Data Science and Artificial Intelligence for Sustainability MSc

• Dr Joanna Zawadzka

Geographical information is increasingly prevalent in our daily life, affecting personal leisure activities as much as business services and the workplace. Geographical information represents a key theme in environmental management and it has been estimated that some 80% of the data used for environmental, business and policy-oriented decision-making is geographical in nature. Such spatial data requires a structured approach in its management if the maximum benefit is to be derived from analysis and dissemination. This module provides a solid introduction to the issues concerning the management of spatial information and the tools to do so, with a predominant focus on ESRI’s software solutions. 

Spatial data models and database structures – vectors, rasters, topology, ordered and indexed lists, hierarchical, network, relational, object oriented, hybrid, metadata.
Mapping fundamentals – geodesy, projections, cartography, abstraction of the real world into map form.
Analysis approaches – database manipulation, reclassification, overlay, spatial modelling.
Data specification and standards – use cases, interoperability, metadata.
Data visualisation.

On successful completion of this module you should be able to:

• Define the functional components of a GIS, projections, data and modelling processes for managing spatial data.
• Organise and integrate, using appropriate data structures, spatial and aspatial data within a GIS.
• Analyse, evaluate and prepare data within appropriate spatial databases structures for information dissemination.
• Examine the role of metadata and FAIR data principles for enabling interoperability between spatial data infrastructures.
• Perform and disseminate the results of analysis and data manipulation via maps, tables and other appropriate media.

• Dr Alice Johnston

The module introduces the context of environmental decision making, providing both a conceptual overview and practical tools for addressing environmental management problems. It aims to promote an understanding of weight-of-evidence approaches used to inform real-world problems by research, industry, and government. Central to this is an understanding of the strengths and limitations of different approaches and aspects of decision science, the social, economic, and environmental trade-offs made during decision-making, and how real-world complexity and future uncertainty is accounted for. Students will learn how to apply systems thinking to evaluate different courses of action in response to environmental challenges related to land, water, and/or energy. 

• Key environmental challenges and the need for decision support tools.
• Different stakeholders in environmental decisions.
• Weighing the evidence for different courses of action.
• Big data; data analytical approaches; mathematical modelling; statistical inference; machine learning; information systems; spatial data science approaches. 
• Environmental, economic, and social trade-offs in decisions. 
• Accounting for real-world complexity and future uncertainty in decisions.

On successful completion of this module you should be able to:

• Evaluate the strengths and weaknesses of decision science approaches for practical environmental management problems.
• Conceptualise complex environmental issues using systems thinking to weigh-up different courses of action.
• Critically appraise the need for stakeholders to trade-off environmental, economic and social concerns, and the limitations of decision science approaches in accounting for real-world complexity and future uncertainty.

• Dr Daniel Simms

Writing code opens-up new approaches to creative problem solving and allows us to move beyond the limitations of any particular software. This module aims to develop your skills and confidence to write your own code in the Python programming language and access a broad ecosystem of packages and tools that comprise the foundations of Data Science.

Fundamentals of programming.
Python syntax, types, calculations, variables, strings, and object-oriented programming concepts.
Branching, iteration, and recursion.
Abstraction, functions, and classes. 
Files, packages, and imports.
Testing, debugging, and exceptions.
Version control.
Writing efficient code.
Python scientific packages for data analysis and visualisation.
Geospatial data.
Jupyter Notebooks and Colaboratory.
Statistics for Data Science.

This module will introduce you to data analytics, overview challenges and solutions in using data analytical tools to solve sustainability problems. The module will present the principles of data ethics, data governance and regulation required to ensure societal good. The lectures and practical sessions will present approaches for knowledge discovery, covering predictive and descriptive data mining, classification, statistical methods, regression models and time-series forecasting methods. You may benefit from knowledge of basic programming and concepts of statistics methods. 

