Exploring Spatial Data Science
Systems and Control
Our research interests in this theme include dynamical systems, control theory and analysis. In particular, the group’s research in dynamical systems include stability, behaviour and performance analysis of linear to nonlinear, infinite-dimensional, switched systems, complex networked and stochastic systems. In control theory, our focus is on optimal, robust, nonlinear, adaptive, stochastic, learning and intelligent control, alongside the development of data-driven and machine learning methodologies in control systems design. This is driven by diverse real-world applications ranging from conservation and ecological management to renewable energy systems. We are also active in complex and functional analysis, fractional calculus and integral transforms, and their connections with signal and image processing, for applications such as neuroscience and cardiology.
View recent outputs from work within this research theme
Active members and significant achievements
- Dr Tim Hughes has resolved numerous longstanding open questions on the analysis and design of passive systems (i.e., systems that do not consume energy), with application to low energy mechanical design and electric circuit synthesis. His research is now included in the definitive textbooks of the field.
- Dr Markus Mueller and Professor Stuart Townley pioneered the use of adaptive and set-point control for management of natural populations.
- Dr Saptarshi Das has developed fractional calculus and artificial intelligence based hybrid methods to discover hundreds of new chaotic attractors and to study synchronization of switched chaotic systems.
- Dr Houry Melkonian has used harmonic analysis to study the generalised Fourier-type series that arise in the study of the (p,q)-Laplacian with Dirichlet boundary conditions, and has developed a new stream of research which studies the behaviour of integrals of generalised sinc functions.
- Dr Mark Callaway has research interests in random dynamical systems and bifurcation.
Epidemiology and Human Health
Our research interests in this theme include complex systems, data science and statistical modelling approaches to resolve big research questions in epidemiology and human health. Within epidemiology and infectious disease modelling, a particular focus is on dengue, malaria and antimicrobial resistance. Within human health data analytics and modelling, we focus on digital technologies, artificial intelligence and machine learning methods for health and well-being, alongside computational problems in neuroscience and cardiology.
View recent outputs from work within this research theme
Active members and significant achievements
- Dr Mario Recker and collaborators made fundamental contributions to a World Health Organisation (WHO) model comparison study on the safety and long-term impact of the first dengue vaccine. He is also active in statistical modelling using individual based models of vector-borne disease systems and immune signature data classification for malaria.
- Dr Markus Mueller and Professor Stuart Townley are theme-leads for modelling of systems design in Smartline and Smartline 2.0, two interdisciplinary projects that explore the relationship between digital technology, health and well-being through research and collaborations with local businesses and social enterprises. Professor Townley and collaborators also developed AI-based tools for predicting the development of acute kidney injury on hospital admission.
- Dr Saptarshi Das and collaborators developed electroencephalogram (EEG), functional magnetic resonance imaging (fMRI), and diffusion tensor imaging (DTI) data processing and assimilation algorithms for studying phase synchronization and functional brain connectivity in autism and anxiety. Dr Das also works on computational challenges in cardiac and neural signal processing using nonlinear dynamics, stochastic processes and machine learning.
Mathematical Biology and Ecology
Research interests within this theme include modelling of ecological dynamics and biological processes in nature.
View recent outputs from work within this research theme
Active members and significant achievements
- Dr Mario Recker developed a modelling framework to predict individual-level risk of mortality following a bloodstream infection with the hospital superbug staphylococcus aureus (MRSA). He also developed computational algorithms to study a notable male-female health-survival paradox in drosophila. In addition, he has made fundamental contributions in the study of antimicrobial resistance for two antibiotics, gentamicin and kanamycin, for Escherichia coli.
- Dr Markus Mueller and Professor Stuart Townley have pioneered the development of robust control tools for the management of natural populations. They have also developed a mathematical theory for analysing transient dynamics in human, animal and plant population models. This theory has recently been applied to predicting invasion by species into non-native ranges, based on their native demographies. Dr Townley used data science tools to study global patterns of sex and age-variation in seabirds, and has contributed to the development of stochastic projection matrix models for studying demographic dynamics of plant population growth.
- Dr Saptarshi Das has developed image processing algorithms for the statistical characterisation of microbial community pattern formation from scanning electron microscopy images for wild pseudomonas biofilms. Together with collaborators, he has also developed novel statistical feature and wavelet transform based pattern recognition algorithms for plant electrical signal processing to identify chemical pollutants in the environment.
