Vibration Analysis, Eigenvalues

[Nazi1400]

Hello,

I have a 3D COMSOL model which I am using for the purpose of vibration analysis.
Up to now I've got analytical eigenvalues using COMSOL with MATLAB. I have to correlate the results with the measured eigenvalues results, using MAC criterion.
The problem is that my analytical results are in real numbers but the measured results are complex numbers and I don't know why :(
I used “mphinterp” command in MATLAB to compute the eigenvalues.
Could anyone give me some hits how to do this correlation?

Cheers,
Nazi

 

[Achy47]

Hi Nazi,

Thank you for sharing your question with us. It sounds like you are working on an interesting project with your 3D COMSOL model and vibration analysis. It is not uncommon to encounter discrepancies between analytical and experimental results, and the MAC criterion is a useful tool for comparing them.

First, it is important to understand why your analytical results are in real numbers while your measured results are in complex numbers. Without knowing the specifics of your model and measurements, it is difficult to say for sure, but here are a few possibilities:

1. Your analytical results may be simplified or idealized, while your measured results are more accurate and include more complex variables. This could be due to simplifications made in your model or assumptions made in your calculations.

2. The measured results may be affected by external factors such as noise or other sources of error, which could introduce complex numbers into the data.

3. There may be a discrepancy in the units used for your analytical and measured results. Make sure to double check that they are consistent.

Once you have a better understanding of why your results are in different forms, you can begin to compare them using the MAC criterion. This criterion is based on the concept of modal assurance, which measures the similarity between two modes of vibration. Essentially, you will need to compare the mode shapes and frequencies of your analytical and measured results and determine how well they match up.

To do this, you can use the mphinterp command in MATLAB to compute the mode shapes and frequencies of your analytical results at the same locations and frequencies as your measured results. This will allow for a direct comparison between the two sets of data. From there, you can calculate the MAC values and determine the level of correlation between the two sets of results.

I hope this helps you get started on your correlation analysis. If you have any further questions or need more guidance, please don't hesitate to reach out. Best of luck with your project!