• Introduction to Data Analytics 
• Application of Data Analytics in sustainability 
• Statistics refresher and data pre-processing 
• Predictive analytics: regression refresher, classification and forecasting methods 
• Clustering and dimensionality reduction 
• Graph analysis and visualization 
• Software and tools for data analytic 
• Case study: application of data pre-processing and data analytical tools for a specific dataset 

On successful completion of this module you should be able to:

• Critically analyse stages of the data analytics workflow; and establish a data analytics workflow based on the available data and formulated requirements.
• Identify, assess and minimise the risk related to data engineering, whilst ensuring ethical practices, data security, privacy and quality.
• Analyse and apply algorithms for discovery of new information from the large data sets, using statistics, regression, classification, forecasting methods.
• Critically evaluate data analysis and visualisation techniques with respect to data analytics stages, using graph analysis and visualisation techniques
  
  

• Dr Da Huo

With more and more measurement and control devices installed in energy systems, data analytics using AI technology to support planning and operation of energy systems has shown significant advantages. The scientific and technical concepts of machine learning and AI methods/tools and their potential advantages in the energy sector will be taught in this module. One example of this is to use smart metering data to analyse a network’s hosting capacity of solar photovoltaic systems, and to analyse a power system’s technical and non-technical energy losses.  The module aims to provide you with data analytical skills from machine learning and AI technology, and evaluate the advantages/disadvantages of their applications in the energy industry. The module also aims to provide you with essential skills (e.g. computer programming and coding in Python) for applying machine learning and AI in the energy industry.

• Design of an appropriate analysis toolkit specific to analyse the examples of applications of machine learning and AI technology in energy industry,
• Analysing the development and scaling/design of the AI technologies by evaluating the advantages/disadvantages of the available examples in the areas of applications in energy systems,
• AI techniques (e.g. Artificial Neural Networks, RNN, Reinforcement Learning), and skills to manage, process and use data to support network operation and planning,
• Techniques to evaluate where AI can be used and the potential benefits to the energy industry. 
  

On successful completion of this module you should be able to:

1. Critically analyse the state-of-the-art of the applications of machine learning (ML) and AI technology in the energy industry,
2. Identify and assess the requirements of different AI/ML techniques and their contributions to improve the planning and operation of energy systems;
3. Implement AI/ML algorithms, estimate their performance in a simulation environment and assess their performance for a realistic case study
4. Evaluate the advantages and disadvantages of particular AI techniques within the context of the energy industry
   
   

• Dr Georgios Pexas

Environmental impact assessment and life cycle analysis are important tools for evaluating the sustainability of complex renewable energy technologies and industrial processes or products. The tools and concepts taught in this module will enable you to assess the sustainability of a case study from an environmental standpoint. Analysis of relevant case studies to demonstrate the assessment process, including how to account for uncertainty and sensitivity analysis.

This module is 10 credits.

• Environmental Impact Assessment,
• Indicator selection and analysis,
• Life cycle analysis and carbon footprinting,
• Social impact assessment,
• The use of appropriate software for lifecycle analysis.

On successful completion of this module you should be able to:

• Critically assess the emissions and waste production throughout the lifecycle of a technology or process,
• Design a framework to ensure compliance of a process, product or service to support the transition to Net Zero that is  compliant with regulatory and voluntary requirements,
• Critically evaluate different environmental and social appraisal metrics,
• Design and implement a strategy to assess the environmental sustainability of a process or technology, and evaluate the associated uncertainties.

• Dr Patrick Verdin

To appraise existing Computational Fluid Dynamics (CFD) techniques and tools for modelling, simulating and analysing practical engineering problems related to renewable energy, with hands on experience using commercial software packages used in industry.

