Due to the latest advancements in monitoring technologies, interest in the possibility of early-detection of quality issues in components has grown considerably in the manufacturing industry. However, implementation of such techniques has been limited outside of the research environment due to the more demanding scenarios posed by production environments. This paper proposes a method of assessing the health of a machining process and the machine tool itself by applying a range of machine learning (ML) techniques to sensor data. The aim of this work is not to provide complete diagnosis of a condition, but to provide a rapid indication that the machine tool or process has changed beyond acceptable limits; making for a more realistic solution for production environments. Prior research by the authors found good visibility of simulated failure modes in a number of machining operations and machine tool fingerprint routines, through the defined sensor suite. The current research set out to utilise this system, and streamline the test procedure to obtain a large dataset to test ML techniques upon. Various supervised and unsupervised ML techniques were implemented using a range of features extracted from the raw sensor signals, principal component analysis and continuous wavelet transform. The latter were classified using convolutional neural networks (CNN); both custom-made networks, and pre-trained networks through transfer learning. The detection and classification accuracies of the simulated failure modes across all classical ML and CNN techniques tested were promising, with all approaching 100% under certain conditions.

The future of machining lies in the fully autonomous machine tool. New technologies must be developed that predict, sense and action intelligent decisions autonomously. Digital twins are one component on this journey and are already having significant impact in the manufacturing industries. Despite this, the implementation of machining Digital Twins has been slow due to the computational burden of simulating cutting forces online resulting in no commercially available Digital Twin that can automatically control the machining process in real time. Addressing this problem, this research presents a machining Digital Twin capable of real-time adaptive control of intelligent machining operations. The computational bottleneck of calculating cutter workpiece engagements online has been overcome using a novel method which combines a priori calculation with real-time tool centre point position data. For the first time, a novel online machine-induced residual stress control system is presented which integrates real-time model-based simulations with online feedback for closed loop residual stress control. Autonomous Digital Twin technologies presented also include chatter prediction and control and adaptive feed rate control. The proposed machining Digital Twin system has been implemented on a large-scale CNC machine tool designed for high-speed machining of aerostructure parts. Validation case studies have been conducted and are presented for each of the machining Digital Twin applications.

The machinability of Ti-6Al-4V alloy has been a constant challenge in the industry, although the material is widely used in the aerospace and medical industries due to its mechanical properties, particularly its strength-to-weight ratio. The current research presents a hybrid laser- and ultrasonic-assisted machining (LUAM) technique to improve the machinability of Ti-6Al-4V alloy in a turning process. This is compared with ultrasonic-assisted machining (UAM), laser-assisted machining (LAM), and convectional turning (CT). The results reveal that UAM and LAM can reduce the cutting forces and surface roughness (Ra) compared to the CT. However, these are achieved mainly at the lowest range of cutting speeds. The hybrid LUAM process demonstrates process improvement with wider range of cutting speeds and depths of cut, which is achieved due to the combined force reduction and thermal softening effect by the hybrid process.