Electromagnetic simulation analysis via COMSOL
COMSOL is one of the best software tools available for analyzing physical theories, offering comprehensive, rigorous, and highly reliable internal physical formulas, especially with its well-developed logical functions for electromagnetism. Therefore, COMSOL is used in this work to perform electromagnetic and thermodynamic analyses of the model.
For this specific model, the AC/DC module in COMSOL, which is suitable for electromagnetic field analysis, is applied. The parameters are set according to the characteristics of the superconductor's Meissner effect, including the magnetic permeability, electrical conductivity, and dielectric constant, to simulate the model.
- 2.
Model Movement Process
As shown in
Figure 8, the movement process of the model is as follows:
- a)
A cylindrical fixed magnet is placed inside a cylindrical superconductor, which is in the normal state, with the moving magnet positioned 52 mm away from the superconductor.
- b)
From 0 to 5 s, the moving magnet moves at a constant speed of 1 cm per second along the negative x-axis, approaching the superconductor. After 5 s, the moving magnet stops and is now 2 mm away from the superconductor.
- c)
At 5 s, the superconductor begins its phase transition to the superconducting state, and by 6 s, the phase transition is complete, with the superconductor in a fully superconducting state.
- d)
From 5 to 8 s, the moving magnet remains stationary.
- e)
At 8 s, the moving magnet starts moving again at a constant speed of 1 cm per second along the x-axis. By 13 s, the moving magnet returns to its initial position in the model.
The simulation model is built on the basis of the geometric model shown in
Figure 9.
- 1)
Parameter Settings
Owing to significant differences in the properties of various superconducting materials, it is necessary to first select a material for the superconductor in this model before setting the parameters. To simplify the analysis, this model is based on a Type I superconductor and uses the parameters of a well-known material, niobium-titanium alloy (NbTi).
The parameter settings for the superconductor model are based on superconducting theory: the magnetic flux inside the superconductor is zero, and the current experiences no resistance. In other words, when the superconductor is in the superconducting state, its relative magnetic permeability is 0, and its electrical conductivity is infinite.
The theoretical foundation follows Ampère's law, making the "magnetic fields" (mf) physics module the most suitable for simulating superconducting shielding. The "magnetic fields" module is specifically designed on the basis of Ampère's law to build electromagnetic fields. The internal equations within "magnetic fields" adhere to existing electromagnetic theory, including the London equations from superconducting theory, which are based on Maxwell's electromagnetic equations. Therefore, the use of "magnetic fields" for simulating superconducting shielding is both appropriate and correct.
For the superconductor in the superconducting state, the relative magnetic permeability is set to 0, and the electrical conductivity is set to infinity. However, in COMSOL, the relative magnetic permeability cannot be set to 0. According to superconductivity theory and the paper on superconducting shielding[
1,
2], for simulations, the relative magnetic permeability is typically set between 10
−2 and 10
−5, and the electrical conductivity is generally set to 10
5. In this model, the relative magnetic permeability of the superconductor is set to 10
−5, and in the normal state, the superconductor is typically paramagnetic, so its relative magnetic permeability is set to 1. The superconductor in this model is assumed to be a niobium‒titanium (NbTi) alloy. In the normal state, the electrical conductivity of the superconductor has little effect on the electromagnetic properties of paramagnetic materials, so the electrical conductivity is also set to 10
5.
The physics interface for the superconductor uses the "Ampere's law in solids" interface from COMSOL's "magnetic fields" module, with the relative magnetic permeability and electrical conductivity set according to the values mentioned above.
For the permanent magnets in the model, the parameters are set according to commonly used permanent magnetic materials in the software. Both the fixed and moving magnets are assigned the material N54 (sintered NdFeB), with the magnetic flux density modulus slightly reduced to 0.2 T. When the moving magnet is replaced with a ferromagnetic material, its relative magnetic permeability is set to 40,000. The surrounding environment is modeled using air.
- 4.
Simulation process
After the materials, physical fields, and interface parameters are determined, the mesh size and precision are set to ensure that the model calculations converge. Since the model is dynamic, to simplify the process and reduce the computation time, the parametric sweep function in COMSOL is used. The time cycle for the model is set to 13 seconds.
- 5.
Explanation of the Contour Plot
After running the model, the simulation results are shown in
Figure 10, which presents electromagnetic distribution contour plots at six key time points.
The results indicate that during the movement process, from 0 to 5 s, the superconductor is in its normal state, behaving as a paramagnetic material, and thus has no effect on the magnetic fields of the two permanent magnets. Between 5 and 6 s, the model is in a transitional state, which can be considered an intermediate state of superconductivity. Since a parametric sweep was used, the intermediate state is not physically accurate (in reality, the phase transition temperature range of Type I superconductors is very narrow, as small as 10--3). Therefore, there is no need to analyze the intermediate state.