• Introduction to CFD: Introduction to the physics and understanding of governing equations (continuity, momentum, energy and species conservation) and state of the art Computational Fluid Dynamics including modelling, grid generation, simulation, and high-performance computing. Case study of industrial problems and the physical processes where CFD can be used,
• Computational engineering exercise: specification for a CFD simulation. Requirements for accurate analysis and validation. Introduction to turbulence and practical applications of turbulence models, introduction to turbulence and turbulent flows, traditional turbulence modelling,
• Advanced turbulence modelling: introduction to Reynolds-averaged Navier Stokes (RANS) simulations and large-eddy simulation (LES),
• Practical sessions: offshore renewable energy problems (flow around wind and tidal turbines) will be solved employing the widely-used industrial flow solver software FLUENT. These practical sessions will cover the entire CFD process including grid generation, flow solver, analysis, validation and visualisation.

On successful completion of this module you should be able to:

• Assemble and evaluate the different components of the CFD process,
• Explain the governing equations for fluid flows and how to solve them computationally,
• Compare and contrast various methods for simulating turbulent flows applicable to civil and mechanical engineering, especially offshore renewable energy applications such as wind turbines and tidal turbines,
• Set up simulations and evaluate a practical problem using a commercial CFD package,
• Design CFD modelling studies of renewable energy devices.

In this world of downsizing, restructuring and technological change, notions of traditional careers and ways of creating value have all been challenged. People are depending more upon their own initiative to realise success. Never, it seems, have more people been starting their own companies than now, particularly to exploit the World Wide Web. There’s no single Government (in either the developed or the developing world), which is not paying at least lip service to enterprise development. The aim of this module is to provide you with knowledge and skills relevant for starting and managing new ventures across the entrepreneurial life cycle. Moreover, it will prepare you on how to prepare a business pitch to an investor.

• Entrepreneurial risk, performance and environment,
• Business planning techniques and their application in entrepreneurial ventures,
• Venture strategy in dynamic markets,
• Start-up and resources to exploit a profit opportunity,
• The evolution of the venture and managing growth,
• Protecting and securing intellectual capital: IPR and antitrust law,
• Financial management for new ventures: financing a start-up,
• The entrepreneurial financing process: buying and selling a venture.

On successful completion of this module you should be able to:

• Assess the impact of the business environment on entrepreneurial opportunity identification and exploitation.
• Critically apply the theoretical underpinning of entrepreneurship to the process of managing risk in new ventures and supporting their development.
• Compare and contrast how managerial challenges vary across the life cycle of an entrepreneurial venture.
• Assess the likely financial needs of a new venture and pitch for finance.
• Develop and write a credible business plan for a new venture.

• Dr Peter King

The module aims to provide you with a deep understanding of the truly multidisciplinary nature of a real industrial project. Using a relevant case study, the scientific and technical concepts learned during the previous modules will be brought together and used to execute the analysis of the case study.

Design of an appropriate analysis toolkit specific to the case study.

Development of a management or maintenance framework for the case study.

Multi-criteria decision analysis [MCDA] applied to energy technologies to identify the most preferred technology.

Energy technologies and systems: understanding the development and scaling/design of the technologies by applying an understanding of the available resources in the assigned location.

Public engagement strategies and the planning process involved in developing energy technologies.

On successful completion of this module a student should be able to:

• Critically evaluate available technological options, and select the most appropriate method for determining the most preferred technology for the specific case study;
• Demonstrate the ability to work as part of a group to achieve the stated requirements of the module brief;
• Organise the single-discipline activities in a logical workflow, and to define the interfaces between them, designing an overall multidisciplinary approach for the specific case study. 

• Dr Gill Drew

The purpose of this module is to provide you with experience of scoping, designing and delivering of a short research project. This requires an understanding of the background literature, as well as relevant analysis techniques. You will need to agree the project scope early on and deliver the project within the two weeks of the module. The module will allow you to draw on the experience and learning from the previous modules. 

Technical requirements specific to project brief and relevant to the MSc course and route.

On successful completion of this module you should be able to:

• Deliver a research project, including identification of research methods,
• Design a data analysis method appropriate to their chosen research topic,
• Execute a short project, analyse the outcomes and provide sound recommendations.