Regex: any 10 digit number preceded by "#" or "# "

[PabloDanielovich]

I have a field that contains text and an account number embedded in the text.  for example

blah bla blah testing ACCT # 3402058931 \bla \ABC\203499\ blah blah

The problem is the string that precedes the account number and is supposed to be the identifier,  sometimes contains typos:
it varies: "ACCT # ", "ACCOUNT#", "AC#" (with or without spaces).  so to parse it out I'm thinking to match any 10 digit number that is preceded by either "# " or "#",  not sure I want to include all of the possible variations. 
So far I have (\d{10}), which will match any 10 digit number (this should work 95% of the time, but it would be good if I could improve it.

[MarqueeCrew]

Regex_replace([field],".*(\d{10}).*",'$1')

That should get it.

[PabloDanielovich]

Thanks Marquee Crew
could you explain that expression?  how does that make sure the account number is preceded by "#" or "# "?  
is the $1 basically saying that we're replacing the entire field value with just the first matched group?   

[MarqueeCrew · Accepted solution]

Regex_replace([field],"(.*#|.*#\s)(\d{10}).*",'$2')

Group 1 is
Any characters followed by a # or
Any chapters followed by a # followed by a space.

Group 2 is
10 digits

Followed by anything

The replacement is group 2

Cheers,
Mark

[PabloDanielovich]

Awesome.  Thank you

[MarqueeCrew]

You're quite welcome.