Abstarct: Abstract In this thesis, the simulation and control of the neural network of the compressed natural gas station process with the aim of constantly working and preventing the on-off mode operation of the compressor is investigated. For this purpose, simulating the process of refueling the compressed gas from the gas storage tanks of the station to the storage tanks of the cars, as well as compensating the reduction of the gas stored in the storage tanks is studied. For the purpose of this research work, trained neural network system with the inputs of tank pressure feedbacks and the demand density in the station and compensation mass flow as the output is considered. In the current simulation, the compressed natural gas supply station for cars has been provided with four independent supply platforms or dispencers with the ability to expand to any number of fueling platforms. On the other hand, due to the necessity of properly reflecting the demand conditions in different days and hours, the randomly generated occupancy factor of compressed gas supply platforms is used for each platform. In this way, it is possible to develop a simulation for any number of supply platforms during different days proportional to the demanded density of fueling. In order to achieve the control goals of the research, three neural network systems with pressure difference inputs of each of the low pressure, medium pressure and high-pressure reservoir tanks with the desired pressure along with the platform occupancy factors as the input and required compensating mass flow rate of the mass of gas consumed from storage tanks is trained as the output baxsed on simulation data. In the following, the proportional mass flow resulting from the estimation of the neural network system along with the application of the process control valve, which follows the second order polynomial rule, is used to compensate the mass consumption of the storage tanks baxsed on the instantaneous mass change of each tank. The pressure of the tanks is also updated accordingly by solving the equation of state. baxsed on the proposed control algorithm of trying to develop the CNG station process simulation and control approach in maximum accordance with the nature of the discussed phenomenon, the results showing the achievement of the main goal of the research, including the continuous operation of the compressor and the possibility of estimating the required capacity, have been achieved.