If you want to fit some PDF, you should have a set of observations. Then you may fit lognormal distribution (or any other). You need to also check how reliablwe your fitting is. In case of pushover analysis, you get a unique result for a selected structure. You cannot, therefore, construct fragility. If you compute response of the system under different motions, then you can process the results to construct fragility. This is my perception.

For further details, please see my paper:

Seismic structural fragilities: Proposals for improved methodology per spectral matching of accelerogram (2016) by AK Banerjee, D Pramanik, R Roy

Engineering Structures 111, 538-551

Thanks

In my structure, i choose my damage state in term of displacement like design displacement which is 12.8cm. I want to calculate corresponding lognormal fragility parameters.

I know the procedure of calculating fragility curve from time history analysis using IDA, MLE, MSA. but i want to calculate fragility parameters from pushover analysis with out time history analysis

I have pushover curve and data that i got from SAP2000.

Now what is my next step?

maybe, you dont need to

Article TO DETERMINE SKEWNESS, MEAN AND DEVIATION WITH A NEW APPROAC...