Abstract: To better control of the pressure drop of filter rods, an intelligent control system was designed for a ZL29 filter rod maker. The system consists of a data acquisition module, a data preprocessing module, a filter rod pressure drop prediction model, a prediction model training module and control models of the filter rod maker. The filter rod pressure drop prediction model is established on the basis of random forest regression algorithm to timely acquire the predicted pressure drop values according to the trends of the pressure drop of finished filter rods and the entire process operational data of ZL29 filter rod maker. The filter rod maker control model is compatible with the Beckhoff PLC controller without impacting the normal operation of the original machine. The designed control system was verified and the results showed that compared with the traditional manual control mode, the intelligent control mode reduced the standard deviation and the coefficient of variation of the filter rod pressure drop by 12.69% and 12.41% respectively, and raised the process capability index by 24.72%. This technology provides support for improving the stability of filter rod pressure drop.