Abstract: According to the “fourpoint” method commonly used in the oilfield actual production and combined with curve moments theory， the invariant curve moments eigenvalue of dynamometer card were extracted as feature parameter vectors. Matterelement model of standard working conditions were established by matterelement analysis method of extension theory whose values of feature parameters were extended to interval forms. Correlation function was used to calculate the correlation degree between the diagnosed sample and the standard working conditions， and then normalization would be carried out. According to the standard working conditions， the correlation degree of diagnostic samples was given by calculations. The maximum correlation degree indicated that the corresponding type of standard working condition should have the greatest possibility. The example illustrated that the proposed method is rational and effective， which is more in line with the actual production requirements.

Key words: dynamometer card diagnosis, matterelement analysis, curve moments, “fourpoint” method, interval feature value