Developing early designs with high confidence in downstream mitigation strategies: use case.
How an aerospace firm used surrogate models to solve target load concerns.
This article is part of our series on High value manufacturing: dealing with the unknown
It is vital to develop early stage design concepts with confidence in how to mitigate risks associated with unknown scenarios. Getting the parameters wrong in the early design stage may mean costly corrections further down the line, or inefficiencies throughout the product lifetime, incurring further cost and lack of market confidence.
High value manufacturing works with complex processes across distributed supply chains, and optimising these processes is key to competitive advantage. In the aerospace industry, for example, mistakes at an early stage may result in expensive redesign, reduced performance and costs associated with not meeting customer requirements.
A case in point: aerospace
At entry into concept, the design is at a very early, unconverged state. Yet 80% of the future development costs are already committed. Continually quantifying, understanding and managing uncertainties enables skillfully committing the 80% as the design progress, referenced in Why UQ&M in Industrial Design?, a presentation given by Sanjiv Sharma, Airbus UK, to RIBA on 11 April 2016.
At a certain point, a number of performance requirements are ‘frozen’. These may include the aircraft target loads; the limiting loads – including flight conditions, wind conditions, turbulence, etc – that the aircraft must be designed to withstand. The analysis process by which the limiting loads are established is very complex and computationally demanding, due to the extremely high number of potential gust and manoeuvring conditions.
The objective is to anticipate the certification load level, scale the calculated loads to match this anticipated level and issue this data as target loads. This provides a stable set of loads for use in the design process, enabling a robust downstream programme while minimising the risk of not meeting the requirements.
There is an important balance to be achieved between risk and design implications. If the load level is too low, it may require costly redesign after the testing phase. If the level is too high there is the risk of unnecessary structural weight, which would lead to not meeting customer expectations.
In this case study, the following question was posed: ‘Can formal Uncertainty Quantification and Management (UQ&M) methods be deployed to improve and underpin confidence in setting target loads, thus reducing conservatism and the weight of the aircraft or component?’
Simulations used to model complex scientific phenomena are usually very computationally expensive, as they address many hundreds of thousands of possible load cases. To meet this challenge, surrogate models are used to replace the computationally demanding model with inexpensive substitutes. The methods can also capture the “uncertainty” introduced by making such a substitution.
Before the design is fully frozen, margins are allocated to account for uncertainties due to lack of definition, so that later decisions will not overwhelm the current design solution. Inverse methods are used to identify the allowable design uncertainties, given specified probabilities of constraint satisfaction and affordable margins.
The key benefit to flow from a demonstrable UQ&M capability is improved confidence levels in the target load setting decision gates, limiting the risk of penalties incurred by either over or under design.
Read more use cases in this series
More information about upcoming UQ&M SIG group activities can be found here.