LABS
Assessing geomagnetic storm risks across power network assets
GIC’s can affect power grid components. This lab explores developing tools that allow users to simulate storm scenarios and visualise the resulting local risk across the UK power network.
Image Courtesy of SOHO/[instrument] consortium. SOHO is a project of international cooperation between ESA and NASA.
LABS
Reducing the time it takes to plan new connections
IConn generates options for where a new generation or demand project can connect to the network providing a breakdown of capacity, cost, time estimates, and transmission line routing.
LABS
Automatically recognising signs of potential open pit mine slope collapse from aerial imagery
We are using AI to detect cracks early helping mine operators take action before a dangerous and costly collapse.
LABS
Making use of prediction uncertainties in neural network outputs
We’re trying to figure out how we can use prediction uncertainties in deep learning network outputs to minimise false positives and prepare more targeted training data.
LABS
Transforming Environmental Management with AI
We use AI to better understand, manage and protect our natural world.
LABS
Transforming Biodiversity Management with AI
We developed a camera system and AI platform for detecting the presence of invasive plant species along linear infrastructure such as rail tracks and roads.
LABS
Detecting leaks and manufacturing imperfections
Keen AI developed simple computer vision based tools to highlight areas in a video feed where leaks are present.
LABS
Identifying Sand Dunes with AI
Shifting sands pose a threat to rail safety and overhead line clearances. We’re using satellite and ground based imagery to detect dunes and encroaching sand and then trying to model how it will move.
Keen AI Labs explore new use cases, technologies and approaches. Many work, others don’t, all are fascinating. Do you have a process that can be improved with AI? Learn more about our Labs process here, and our collaborators.
Partners:
Image source: SOHO (ESA & NASA)
active
Assessing geomagnetic storm risks across power network assets
Synopsis:
Geomagnetic storms, driven by solar activity, induce electric currents known as Geomagnetically Induced Currents (GICs) in power grids. These currents can affect a range of network components, including transformers, transmission lines, capacitor banks, and protection relays. Although aspects of this process are well-studied, there is a clear need for tools that allow users to simulate different storm scenarios and visualise the resulting local risk across the UK power network.
Outcomes:
We have developed an automated tool, SolarStorm.Space, that integrates real-time geomagnetic data, machine learning, and UK-specific infrastructure information to evaluate the risk of GIC-induced damage across various power network assets. This tool provides asset managers with clear visualisations of potential impacts, enabling informed decision-making to mitigate risks and enhance grid resilience.
Further Reading:
Researchers:
Cristian Silva-Perez
Ruby Pearce-Casey
active
Reducing the time it takes to plan new connections
Synopsis:
Networks need to be more resilient to climate change and demand for new connections is rapidly growing. Initial viability evaluations are much harder and take longer than they should.
Outcomes:
IConn enables more holistic decision-making, reducing the burden on system design engineers and providing reliable information earlier on in the process.
Further Reading:
Researchers:
Cristian Silva-Perez
Hamzah Reta
active
Automatically recognising signs of potential open pit mine slope collapse from aerial imagery
Synopsis:
Soil cracks running parallel to the pit edge are a critical indicator or precursor of slope instability in open-pit mining operations. Detecting cracks early can help mine operators take action before a dangerous and costly collapse.
Outcomes:
By visually inspecting multiple examples of detected cracks, we can assess the pipeline’s ability to accurately identify and segment cracks based on their risk of slope instability.
Further Reading:
Researchers:
Cristian Silva-Perez
Amjad Karim
active
Monitoring sand dunes using satellite and train mounted camera data
Synopsis:
Shifting sands pose a threat to rail safety and overhead line clearances. We’re using satellite and ground based imagery to detect dunes and encroaching sand and then trying to model how it will move.
Researchers:
Dr. Ahmed Mahmoud
Amjad Karim
Petar Gyurov
Complete
Robots looking for bubbles to detect leaks and spot manufacturing imperfections
Synopsis:
Keen AI developed simple computer vision based tools to highlight areas in a video feed where leaks are present and present to a human who can repair the leak.
Outcomes:
Further Reading:
Researchers:
Hamzah Reta
Amjad Karim
Complete
Neural networks and camera systems for detecting invasive plant species
Synopsis:
We developed a camera system and AI platform for detecting the presence of invasive plant species along linear infrastructure such as rail tracks and roads.
Outcomes:
We trained models able to detect Himalayan Balsam and Ragwort from photographs taken from the car mounted system. Vehicle speed had to be kept under 30 mph for safety and for contiguous footage. Future work to focus on increasing speed to at least 50 mph.
Further Reading:
Researchers:
Hamzah Reta
Amjad Karim
Petar Gyurov
Dr. Tom August
cOMPLETE
Bootstrapping computer vision model training using synthetic data
Synopsis:
Models require data to train on. Labelling this data manually is time consuming or the data may not exist. We’ve been working with a variety of techniques for generating synthetic data.
Outcomes:
Further Reading:
Researchers:
Thomas Emment
Hamzah Reta
Petar Gyurov
Amjad Karim
COMPLETE
Using model uncertainty to reduce manual labelling effort
Synopsis:
Researchers:
Qifan Fu
Prof. Greg Sablaugh
Petar Gyurov
Amjad Karim
Complete
Identifying ash trees at risk of ash dieback
Synopsis:
Following the learnings from our previous study on identifying invasive species from roadside images, and continuing our partnership with UKCEH, we set out to devise an effective operational pipeline for surveying the ash tree population and diagnosing ash dieback.
Outcomes:
Key insight was determining when in the year to collect footage. Ash tree seed pods are very distinctive and easier to detect in photos collected in late autumn.
Further Reading:
Researchers:
Petar Gyurov