Coupling of Real-World Data and Fast Response Algorithms to Improve Simulation Correlations and Optimise Construction

The project will develop and deploy MODSUITE within the Caterpillar UK Engines Company Ltd. (CAT). MODSUITE is a
novel data analysis and optimisation tool and will be applied to analyse engine and machine test data against predictive
physics-based models of the processes occurring within the engine. The tool will apply novel optimisation and fastresponse
algorithms to systematically refine and quantify the uncertainty within the models. This will enable CAT to use the
make more effective use of the data and use the models to help optimise engine performance, including gaseous and
particulate (soot) emissions.
The project is split between four main partners. The Computational Modelling (CoMo) Group at the University of Cambridge
will perform the fundamental research and development required to apply MODSUITE to the applications presented by
CAT. Cambridge Computational Modelling Ltd. (CMCL), an engineering software and services company, will focus on
developing a user interface and testing the application of the software. CAT and BorgWarner Ltd. (see
www.borgwarner.com) will provide experimental data and models for real applications to support the software testing.
Federal Mogul Ltd. (see www.federalmogul.com) will act as a subcontractor to CAT for some of the model development.
The software will be developed and tested using three demonstration applications, at increasing levels of system
complexity. The initial phase of the project will be performed at CMCL facilities in Cambridge. Following successful
completion of the initial testing, the software will be deployed at CAT via an internet based user interface. This will enable
the software to be run at CMCL whilst it is still in development, whereas the application models will be run at CAT using
distributed computing technology, allowing CAT to harness the large computing resource at their disposal and maintain
control over the models and data. The web-based interface and distributed computing design offer a simple, but powerful
solution.
The CoMo Group will initially contribute to the project by investigating optimisation methods that are not currently
implemented within MODSUITE. The ones that are most relevant to the applications presented by CAT will be identified
and added. The group will extend MODSUITE so that it can automatically read and process the large quantities of
experimental data made available by CAT, facilitating the creation of data driven models. A wider range of different
response surface methods and an automated response generation and selection method will be investigated and
implemented. This will increase the versatility of the tool such that suitable response surfaces can be generated for each
specific test application. These improvements will facilitate the ability to generate suitable data driven models as well as
fast surrogate models. Advanced optimisation methods will be developed with a focus on self-calibration and robustness to
provide consistent and reliable results without the need for expert knowledge of the specific algorithms. These advanced
algorithms will be combined with uncertainty propagation and analysis tools to quantify the uncertainties in the model. The
MODSUITE code will be adapted to allow it to run over a distributed computing system.

The beneficiaries from the proposed research fall into four categories:
1. Academia. Academics working internationally in the science and engineering communities have shown strong interest in
the novel response surface, optimisation and parameter estimation algorithms behind MODSUITE. The algorithms are
suitable for a wide range of applications and it is anticipated that many academics will apply the methods to investigate
data in their own fields of research. A key example of the potential impact is the development of intelligent experimental
design methods. This is where data are assessed against a model and used to suggest which future experiment would best
inform the model.
The project will establish a core expertise in statistical parameter estimation and uncertainty analysis at Cambridge and will
train a highly skilled post-doctoral research associate. The research associate will gain extensive experience of teamworking,
presenting and communicating the research with both academic researchers and, in particular, with the
employees at the industrial partners. These transferrable skills will be invaluable to the individual, who will be well equipped
to start a successful career in any employment sector.
2. UK research and manufacturing industry. The project will develop and deploy a cutting-edge tool to systematically
analyse vast quantities of existing, and frequently underutilised, data held by the project partners. This will bring an
immediate direct benefit to the industrial partners, where the models and insight developed during the analysis will enable
CAT and BorgWarner to identify cross-platform trends within their data, ultimately helping them improve their products and
achieve better return on investment from their experimental programmes. The project will train the engineers and managers
working at the project partners. The training will be via a combination of presentations, training courses, meetings and
individual tuition and will equip the partners with the knowledge and skills to continue to make beneficial use of the tool long after the completion of the project.
CMCL will directly benefit from the experience of working with two large industrial partners. The project will give CMCL the
experience and track record required to attract clients with similar needs in other fields and will deliver an innovative set of
methods that offer the capability to enhance the research effectiveness, knowledge and skills of many research and
manufacturing organisations, either by companies working directly with the CoMo Group or CMCL, or simply by applying
the methods themselves. Over the longer term, it is hoped that this will establish a cutting-edge capability that brings a
competitive advantage to the UK, making the UK a more attractive prospect for investment from global businesses.
3. Government regulators. The methods are suitable for a wide range of applications. In the longer term they offer the
potential to contribute towards evidence based policy-making by providing quantitative information on the uncertainties
within a set of data or a model, potentially influencing legislation at a local, regional, national and international level.
4. The wider public. The wider public will primarily benefit indirectly from the research via the benefits to CAT and
BorgWarner. The benefits will take the form of reduced energy consumption by more efficient better value-for-money
vehicles, contributing to improved environmental sustainability and better financial performance of each company. In the
longer term, the research will form the basis of a number of further projects. Examples include projects within the
automotive and pharmaceutical industries. In this way it is hoped that the same type of benefits can be derived much more
widely within the UK, ultimately improving the environmental sustainability and contributing to job security and the nation's
wealth.