I updated the tags on this for better visibility .
As a rule of thumb I would always ignore the intercept.
Reason being is that you will be including a result that you have not measured.
You are using a blank for your lower end concentration, meaning that the blank result (if there is one) will be included in the calibration.
NOTE: This is a statistically debated area. Personally, though the calibration curve you produce may not be straight line linear at the lower end concentrations (due to noise interfearance).
Hope this helps.