Dr C Chalmers, Dr P Fergus
No more applications being accepted
Funded PhD Project (European/UK Students Only)
About the Project
Project Description:
Data exchange, automation and intelligence are the leading principles driving the industry 4.0 evolution. The purpose of this smart infrastructure is to facilitate the communication between machines, devices, sensors and people via the Internet of Things (IoT). However such evolution requires the use of advanced data analytics tools to extract accurate and meaningful information. This intelligence enables businesses to address uncertainties while making more informed decisions regarding their processes and services.
E-maintenance and service innovations are significant challenges facing both manufacturing and service industries, yet both the use of big data and smart predictive tools are imperative for both efficiency and productivity. Currently machinery and associated control equipment lack the ability to analyse and process large volumes of data for the purpose of predictive maintenance and anomaly detection. Wireless smart sensing technologies with efficient low energy communication infrastructure such as narrowband IoT provide the foundation for predictive technologies. Intelligent algorithms can be used to predict system failure, degradation and enable preventative maintenance by transmitting sensory data to cloud infrastructure for analysis.
The proposed project will be a joint undertaking between Liverpool John Moores University and Central Group PLC. The main objective of the project is to research, develop and implement a novel intelligent prediction system to monitor industrial infrastructure. By using sensing technology, machine learning and a cloud infrastructure the platform will be able to detect anomalies during operation while predicting faults and aiding in diagnostics and maintenance. Although each algorithm will be tailored to the individual device the platform will facilitate automatic training for new devices enabling both scalability and flexibility. In order to generate the necessary training data sensors will be fitted to each device to obtain measurements during both normal and abnormal operation. Example measurements might include temperature, vibration and electrical load. A large component of the machine learning process is the feature extraction stage. However this requires specific domain knowledge for each application which can be both costly and time consuming. During the project alternative methods will be investigated to streamline and optimise this process, such as the use of deep learning and stacked autoencoders.
In addition the project will investigate the optimal way for communicating the acquired intelligence to the service users. In order for the student to address these challenges and gain industrial experience it is proposed that the student will spend at least one day in the engineering facilities at Central and their remaining time at the university. During the course of the project emphasis will be put on the applied aspects of the research which will both validate the approach and its feasibility for use within industry.
Funding Notes
Requirements:
The successful candidate will have a minimum UK Honours Degree at 2.1 or Masters in computer science or related subject.
Good knowledge of Machine learning, statistical modelling and cloud computing.
Programming or software development experience is essential.
Funding notes:
This is a fully funded student scholarship - “This Studentship is Subject to Contract”