Thermal control of a mould is the key in the development of high efficiency injection moulds. For an effective thermal management system, this research provides a strategy to identify the thermal dynamic model for the design of a controller. Using neural networks and finite element analysis, system identification is carried out to deal with various cycle-times for moulding process and uncertain dynamics of the mould. Based on the system identification, a self-adaptive PID controller with radial basis function (RBF) is designed to tune controller parameters. The controller’s performance is studied in terms of tracking accuracy under different moulding processes.