calculate. A drive current of 1 amp at the drive frequency will give rise to a voltage drop across the transmission coil,$$i\omega LI$$.

 

[berkeman]

Thread closed temporarily for Moderation...

 

[berkeman]

Okay, I will reopen this thread provisionally.

@Hasan2022 -- This is basically the same question you asked back in 2022: https://www.physicsforums.com/threa...ar-numerical-model-for-magnetic-flux.1015297/

At that time, I asked about the motivations for the ferrite geometries and about where the coils were, but you were more interested in COMSOL simulation help and abandoned that thread.

After looking though the attached paper, I can finally understand a bit more about the geometry and the motivations behind it:

** The two "washers" shown in the figure you copied from the paper are actually the Tx and Rx coils themselves. This picture from the paper (which you did not include) helps to understand the coil geometries:

So the lower radial ferrite bars and upper ferrite plate are meant to help guide and contain the flux that is generated by the Tx coil and help to couple it better to the Rx coil.

** The reason for the goofy radial ferrite bar geometry instead of another ferrite plate is to save on ferrite material, while still trying to maintain a reasonable guidance of the Tx flux. From the paper:

"The aim of this paper is to present a new design of ferrite core made of ferrite bars for inductive wireless power transfer. The authors wanted to answer the question on how many ferrite bars use the least ferrite material for the same coupling coefficient k. This is a very important from economical reason, especially in serial production where the cost and consumption of ferrite material is the required information when designing new components.".

So now that we understand all of that (it would have been nice if you would have explained it when I asked in your original thread), what are you asking now? Just for help on the COMSOL simulation part?

 

[Hasan2022]

Okay, I will reopen this thread provisionally.

@Hasan2022 -- This is basically the same question you asked back in 2022: https://www.physicsforums.com/threa...ar-numerical-model-for-magnetic-flux.1015297/

At that time, I asked about the motivations for the ferrite geometries and about where the coils were, but you were more interested in COMSOL simulation help and abandoned that thread.

After looking though the attached paper, I can finally understand a bit more about the geometry and the motivations behind it:

** The two "washers" shown in the figure you copied from the paper are actually the Tx and Rx coils themselves. This picture from the paper (which you did not include) helps to understand the coil geometries:

View attachment 354847

So the lower radial ferrite bars and upper ferrite plate are meant to help guide and contain the flux that is generated by the Tx coil and help to couple it better to the Rx coil.

** The reason for the goofy radial ferrite bar geometry instead of another ferrite plate is to save on ferrite material, while still trying to maintain a reasonable guidance of the Tx flux. From the paper:

"The aim of this paper is to present a new design of ferrite core made of ferrite bars for inductive wireless power transfer. The authors wanted to answer the question on how many ferrite bars use the least ferrite material for the same coupling coefficient k. This is a very important from economical reason, especially in serial production where the cost and consumption of ferrite material is the required information when designing new components.".

So now that we understand all of that (it would have been nice if you would have explained it when I asked in your original thread), what are you asking now? Just for help on the COMSOL simulation part?

@Berkman thank you for make it more understandable. I understand now why you have closed and re-open the post.

I have been successfully simulate this problem in COMSOL but get stuck if I change the coils type as hexagonal.

Let's say we consider the Tx and Rx coil but not interested to the number "n".

I want you to figure out how we can present this WPT phenomena with different mathematical way or any important factor or parameters that can possible to define with co-relation.

Perhaps I am not interested to make material characteristics for ferrite core. I want you to participate here how I can make something presentable in different way.

I have some mathematics to represent the Hexagonal shape. And there is some point if the coils are in single loop or spiral.