The third Uncertainty Quantification and Mangagement Study Group with Industry will take place at Warwick Centre for Predictive Modelling from 13-15 December 2017.
The participant information pack contains full details of the event including industry use cases, schedule, list of participants.
Industry Use Cases
The industrial challenge problems to be addressed are as follows:
- AWE Developing a Surrogate Model for Predicting Metallic Corrosion Events.
The problem involves hydrogen transport through a multi-layered material substrate. Transport is hindered by the presence of trap sites which lead to nonlinear kinetic terms arising in the diffusion equation. The model predicts the time-dependent concentration within the material and subsequent nucleation of corrosion sites. The desired outcomes are 1) the development of a surrogate response surface that can be interrogated quickly by higher-level models, and 2) a proposed set of validation experiments which would optimise the model assessment process.
- AstraZeneca Improved Drug Discovery Through Better Machine Learning Models.
There is a strong interest in machine learning methods for drug discovery. While proven useful in QSPR (Quantitative Structure Property Relationships) modeling to predict physicochemical properties like lipophilicity, there is a renewed interest to increase the application domain of machine learning as well as take a fresh look at classical machine learning tasks to predict the bioactivity of a compound and the ADME (Adsorption, Distribution, Metabolism and Excretion) of a compound. AstraZeneca are interested in investigating different methods that could increase their predictive accuracy and make a comparison to their current standard approach. One approach is to investigate advances in deep learning and molecular graph convolution. Participants are encouraged to build models either with scikit-learn or for deep learning within the DeepChem framework. Training and test datasets will be provided.
- HS2 Visualising, Communicating and Managing Risk in Large Infrastructure Projects.
The primary objective of this challenge is to develop improved communication and transparency of risk information in a succinct manner but with highly flexible, easy to use, interactive drill-down (and/or alternative views) to promote engagement and understanding of the risk "big picture" and offer the prospect of gaining greater strategic insights.
- Zenotech Machine Learning for Wind Energy Modelling. The United Kingdom is one of the best locations for wind power in the world, and is considered to be the best in Europe. Wind power contributed 11% of UK electricity generation in 2015, and 17% in December 2015. Allowing for the costs of pollution, onshore wind power is the cheapest form of energy in the United Kingdom. The use of computational fluid dynamics (CFD) to assess the wind energy resource for a prospective new site is an established method that has been used for many years, however the inclusion of wake interaction effects - particularly for larger arrays of turbines - is less mature. The SWEPT2 consortium has been developing new tools to improve the utility and accuracy of CFD-based wake interaction modelling. Large amounts of data can now be generated and consolidated to inform engineering and investment decisions. The objectives of this challenge is to explore how well machine learning methods might be able to infer the interaction patterns and consequent power production from sites, given sufficient training data.
Study Group Programme
All participants will be provided (at no further cost other than the initial £15 - £60 registration fee) with lunch, dinner and coffee from Wednesday morning until the workshop finishes on Friday afternoon.
Accommodation will be provided in the Arden Conference Centre at the University of Warwick. If you have requested accomodation you will have been sent details of how to reach the conference center and when you can check in.
The event is supported by the Uncertainty Quantification and Management in High Value Manufacturing special interest group of the Knowledge Transfer Network and by the School of Engineering at Warwick.