I have read that any kind of artificial light may emit some energy of IR..my question here that how one can make measurements or computation to find the proportion of IR in the spectrum distribution of the sources...As you know for incandescent bulbs,Planck's law helps in this regard,however,for LEDs or fluorescent lamps, I could not find a way to compute the power spectral distribution because I am working on a formula that needs that power of IR.
You cannot "compute" the IR component of the spectrum of an LED without knowing its construction. Typically the white color of the LED is the result of phosphors that are excited at higher frequencies, and that emit different colors at lower frequencies. It is possible to create a white LED with virtually no IR emission (maybe just a bit from the heating of the LED). However there's no way to calculate it that I can think of - you could measure it by using an IR filter and a photodiode with known sensitivity curve.