Hi all
I was struggling with the same issue, and managed to finally solve it, so I'll leave my solution here.
A few introductory words about my use case first.
I have a PowerApps application that stores data in a SharePoint list. I was looking for a way of automatically generating an Excel output based on that data, but with the added difficulty of having to also run 2 Power Queries where one result is used to filter the other result.
In the end, I added the 2 Power Queries to a Power BI and just did a basic table, just to be able to map the required data:


When this was done, well, basically I followed this video from Reza : How to Export Power BI Data to Excel | Query against a dataset (youtube.com)
Unfortunately, for what I needed, large export, not launched from the Dashboard, Reza and all others I could find, only exported to CSV, not Excel.
Still, it was a good starting point.
Now, here comes my twist to the solution, and a massive shoutout to a colleague from work who helped me with this issue.
Instead of CSV table, we used the Parse JSON node as, well, the output of the "Run a query ... " is actually a JSON.

The "catch" here is that you have to generate your own Schema for this to work.
In order to generate the Schema, I used this webpage: Excel to JSON Converter: Convert Excel sheet to JSON table format (codebeautify.org)
But you can obviously use whatever solution you find.
Once that is done, it's just a matter of mapping you excel table rows to the output:

Now, as a last piece of advice, save yourself the pain, and check if values exist (hence all the if's).
It really doesn't like missing data.
As that was another pain point (ask me how I know), here's an example Expression of how to check if there is a value:
if(contains(item(), 'your item here'), item()['your item here'], '')
So, that's pretty much it and solves all my problems, no premium connectors, no conversions or anything, just built in functionality.
Hope this saves someone a few days of their lives (again, ask me how I know 😁)