Abstract:

Probable productive formations are recognized by log data standardization, log curve automativc layering and establishing recognizing models; the parmeters such as effective porosity of The formaions, the absofute and effectivePermeabilities of gas, the index and the coefficient of productivity, and the comprehensive index are calculated. The sedata are used as evaluation porters to detendne pwhctive formations. The standard model of productivity gradecan be obtained by cluster analysis, sublayer productivity grad could be connected to the reservoir productivity(testnonresistance out flow), and then the evaluation and the prediction of reservoir productivity could be realized.

Key words: gas reservoir, log evaluation, productivity prediction, fuzzy cluster