This presentation was made at the NAFEMS European Conference on Simulation-Based Optimisation held on the 15th of October in London.
Optimisation has become a key ingredient in many engineering disciplines and has experienced rapid growth in recent years due to innovations in optimisation algorithms and techniques, coupled with developments in computer hardware and software capabilities. The growing popularity of optimisation in engineering applications is driven by ever-increasing competition pressure, where optimised products and processes can offer improved performance and cost-effectiveness which would not be possible using traditional design approaches. However, there are still many hurdles to be overcome before optimisation is used routinely for engineering applications.
The NAFEMS European Conference on Simulation-Based Optimisation brings together practitioners and academics from all relevant disciplines to share their knowledge and experience, and discuss problems and challenges, in order to facilitate further improvements in optimisation techniques.
In today's engine market, optimising durability and performance whilst reducing cost is essential to remain competitive. This is a complex trade-off; thus, we ensure our engines are durable through the combined use of simulation and testing.
One of the key drivers in the diesel engine business is to reduce the fuel consumption. Engine down-sizing, replacing a larger engine with a smaller swept volume engine, offers one solution to meet this need. The challenge associated with down-sizing is increased in-cylinder thermal loading. Customers still however expect a durable product and hence there has been increased focus on optimising the design of these down-sized engines.
Optimising the complexity of a cylinder head is not economically viable nor fast enough using test methods alone. This, and many other conflicting requirements, has driven the increased use of simulation methods to support product development through generating a detailed understanding of the component and system performance and their associated interactions.
In Simulation led design the design space is explored, and the design informed by that exploration, before the geometry is finalised for full simulation validation. Simulation based design studies have been used for several years at Caterpillar, growing in complexity from trial-and-error to One-Factor-At-a-Time (OFAT) to Design of Experiment (DoE). At all stages a unique combination of parameters is specified resulting in the need to update one or many models, before solving them and extracting the results. This results in a process with significant non-value-added time, and a high risk of systematic error due to the repetitive creation of the simulation and extraction of the results.
Multi-disciplinary Optimisation (MDO) tools allow for the creation of workflows that can automatically run a simulation study such as a DoE or an optimisation. These tools significantly reduce the non-value-added time and the opportunity for systematic errors by automatically creating simulation models and extracting results. This provides an opportunity for junior engineers to accelerate their learning of the inherent engineering in the workflow process, whilst the expert decision-maker can get access to results of a deeper level quicker or explore a much greater design space than previously possible. Incremental benefits include the reuse of the MDO workflow in similar projects, reducing the time to deliver that project and minimising project cost.
This paper will look at the use of MDO as part of simulation led design to conduct an optimisation of a simplified parametric cylinder head. The objective of the project was to find an acceptable trade-off between durability and cost, whilst meeting performance targets.
|15th October 2019