The relevance of the research is caused by the need to predict the internal physical condition of power transformer windings. Windings displacement leads to a change in windings insulation distances. As a result, the service life of the winding insulation decreases and finally interturn fault occurs. Detection of minor faults in the transformer is essential and difficult task that has not been solved yet. The most sensitive and perspective method as known is nanosecond low-voltage impulse testing, but when response waveforms are obtained and compared, final conclusion (using by visual inspection) is not always clear to make a decision about failures in the transformer windings. The main aim of the research is to develop a method for processing the response waveforms based on the results of diagnosing the internal physical condition of power transformer windings using nanosecond low-voltage impulses. Objects: the response waveforms based on the results of diagnosing the internal physical condition of power transformer ТМ-160/10-У1 by the nanosecond low-voltage impulse testing. Methods: nanosecond low-voltage impulse testing for obtaining the response waveforms; analysis and comparison of the obtained pulse response waveforms by calculation of determination coefficient; coefficient of determination to quantify the differences between the response waveforms. Results. The response waveforms of power transformer ТМ-160/10-У1 were processed using the coefficient of determination, which allows quantifying the comparison of waveforms. The method that allows correctly quantifying prediction of the internal physical condition is developed.