Change to the way Power Query reads Excel files?

Thank you.

I'm afraid we've tried these suggestions and while we have been able to get some of our files back to refreshing, there is still one scenario we can't quite work out.

We have a need to analyse multiple weekly excel exports in the same report. These Excel exports are stored in SharePoint (same folder - no changes). They are exported from a source system and settings have been consistent throughout. We ingest data from around 50 excel docs all structured identically and named consistently. 

Where previously we were able to refresh this model without any trouble, this one also stopped working at the same time as all of this excel strangeness began (7 July). We've tried open and closing the workbooks, resaving them, restructuring our Power Query - pretty much everything we can think of. The problem I am finding is when we go through the navigation steps in Power Query, some of the files (when looking at them one-by-one) return the columns you'd expect - [Name], [Item], [Kind], [Hidden], but other files in the list are only returning [Name] and [Data]. 

When we try to process the whole lot of them, we are now consistently getting the message "DataFormat.Error: We were unable to load this Excel file because we couldn't understand its format. File contains corrupted data." However, each of the files, when opened individually, seems to work.

Again, I want to make it clear, that this whole process was working for well over a year and just stopped working earlier this month. Now this particular model appears to refresh in Desktop but will not refresh in the service. Again, for additional context, going through applied steps in the service, everything appears to work, but when I select the dropdown in the column that shows the file names, once the whole query has processed, it is there that I get the corrupted data message.