From 6 to 13 s, the superconductor is in the superconducting state, the internal magnetic flux is expelled, and the magnetic flux inside the superconductor becomes zero. A counteracting magnetic field forms on the inner and outer surfaces of the superconductor, opposing the external magnetic field, which is consistent with superconducting theory. At this point, the superconductor not only shields the magnetic field of the fixed magnet from its interior but also generates an opposing magnetic field outside, which exerts a positive electromagnetic repulsive force in the x-direction on the moving magnet.
To facilitate observation, the moving magnet remains stationary during the period from 5--8 seconds. At 8 s, the moving magnet begins to move away from the superconductor, and the opposing magnetic field outside the superconductor starts to weaken. This behavior is fully consistent with the Meissner effect of the superconductor (excluding the pinning effect of Type II superconductors here).
The contour plots of the model during the movement process align perfectly with the theoretical analysis of the model discussed earlier.
- 6.
Data analysis
After completing the simulation of the full model, the first step is to verify whether the model satisfies the principle of electromagnetic superposition. Using COMSOL, simulations were performed for the combination models described in Part 3, and the results were analyzed through superposition.
From the simulation results of the three combination models, two important electromagnetic parameters are extracted for analysis:
1. The x-direction electromagnetic force is exerted on the moving magnet during the operation of the model.
2. The magnetic flux density modulus at point 1 of the model is shown in Figure 9.
By comparing the variation curves of these two parameters across the three combination models, the applicability of the superposition principle to this model is analyzed.
Figure 11 and
Figure 12 show the two combinations of the moving magnet and superconductor and the fixed magnet and superconductor, respectively.
Figure 13 and
Figure 14 display the simulation result contour plots for these two combinations.
After the model is broken down, one more combination involving the fixed magnet and moving magnet remains, which considers only the interaction between the two without involving the superconductor. This will not be discussed at this stage and will be addressed separately later.
After the simulation results are computed, the parameter curves are extracted.
Figure 15 shows the magnetic flux density modulus curve at point 1 in the complete model.
Figure 16 shows the magnetic flux density modulus curve at point 1 in the model without the fixed magnet.
Figure 17 presents the curves from different combination models for comparison and analysis.
In
Figure 17, during the 0–5 second interval, the superconductor is in the normal state, is unaffected by the magnetic field, and does not generate an opposing magnetic field. Furthermore, from 0–5 seconds, it is clear from the three curves that the sum of curves 1 and 2 equals the data of curve 3, which proves that the simulation results are consistent with the principle of electromagnetic field superposition, matching the actual behavior.
Now, let us focus on analyzing the 6–13 second segment of the curves. Between 6–8 s and 8–13 s, curves 1 and 2 almost overlap, showing that the magnetic flux density moduli for both curves are equal. This indicates that the external magnetic field of the superconductor is the same in both states, meaning that the magnetic field of the fixed magnet has been completely shielded by the superconductor. The magnetic flux density on the outer surface of the superconductor is generated entirely by the moving magnet, and the fixed magnet no longer has any influence. Curve 3, which is zero after 6 s, further confirms that the fixed magnet no longer affects the external field of the superconductor.
From a reverse perspective, the magnetic field of the moving magnet affects only the inner surface of the superconductor, whereas the fixed magnet affects only the outer surface of the superconductor. This means that the electromagnetic effects on the superconductor in the model of
Figure 11, combined with the effects in the model of
Figure 12, equal the total electromagnetic influence of both permanent magnets on the superconductor in the complete model shown in
Figure 9.
To summarize the previous analysis, the electromagnetic influence of the permanent magnets on the superconductor in the models of
Figure 11 and
Figure 12, when superimposed, is entirely equivalent to the combined influence of both permanent magnets on the superconductor in the model of
Figure 9.
Additionally, the overall energy in the models of
Figure 11 and
Figure 12 is conserved during the operation process. On the basis of the equivalent analysis, the combined electromagnetic influence between the two permanent magnets and the superconductor in the model of
Figure 9 is also energy conserved.
Since the electromagnetic interaction between the two permanent magnets and the superconductor has been analyzed, let us focus on the interaction between the moving magnet and the fixed magnet during the operation of the model.
The contour plots and the curve in
Figure 17 clearly show that during the 0–5 second interval, the fixed magnet exerts an attractive electromagnetic force on the moving magnet and performs positive work on it. After 6 s, as shown by curve 3, the magnetic flux density modulus of the fixed magnet outside the superconductor drops to zero, meaning that it no longer exerts any electromagnetic influence on the moving magnet. Therefore, over the entire time period, the fixed magnet performs positive work on the moving magnet.
