The study of pneumatic control valve with hysteresis phenomena has significant importance in process control industries. It is caused by the friction between the stem and packing. It is a nonlinear phenomenon and cannot be expressed by a transfer function. In the present study, a liquid-flow process is considered with an equal percentage pneumatic control valve as a final control element. The control valve has excessive hysteresis. Valve hysteresis strongly influences the dynamic behaviour of a liquid-flow process. Due to hysteresis the model of the open loop flow process is constructed in upward and downward motion of valve stem using neural network with backpropagation technique and general polynomial function. NATIONAL INSTRUMENTS™ based hardware and software tools (LabVIEW™) and MATLAB® R2008a are used for precise and accurate acquisition, flow measurement, control action and simulation. The closed-loop control of a liquid-flow process is simulated with the constructed model. The effectiveness of the simulated model is verified by comparing the simulation results with the real time results in closed loop. It is observed that simulation results are very close to the real time ones.