Neurofuzzy Drilling Force-based Control in an Ethernet-based Application

This paper reports on the design and implementation of a neurofuzzy system for modelling and controlling drilling processes in an Ethernet-based application. The neurofuzzy system in question is an Adaptive Network based Fuzzy Inference System (ANFIS), where fuzzy rules are obtained from input/output data. The design of the control system is based on the internal model control paradigm. The main advantages of the suggested approach are that its use of a neurofuzzy system to deal with nonlinear drilling process behaviour and process uncertainty eliminates the need for an exact mathematical model to design and tune the control system, and that it offers a simple and computationally efficient procedure for real-time applications. The results are positive in both simulation and in the real-time application of networked control. The case study indicates that the proposed method outperforms a PID control strategy and an optimal fuzzy controller. This improved behaviour is verified by several performance indices.