Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Machine Learning for Technical Report and Data Insights

Digital and data-driven technologies are shaping the future of engineering, from the deployment of digital twins for structural integrity to the application of machine learning and artificial intelligence for the design of new alloys. However, one of the biggest challenges for data-driven technologies is the availability of databases of sufficient breadth and depth.

TWI has a wealth of archived data with exceptional depth and breadth from previous projects, which can be leveraged in an effective and secure way. The aim is to demonstrate data-driven technologies in three high-interest areas; natural language processing for failure investigations, automatic image processing for additive manufacturing micrographs and machine learning applied to mechanical properties.

Utilising existing TWI datasets and open source software, TWI will demonstrate the benefits of accelerating the adoption of digital technologies for engineering applications, leveraging TWI’s multi-disciplinary expertise across the technology sections.

TWI is developing capabilities for the automated key-wording of new Weldasearch articles through the application of natural language processing. Automating this process will save time and facilitate an improved search capability to interrogate past reports, enabling greater access to knowledge.

Addressing one of the fundamental challenges in additive manufacturing, TWI is developing the capability to automate the image processing of micrographs and characterisation of additively manufactured samples. This work will lead to improvements in the understanding of the link between the process parameters and properties of additively manufactured samples.

There is significant value in historical datasets and TWI has been performing tensile strength and fracture toughness testing for many years, accumulating a wealth of data in the process. This work will analyse the past test results, extract insights and develop predictive capabilities of future test outcomes. These developments will inform and provide insights to operators.

The specific activities mentioned above are being undertaken under TWI’s Core Research Programme.

}