Renewable Energy, Sustainability and Fluid Mechanics
Our research in this theme involves collaboration with the world-class Renewable Energy group and the Exeter Energy network. In particular, Drs Hamid Alemi Ardakani, Markus Mueller, Saptarshi Das and Professor Stuart Townley are active in modelling and design of wave energy converters and smart grids; energy data analytics and machine learning; and data driven modelling, control and optimization of renewable and sustainable energy systems.
View recent outputs from work within this research theme
Active members and significant achievements
- Dr Markus Mueller works on modelling and control design for marine renewable technologies. He led the control design work package for OPERA, an EU Horizon 2020 Research and Innovation programme grant aimed at technology readiness for oscillating water column wave energy converters. He also developed spatio-temporal life cycle analysis methods to determine localised emissions and wind energy over time. In addition, Dr Mueller works on control strategies of ocean wave converters, turbines and wave power plants.
- Dr Alemi Ardakani has research interests in mathematical analysis, numerical analysis, structure-preserving algorithm design, and theoretical fluid and continuum mechanics. He has recently developed new variational principles for the generalized Green-Naghdi and Whitham equations, and Euler-Poincare variational principles for three-dimensional interactions between ocean surface waves and a floating rigid-body, dynamically coupled to its interior fluid sloshing.
- Dr Saptarshi Das works on energy and power systems dynamics, control, optimization and data analytics with a focus on clean energy and sustainable technologies. In two recent highly cited papers, he developed fractional calculus based intelligent control approach using chaotic swarm optimization and surrogate based optimization for hybrid power systems and microgrids with renewable generation and energy storage. Dr Das also used robust optimization and kriging based surrogate models to handle complexity of load and renewable generation patterns for distributed networked control of energy storage elements in smart grids, and he has also developed chaotic multi-objective optimization based load frequency fractional order control methods for interconnected power systems. Dr Das is currently using machine learning and artificial intelligence approaches for the development of battery management algorithms in smart grids, in particular islanded micro-grids with solar photovoltaics. This research is making economic impact through the Cornwall New Energy project.
Mathematics for Innovation and Business
Within this theme, our focus is to develop mathematical, statistical and computational methods and tools needed for sustainable growth of small and medium size business, working with the energy sector, mining, renewable energy trading, industrial and domestic waste management, and other public and charity sector organisations focussed on conservation ecology and environmental sustainability.
View recent outputs from work within this research theme
Active members and significant achievements
- Dr Markus Mueller and Professor Stuart Townley have completed two Knowledge Transfer Partnerships (KTP) with NJW Ltd. In the built environment and with Hirst Magnetic Instruments Ltd. on magnet characterisation. Drs Saptarshi Das and TJ McKinley are currently collaborating in a KTP with Chelonia Ltd. on data analytics and statistical modelling for marine-life monitoring. The centre also hosts two industry-funded PhD project students.
- Through the projects Marine-i, Smartline and Tevi, the group has established substantive engagement with regional SMEs. In Tevi, Dr Markus Mueller leads the electric vehicle charging network challenge, which draws on multi-objective optimisation methods to enable the decarbonisation of transport in Cornwall. In Smartline, Dr Mueller works with Buzz Interactive Ltd. on the development of a self-assessment and assistance application for Meniere’s Disease.
- Dr Saptarshi Das works on seismic data processing, imaging and fast computational and statistical methods for geophysical inversion, relevant for the oil and gas, mining and carbon capture and storage industries. Recently Dr Das and collaborators have developed statistical pattern recognition algorithms for field emission scanning electron microscope (FESEM) images to study crystal growth modelling in complex biopolymers, with has potential impact in pharmaceuticals. Dr Das and collaborators have also developed artificial intelligence based predictive models for fluidized bed reactors from experimental data for solid waste management. In addition, Dr Das used multi-objective optimization methods to resolve control conflicts to study design trade-offs in automatic voltage control of synchronous generators in power plants. He has also recently developed several time delay handling methods relevant for process control and industrial automation to obtain stability regions for designing robust stabilizing controllers using random search, optimization and sampling approach in fusion with unsupervised machine learning for identifying the robust stable solutions amongst many conflicting control objectives. Drs Tim Hughes and Saptarshi Das have interests in financial systems modelling with deterministic and stochastic models using dynamical systems theory and time series forecasting methods using machine learning and statistical signal processing approaches.
