Abstract: The risk of abnormal high pressure in hydrocarbon generating formations of the Paleogene is becoming increasingly prominent, and the engineering complexity caused by inaccurate monitoring of drilling pressure is gradually increasing, seriously affecting the safety and efficiency of drilling operations. It is necessary to study more accurate monitoring methods to reduce the risk of drilling operations. By analyzing the characteristics of abnormal pressure in the hydrocarbon generating strata of the Paleogene in the study area, Pearson correlation coefficient method was used to optimize seven characteristic parameters from engineering logging, geochemical logging, and elemental logging. A genetic algorithm optimized random forest while drilling quantitative formation pressure monitoring model was established, and the model was used to predict the formation pressure coefficients of nine wells in the study area. The average relative error between the predicted value and the actual value is 7.4%, and the application qualification rate reaches 88.9%. Compared with conventional monitoring methods, it has increased by 22.2%, and the average relative error has decreased by 3.9%, proving the reliability and effectiveness of the model. The conclusion and suggestion provide a solid technical guarantee for the complete layer capture, drilling fluid performance adjustment, and drilling operation safety during drilling.