It's not ordinarily done the way you are proposing. I'll assume you mean ppm "by weight" b/c it could also be by mole.
First: you need to know the NO/NO2 ratios of your NOx. Then also realize that EPA assumes all the NO goes to NO2 in the atmosphere. E.g. if you produce NOx with 90% NO, you will be "credited" with .1 * x + .9 * (46/30) x, or 1.48 g NOx emissions per 1.0 g NOx measured.
So - you measure 100 g NOx, convert to 148 g NOx equivalent, and then you have to know how many hp-hr were used in generating that much NOx.
e.g. - let's say your engine runs 23,000 g/min exhaust flow at 200 hp, and you measure 200 ppm (weight) NOx; your g/hp-hr is: 23,000 * 60 * (200/1,000,000) * 1.48 / 200 = 2.04 g/hp-hr