Error in Converting xml file into ABAQUS inp file

Hi AnyBody team,

I had a problem in converting xml file into ABAQUS inp file using AnyFE Converter Tool for Abaqus. I have been successfully using such Converter Tool for hundred times. The only difference that I have for this ONE is that I have a huge xml file (about 300 MB) with an error of "failed to parse the xml file". I have also tested that if I change xml to a smaller one (20 MB) that I worked with before, the Converter works fine. I am wondering whether there are other options to convert the xml file into ABAQUS inp file without using AnyFE converter Tool. Thank you so much in advance. The reason that xml file is large is because we are exporting the force data for 20 strides, with each stride has around 100 frames. Any suggestions would be greatly appreciated. Thanks.

Junfei

Hi Junfei,

I'd make several sequential xml files (by changing tStart, tEnd) and analyze strides separately. Alternatively you can export data from the XML file yourself and write Abaqus input decks.

Kind regards,
Pavel

Hi Pavel,

Thanks a lot for the suggestion. For the second option, how should I assign the radius for each coupling load? I don't see that from the xml file. Thanks.

Junfei

In the converter we use a default value of 1cm if I remember correctly. It probably works well for the muscle forces, but you may need a larger radius when it comes to joints and reaction forces. It normally depends on your problem and what you want to do exactly.

P.

Hi Pavel,

Got it. Thanks a lot for your help. By the way, have you ever looked into the impact of radius size on the FEA results? Thank you so much.

Junfei

I typically adjust radii to the expected area of function to best of my knowledge. But it would have local effects on the strain/stress. If you apply the forces to a single node - the stress concentration around it will be large. The question here is what is your actual region of interest and how it relates to the radius of influence of a particular element.

Hi Pavel,

Thanks a lot for the info. Considering that we have 57 radiuses need to be determined, it would be lots of manual work. In addition, if we do it manually, it more or less involves bias depends on the user experience. Any suggestions to simplify the procedure in a more consistent manner? Thanks a lot.

Junfei

Not from the top of my head - would need to check literature, which you could also do.
Or you could make a small sensitivity analysis since you will have an automated procedure.

P.

Got it. Thanks a lot for your suggestions, Pavel.

Junfei