Abstract: According to the basic requirements of coalbed methane drainage control system, coalbed methane intelligent extraction control method based on neural network is proposed. This method uses neural network fault diagnosis classification principle; based on the gas concentration data collected from drainage pipe by gas concentration sensor, a programmable logic controller (PLC) is used to adjust the valve opening of extraction pipeline under different concentrations and CBM flow to achieve safe production. To achieve the intelligent control of ground gas drainage pumping station on coal bed methane extraction and to solve the deficiencies of gas extraction manual control and semi-automatic control, this method can achieve intelligent control requirements of multi-line extraction.