If the two permanent magnets were not in the model, the reciprocating motion of the moving magnet would result in the fixed magnet performing alternating positive and negative work of equal magnitude on the moving magnet, resulting in zero work. However, within the model, the situation is different because of the superconductor’s magnetic shielding function, which effectively isolates the magnetic fields. The magnetic flux does not reach the opposite magnet, and thus, the magnetic force disappears. According to the model, the moving magnet experiences a magnetic force from the fixed magnet when approaching but no force when moving away.
In summary, in the complete model shown in
Figure 9, the remaining effect is the positive work performed by the fixed magnet on the moving magnet during the 0–5 second interval. This indicates a change in the system’s energy, and the model generates positive work, increasing the free energy.
Next, we further analyze the model’s energy change by examining the electromagnetic force experienced by the moving magnet.
Figure 18 shows the electromagnetic force curve for the complete model,
Figure 19 shows the curve for the model without the fixed magnet, and
Figure 20 compares the electromagnetic force curves for the two models.
The electromagnetic force provides a more intuitive explanation of the superposition principle. After 6 s, when the superconductor shields the fixed magnet, the two curves almost overlap, indicating that the electromagnetic force on the moving magnet is nearly the same in both models. This confirms that the electromagnetic force acting on the moving magnet in the complete model at that point is entirely provided by the superconductor’s opposing magnetic field, which aligns with the previous superposition theory analysis.
Figure 19 shows that in the model without the fixed magnet, the electromagnetic force curve during the 0–5 second interval is small and gradually increases in the positive x-direction electromagnetic force. Initially, it was thought that there might be a parameter discrepancy in the superconductor settings, causing the superconductor in its normal state to exert some electromagnetic force on the moving magnet. To verify the correctness of the superconductor parameter settings, the model was modified by setting the superconductor material to air. After running the simulation and outputting the electromagnetic force curve, it was found to match exactly with the curve where the relative magnetic permeability of the superconductor in the normal state was set to 1. This demonstrates that even if an object is set as air in COMSOL, it still exerts a small electromagnetic force on the magnet. Importantly, this proves that the superconductor parameter settings in the normal state were correct. This finding also suggests that solid paramagnetic materials exert a slight electromagnetic force on permanent magnets.
The electromagnetic force analysis further confirms the correctness of the electromagnetic superposition principle in the model, supporting the accuracy of the earlier theoretical analysis.
By using the electromagnetic force curves, the work done by the electromagnetic force can be calculated through integration over the displacement variable. Since COMSOL does not directly offer the functionality to integrate electromagnetic force along the displacement direction, this paper extracts the electromagnetic force data from the model and uses EXCEL or MATLAB to perform displacement integration for each segment of the curve, obtaining the electromagnetic work for each time interval. As shown in Figure 22 to 25, excluding
Figure 22, which shows the electromagnetic work of the superconductor on the moving magnet,
Figure 24 displays the remaining work.
Figure 25 shows the cumulative numerical results for
Figure 24.
When comparing
Figure 23 with
Figure 22, the values are almost identical, indicating that during the 8–13 second interval, the superconductor completely shields the magnetic field of the fixed magnet, and the electromagnetic work exerted on the moving magnet is entirely provided by the opposing magnetic field of the superconductor. The fixed magnet no longer exerts any negative work on the moving magnet.
The simulation model not only confirms the correctness of the electromagnetic superposition principle but also directly calculates the remaining work generated during the model's operation.
Figure 22.
Electromagnetic Work on the Moving Magnet in the Complete Model During 8–13 Seconds.
Figure 22.
Electromagnetic Work on the Moving Magnet in the Complete Model During 8–13 Seconds.
Figure 23.
Electromagnetic Work on the Moving Magnet Without the Fixed Magnet During 8–13 Seconds.
Figure 24.
Electromagnetic Work on the Moving Magnet in the Complete Model During 0–5 Seconds.
Figure 24.
Electromagnetic Work on the Moving Magnet in the Complete Model During 0–5 Seconds.
Figure 25.
Cumulative Electromagnetic Work on the Moving Magnet in the Complete Model During 0–5 Seconds.
- 7.
Simulation Conclusion
The simulation analysis results show that the electromagnetic interactions between the superconductor and the two permanent magnets conform to the principle of electromagnetic superposition, proving that the derivations and analysis in the second part of this paper are entirely correct. Additionally, the simulation directly calculates the positive work done by the fixed magnet on the moving magnet, indicating that the model produces positive work during operation, leading to an increase in energy.
Figure 26 presents an integration of the simulation contour plots and the calculated data.