Dr Edoardo Patelli

ZZ (DO NOT USE) was Civil Engineering and Industrial Design

Research

Research Overview

* LAST UPDATE: SEPTEMBER 2019 *

Generalised probabilistic approaches and advanced stochastic computational methods

Effect of the uncertainties
Effect of the uncertainties

In engineering practice, the evaluation of risk is often performed adopting over-simplified models and subjective judgements of experts which reduce considerably the credibility of the predictions in quantitative terms. Research is therefore required to develop more accurate models for risk assessment that are able to include vague and imprecise information and to identify the features, events and processes that influence the system integrity. Such models require in turn the availability of very efficient uncertainty quantification tools required to estimate the probability of occurrence, or intervals of probability, of events and their consequences.

Current research projects are:
* Uncertainty and Reliability of Systems and Networks
* Robust Design Optimization of Structural Systems
* Efficient Numerical Methods for Uncertainty Quantification in Engineering

General purpose software for uncertainty quantification and risk analysis

Needs of an innovative software
Needs of an innovative software

Computer-aided modelling and simulation is now widely recognised as the third `leg` of scientific method, alongside theory and experimentation. Many phenomena can be studied only by using computational processes such as complex simulations or analysis of experimental data. One of the greatest challenges of virtual prototyping is to improve the fidelity of the computational analysis. This can only be achieved by explicitly including variability and uncertainties from different sources.

Stochastic methods offer a much more realistic approach for analysis and design, but they are generally computational expensive. Hence, scalable computational tools are necessary, i.e. by making use of the computational power of a cluster and grid computing.

A multi-disciplinary software suite for uncertainty management and risk analysis is under development. The computational tools satisfy the industry requirements regarding numerical efficiency, flexibility, scalability and analysis of detailed models that can be used to solve a wide range of engineering and scientific problems. The availability of such software is particularly important for the analysis and design of resilient structures and safety critical systems. COSSAN software

Current research projects are:
* Reliability/robustness-based approaches and computational tools for multidisciplinary systems under mixed aleatory and epistemic uncertainty
* Efficient and Energy-Aware Software for Stochastic Analysis on Large-Scale Systems
* Efficient Numerical Methods for Uncertainty Quantification in Engineering

Risk analysis, Nuclear Safety and Probabilistic Risk Assessment

Risk is the potential of experiencing a loss when a system does not operate as expected due to the occurrence of uncertain, and difficult-to-predict events. Risk assessment requires the quantification of not only the direct cost of system failure but also the accompanying multi-faceted failure consequences that cascade across the boundaries of every disciplines and sectors of society. This is illustrated by a well known disaster such as the Fukushima nuclear incident, where natural events caused an avalanche of inter-related effects on the safety systems of the plant and subsequent contamination of the environment.

Probabilistic safety assessment methods is a generic term for the integrated analysis of risks arising from plant and processes which are undertaken by applying structured and systematic analysis techniques. Although some issues have been addressed in recent years, there is still a need to continue improving the probabilistic safety assessment methods. Every disaster is unique, and the availability of robust and fast predictive models able to deal with scarce and limited data is of fundamental importance, with the aim of having more realistic analysis that can support safety related decision at nuclear installations and mitigation actions will result in fewer deaths and less damage in case of severe accident.

Current research projects are:
* Smart Online Monitoring of Nuclear Power Plants
* Uncertainty Quantification and Risk Assessment Methods for Natural Hazards Triggering Simultaneous Failures
* Risk mitigation: Human errors and preventive design and project management
* Optimal Risk and Benefit Sharing and Management in Large Energy Projects

Research Grants

Informed mining: risk reduction through enhanced public and institutional risk awareness (IM AWARE)

ECONOMIC AND SOCIAL RESEARCH COUNCIL

November 2019 - March 2023

A Resilience Modelling Framework for Improved Nuclear Safety - NuRes

ENGINEERING & PHYSICAL SCIENCES RESEARCH COUNCIL

January 2019 - September 2021

Uncertainty Quantification in Fusion Power Plant Design

CULHAM CENTRE FOR FUSION ENERGY (UK)

October 2018 - October 2022

Realistic model prediction for managing risk in nuclear decommissioning Generating realistic synthetic plant data for future scenario analysis CDT

NATIONAL NUCLEAR LABORATORY LTD (UK)

October 2017 - September 2021

Smart on-line monitoring for nuclear power plants (SMART)

ENGINEERING & PHYSICAL SCIENCES RESEARCH COUNCIL

December 2015 - May 2018

Climate Change and Extreme Weather Impact on Hydropower Installations.

DEPARTMENT FOR BUSINESS, INNOVATION AND SKILLS (UK)

April 2016 - March 2017

Large multipurpose platforms for exploiting renewable energy in open seas (PLENOSE).

EUROPEAN COMMISSION

May 2014 - April 2018

Research Collaborations

Jiamei Deng

External: Leeds Beckett University

Smart Online Monitoring of Nuclear Power Plants

Rong-Jiun Sheu

External: National Tsing Hua University

Radiological safety evaluation and uncertainty analysis related to spent fuel storage installation or nuclear decommissioning. Dual PhD programme in Nuclear Safety

Zhan KANG (亢战)

External: Dalian University of Technology

Structural optimization under non-probabilistic uncertainties.

Virtual Engineering Centre

Internal

Robust Design. Uncertainty quantification.
Developing and exploit a general purpose software for uncertainty quantification and risk analysis