The global CO2 emissions are expected to rise by almost 5% this year to 33 billion tonnes, as per IEA’s Global Energy Review 2021. The 1.5 billion tonnes increase in Carbon emission is primarily attributed to excessive coal usage in the power sector. “This is a dire warning that the economic recovery from the Covid crisis is currently anything but sustainable for our climate,” said Fatih Birol, IEA Executive Director.
Recently, a research team from IBM has combined high-performance computing (HPC) and hybrid cloud to intensify carbon capture and storage (CCS) goals. The team has developed algorithms to identify and approximate the empty spaces in porous rocks – capillary networks – to utilise the spaces for storing captured CO2 from different emission sources.
The captured CO2 will be compressed to convert it into a liquid state. Next, the liquid CO2 will be injected inside the tiny rock pores, and then it will be mineralised – turning it into solid for efficient storage. It will remain there for decades or even centuries.
“Methodically, the application of X-ray microtomography to rock samples provides a series of images of the spatial distribution of the pore space from which 3-D digital rock representations are created. Once a full series of microscopic rock images is acquired, the image series undergoes a sequence of processing steps with regards to noise, segmentation and morphology, producing a data cube (digital rock) containing voxels that either represent solid or void space of the imaged rock,” according to the IBM paper.
Measurement of rock permeability was done with core plugs. Permeability simulations were performed of the digital rock with a side length of 2.25 mm – acquired from the same rock sample with the help of X-ray micro-CT.
As the spatial distribution of the connected pores is known, a flow model was initiated to predict fluid permeability. Quantifying fluid flow and CO2 processes within a rock’s pores requires the ability to precisely simulate its capillary networks. This is where HPC and cloud come in handy.
Together with São Carlos Institute of Physics-University of São Paulo and Petrobras – a Brazilian multinational corporation in the petroleum industry-IBM has performed the analysis of rock formation, thereby cutting down the time required to do so from months to just days, reducing risks of geological carbon storage.
The work is a part of IBM’s commitment to addressing climate change. Moreover, researchers intend to expand the rock network simulations to incorporate chemical processes and improved materials to enhance CO2 mineralisation.
Countries need to remove between 100 billion and one trillion tonnes of CO2 from the atmosphere to avoid the worst effects of climate change, as per International Panel on Climate Change (IPCC). Additionally, as per BCG estimates, the total value gained by applying AI to corporate sustainability by 2030 could range from $1.3 trillion to $2.6 trillion in increased revenues and cost savings.
Image Credits: BCG
AI to rescue
- Monitoring Emissions: Companies can collect data from their operations, IT equipment, corporate travel and other aspects of the entire value chain. Data from new sources, such as satellites, can be used by AI. AI can also develop approximations of missing data and evaluate the level of certainty of the results by layering intelligence onto the data.
- Predictive AI: It will help predict future emissions across a company’s carbon footprint, taking existing efforts into account, new carbon reduction strategies, and future demand. Companies can more precisely set, alter, and accomplish reduction targets.
