Due to the fine resolution and continuity of the well log

it plays an irreplaceable role in the high frequency sequence stratigraphic division. By the adoption of the maximum entropy spectrum decomposition technology to analyze the predication error of the well log

the prediction error trend curve is obtained by further calculation. The wavelet transform technology transforms the one-dimensional depth domain logging signals into the two-dimensional depth and the scale domain logging signals. And the organic combination of the two techniques can well identify the multi-order sedimentary cycles of the stratum

so the interferences of the human factors are greatly reduced. The high frequency stratigraphic sequence division of the actual log data in Bayanhushu Sag is conducted by using this method

and the application effect is good.