I consider that you can not solve this issue, due to different corrosion[1] (history) path of (some of) your specimens. We can solve this issue only for the NEW specimens e.g. all specimens inserted (begin with) in your measurement from the SAME (materials and surfaces) conditions as starting point.

1. Corrosion of Embedded Metals  http://www.cement.org/for-concrete-books-learning/concrete-technology/durability/corrosion-of-embedded-materials

If you want to find the polarisation resistance this way you have to use CV, not LSV: from OCP to OCP-15 mV to OCP+15 mV to OCP (repeat at least 5 times).

A better option is EIS (electrochemical impedance).

.

 Thanks Christian Punckt and Cornel Radu for your replies.

I am monitoring OCP just before doing Linear Polarisation Resistance. Even if my OCP is stable, for some specimens I couldn't see the curve crossing zero current line in the E vs I curve. I will check it once again. Thanks a lot for your time.

as Cornel Radu : A better option is EIS (electrochemical impedance).

Thank you Ioannis Samaras sir for your suggestion

I want to add some answer to this question.

First stabilized the OCP at least for 1 hr and then give the potential input +- 20 to 15 mV with respect to OCP.

Or to calculate PR (polarization resistance), FRA (EIS) is the best techniques.

By suitable equivalent circuit fitting, you will get addition information like capacitance behavior, solution resistance, polarization Resistance at OCP.