Monitoring of CO2 Flow Under CCS Conditions Through Multi-Modal Sensing and Machine Learning
Why is this research needed?
Measurement and monitoring of CO2 across the entire CCS chain are essential to gain an in-depth understanding of the physical and chemical characteristics of CO2 flow in pipelines and leakage plums over storage sites across different length and time scales. Advances in these areas will also ensure the success of large-scale injection projects and enable the long-term monitoring of captured CO2 in storage sites. This particular R&D need has been identified in several recent status review reports and policy documents, such as the Reports of the Carbon Capture, Utilization and Storage Experts’ Workshops, as well as in the recently released Mission Innovation Action Plan.
What is this research investigating?
This project aims to evaluate a cutting-edge technology for the monitoring of CO2 flow under CCS conditions. The project will start with the establishment of a novel multi-modal sensing system incorporating Coriolis, ultrasonic, differential-pressure and acoustic emission sensors as well as temperature and pressure transducers. Data-driven models based on convolutional neural network and Bayesian statistical learning algorithms will be developed to fuse the outputs from the sensors in order to provide the measurements of flow characteristics in CO2 pipelines or detect the leakage plums from confined CO2 units. An extensive experimental programme will be undertaken to acquire practical datasets under a wide range of CCS conditions to train and validate the machine learning models.
What does the research hope to achieve?
The outcome of the proposed research will impact directly on the rapid deployment of CCS technologies, leading to significant reductions in CO2 emissions from fossil fuel fired power
generation and other energy-intensive industrial processes. This project aims to assess a cutting-edge technology for the monitoring of CO2 flow and plums under CCS conditions. The project objectives are:
- To establish and construct a novel multi-modal sensing system incorporating Coriolis, ultrasonic, differential-pressure and acoustic emission sensors as well as temperature and pressure transducers.
- To develop data driven models and validate them with extensive experimental results for the real-time mass flow monitoring and leakage plum detection of CO2 under realistic CCS conditions.
- To disseminate findings to the UKCCS community via journal publications and conference presentations.