This case study will address:
- Bequeathing a rich legacy of new or improved data sources, technology, infrastructure, facilities, practices, skills or partnerships.
- Low and zero carbon futures (energy)
- Data sharing ecosystems
- Development of skills, including: technical, managerial and methods
With support from Pitch-In, the aim of this project was to use IoT to automate some of the labour-intensive processes involved in collecting big data. The objective was to remove barriers to data access; facilitate research and collaboration between universities and industry; and speed up battery design and development.
Lithium-ion batteries are vital in industries such as electric vehicle manufacturing and grid energy storage. While batteries have continued to improve in terms of their cost and energy density, there is still a way to go to improve battery performance and gathering data about them is crucial to progress in this field.
Vast amounts of data exist about batteries – known as big data. Yet the key question is how to gather this data from its diverse sources? How to declutter and manage it with standardised processes? These are just some of the issues that this Pitch-In project was designed to address.
What were the problems or barriers?
- Testing battery lifetime and performance is time-consuming, expensive and requires specialist equipment.
- A wide range of testers are available, but there is no standard way to share and store data about them.
- In the energy sector, there is a large amount of data for the research community to work on. However access issues, including security and data sharing are major barriers to research and development
- Access for everyone involved in researching solutions can be challenging when they are spread across different functions in an organisation and also when they are geographically spread out.
- There is a lack of specialists with the technical know-how in IoT as well as a lack of skills in other disciplines vital to furthering developments in this area including front and back-end software development, analytics and connectivity.
- There is a lack of knowledge of how analytical tools can be used for leveraging IoT data.
What did you do?
Finding ways to access battery data
Researchers at the universities of Oxford, Newcastle and Sheffield came together to identify the barriers faced by the battery community in relation to IoT. Data de-cluttering and management were identified as major obstacles and researchers at the University of Oxford investigated the different analytical tools and software solutions available to tackle this problem.
Subsequently, Professor Howey’s Battery Intelligence Laboratory worked with engineers at the Oxford Robotics Institute to build a software package which they called Galvanalyser. This is software to extract data from battery testers, parse it into a standard format and store it in a database. Working with the IT infrastructure team at Oxford’s Department of Engineering Science, a server was built and housed by the Oxford e-Research Centre. The server was designed for storing and managing data through a database, to enable querying and data plotting through a web application and handling of the connection between browser clients and the webapp. Further work with research software engineers at the Research Software Engineering Group in the Department of Computer Science streamlined the installation process and user interface.
The software was also developed to enable easy backups to be made and for users to conveniently access data through a web application or directly using an API. Work was done to improve its functionality including security, usability, and the management of metadata.
Finally, improvements to useability continued following user feedback and dissemination amongst the battery modelling community. The software was discussed and reviewed with over 600 battery researcher partners in the international battery modelling community.
What was the result?
A tool was created for processing and storing large amounts of data produced by different battery testers. It has already received wide interest from members of the international battery modelling community.
The project delivered on three core measures of success:
1. Delivery of validated tools and software for processing and storing large amounts of data: A software suite was developed that processes and stores large amounts of data from different battery testers. As a result, data can be exchanged and analysed more easily, speeding up battery design and development work.
2. Demonstration of storage, access and processing of a battery data set in line with best practice in other industry sectors: The software is based on a popular database system (PostgreSQL), which is widely used across many sectors.
3. Publishing an example of data storage and processing: An introduction to Galvanalyser, together with an illustrative example have been published by the open access service ArXiv, arXiv:2010.14959 .
- As the commercial success of batteries continues, the value of improving our understanding of these complex devices and accelerating efforts to learn from data goes up.
- Extensive discussions involved a wide range of academic and industry collaborators, both nationally and international. This led to a common understanding of the issues around managing battery testing data. This enabled a clearly defined common problem.
- The creation and use of a new specialist software development group in computer science helped to make software development more professional.
- The experts involved are usually specialists in batteries or data. Therefore the use of a multi-disciplinary team including battery, data sciences and software specialists was key.
- Connecting capabilities: It was vital for us to involve professional research software engineers to help with development.
The development of a database solution is in the process of being rolled out to beta testers.