Multiple sensor data Integration and analytics for directed energy deposition additive manufacturing (DED-AM) processes and systems PhD

Direct energy deposition additive manufacturing (DED-AM) has been employed by both academia and industry as a driving element of Industry 4.0. The benefits of DED-AM production have been proved in the last decade, in terms of customisation, complicated products, efficient supply chain, shortened leading time and environmental sustainability. Since data becomes one of the most important elements in the additive manufacturing (AM) research, various process monitoring, system monitoring and data collection technologies have been adopted for different AM process systems relying on multiple sensors, in terms of electrical sensor, gas flow meter, pyrometer, process camera, oxygen sensor, etc. However, most of monitoring data collected from these sensors is focused independently and not been integrated together for deep analytics.

Multi-sourced data integration and analytics has been approved to exploit and discover the hidden knowledge in many additive manufacturing research fields. However, there is still a lack of data integration and analytics for DED-AM. The main challenge of such research is that the DED-AM monitoring dataset involves various formats, types, features, and dimensions which tend to be nested as multiple hierarchical structure. Without a comprehensive understanding of every dataset and process domain knowledge, it is impossible to integrate the multiple sensor data and mine the meaningful knowledge efficiently and logically. This PhD project will study multi-sourced data from DED-AM processes and systems, develop and validate a data integration and analytics methodology and application. The overall project objectives are to:

• Examine and study the multiple sensor data collected from DED-AM process and system. 
• Design, develop and evaluate a multiple sensing data integration method based on the data attributes for the DED-AM process and system.
• Understanding and processing the key system and process monitoring data to discover the hidden correlations between variables and to explain the DED-AM process and system. behaviours.
• Develop a knowledge-enhanced data-driven modelling method to predict process/system status and improve the relevant DED-AM process/system.
• Method validation experiment.

The student will be based at the Welding and Additive Manufacturing Centre. The Centre is recognised for the impact of its research into advanced fusion-based processing / manufacturing methods on industry, through extensive MSc and PhD research, and its rolling technology development programme on large-scale additive manufacturing. This project will have close links to Sustainable Additive Manufacturing, Thermal monitoring instrumentation for metal additive manufacturing – PYRAM, and I-Break. The student will be integrated in a diverse and vibrant researcher community. In addition, opportunity for working with the Centre’s industrial partners (e.g. WAAM3D and WAAMMat) would be also provided.

The student is expected to acquire the following (including but not limited to) knowledge and skills from research in this project. 

• Sensor integration and data acquisition, transferring and storage technologies. 
• Advanced data analytics methods, like machine learning and Image processing techniques. 
• Techniques, requirements, and applications of metal additive manufacturing. 
• Reviewing literature, planning, and managing research, writing technical report, paper, presenting in meetings, conferences, and teamwork.