UK Coal

The acid production reaction between CO2 and water in rocks changes the pore structure, thereby altering the rock mechanical properties. To ensure the safety of CO2 storage, it is critical to identify the rock damage mechanisms, clarify the variation patterns of mechanical properties under damage conditions, and investigate the flow characteristics of two-phase CO2-water systems under complex mixed-wettability. This study established a constant-temperature and -pressure multiphase CO2-rock immersion experimental system. Additionally, a comprehensive experimental procedure integrating CO2 immersion, CT imaging, and uniaxial compression was designed. A comparative study revealed that reactions involving gaseous CO2 and ScCO2 with water increased the core porosity by 1.2% and 2.9%, respectively. During the immersion period, gaseous CO2 accumulated on the upper section of the rock, causing a sharp increase in the surface porosity of the upper section (Δϕ = 0.9%). By contrast, ScCO2 accumulated on the middle section, causing a substantial increase in the surface porosity of the middle part (Δϕ = 2.8%). After the CO2 saturation in different phase states, the pore connectivity of the core was enhanced. When exposed to gaseous CO2, some isolated micropores became interconnected, thereby increasing connected porosity from 13.0 to 13.9%. When exposed to ScCO2, the connected porosity increased from 11.3 to 11.7%. Under damage conditions, the primary factors contributing to the deterioration of mechanical properties were the expansion of the pore volume and the increase in the number of connected pores. Compared to gaseous CO2, ScCO2 generated stronger carbonic acid upon reaction with water. After the acidification reaction, the number of both interconnected and isolated pores within the core increased, leading to considerable changes in the mechanical properties. Specifically, the compressive strength and elastic modulus decreased by 19.08 and 16.2%, respectively, and the Poisson’s ratio increased by 26.9%. Under the single-phase wettability, the displacement efficiencies of gaseous CO2 and ScCO2 were enhanced under weak and strong wettability conditions, respectively. The displacement efficiencies of gaseous CO2 and ScCO2 under complex mixed-wettability decreased by 5.7 and 15.3%, respectively, compared with those observed under single-phase wettability. Therefore, neglecting the complex mixed-wettability of pore wall surface often leads to the overestimation of displacement efficiency.

As the key channel in the operation of coal mine, the structural stability of the shaft is directly related to the safety production of mine. In order to improve the accuracy and efficiency of wellbore deformation monitoring, 3D laser scanning system, as a new technology, is gradually applied to wellbore deformation monitoring. Due to the existence of lifting equipment, pipelines, cables and other auxiliary facilities inside the wellbore, the application of automatic 3D laser scanning data is often affected by noise interference and occlusion, resulting in the traditional filtering method facing great challenges in accuracy and robustness when extracting wellbore wall point cloud. In order to improve the accuracy and robustness of wellbore deformation monitoring data, this study improves the quick hull algorithm and constructs an efficient filtering method for ancillary facilities by introducing an adaptive distance threshold judgment mechanism. This method first slices the wellbore point cloud data and then extracts the wellbore boundary by using quick hull. Instead, it automatically calculates and adjusts an appropriate threshold according to the local characteristics of the currently processed data (point cloud density, distribution) to effectively separate the wellbore and ancillary facilities point cloud. Experimental results show that, compared with the circular fitting filtering method based on random sampling consensus algorithm (RANSAC), the proposed method has significant advantages in terms of operation efficiency, and the processing speed is more than doubled. At the same time, in terms of fitting accuracy, the standard deviation of the results obtained by this method is reduced by 16.4% compared with RANSAC method, which shows that it can more effectively retain the real wellbore structure data, and significantly suppress the interference noise introduced by ancillary facilities and other factors, thus reflecting better filtering performance and reliability.

Abstract Artificial intelligence (AI) in healthcare holds transformative potential but risks exacerbating existing health disparities if inclusivity is not explicitly accounted for. This study addresses the disconnected discussions on inclusive medical AI by developing a comprehensive framework, PREFER-IT. This framework is based on the outcomes of a five-day transdisciplinary co-creation workshop that involved 37 experts from diverse backgrounds, including healthcare, ethics, law, social sciences, AI, and patient advocacy. For this workshop, we used design thinking and participatory methodologies to develop a framework for realising inclusive medical AI. We identified three key challenges for realising inclusive medical AI: integrating the lived experiences and stakeholder voices across the AI lifecycle, designing data collection practices that promote fairness and prevent inequalities, and fostering regulatory frameworks to uphold human rights and promote inclusivity. The analysis of participants’ perspectives informed the development of eight key thematic clusters of PREFER-IT: Participatory and co-design approaches (P), Representative and diverse data (R), Education and digital literacy (E), Fairness (F), Ethical and legal accountability (E), Real-world validation and feedback (R), Inclusive communication (I), and Technical interoperability (T). These elements were mapped across structural layers of AI (humans, data, system, process, and governance) and the AI lifecycle to guide inclusive design, development, validation, implementation, monitoring, and governance. This framework fosters stakeholder engagement and systemic change, positioning inclusion as a guiding principle in practice. PREFER-IT offers a practical and conceptual contribution for how to include ethical, legal and societal aspects when aiming to foster responsible and inclusive AI in healthcare. Author Summary Artificial intelligence (AI) is being used more and more in healthcare to improve diagnosis, treatment, and personalised care. However, if not designed carefully, these technologies can unintentionally increase existing inequalities and exclude certain groups from their benefits. In our study, we brought together experts from healthcare, ethics, law, social sciences, and patient advocacy to explore how AI in medicine can be made more inclusive. Over five days, we worked together to identify key issues and come up with practical solutions. We focused on three main areas: 1) Ensuring diverse voices are heard during the development of AI tools; 2) Making data collection fair and representative; and 3) Creating regulations that protect human rights. From the discussions of the workshop, we created the PREFER-IT framework, which outlines eight key principles for inclusive AI: P articipatory and co-design approaches R epresentative and diverse data E ducation and digital literacy F airness E thical and legal accountability R eal-world validation and feedback I nclusive communication T echnical interoperability This framework helps guide developers, policymakers, and healthcare professionals in creating AI systems that are not only effective but also fair and respectful of all users. Our work emphasises the importance of involving patients and communities in shaping the future of AI.

To investigate the influence of colloidal clay particle migration on the physical and mechanical properties of loess during cyclic humidification and dehumidification, tests such as apparent morphology, shear strength, scanning electron microscopy, mapping and X-ray fluorescence (XRF) were carried out. They compared and analysed the variations in sample mass parameters, surface crack rates, shear parameters, microstructure characteristics, and particle occurrence and content under different colloidal clay contents and different dry‒wet conditions. The aim was to reveal the internal relationships between cyclic humidification and dehumidification in different environments and structural mechanics, and structural causes. The results show that the mass change parameters and apparent fracture rates of the samples in the open environment increase with increasing dry‒wet frequency and colloidal clay particle content. The overall change parameters and the total change parameters of the closed environment are smaller, fewer cracks, and cracks are only produced after a large amount of accumulation of humidification and dehumidification. Under the same positive pressure conditions, the shear strength of the open environment decreases with an increasing number of dry‒wet cycles, whereas the shear strength of the closed environment increases with an increasing number of cycles. The microscopy images revealed that the image brightness of sodium (Na) in the open environment gradually darkened, whereas the image brightness of high-frequency dry and wet loess in the closed environment was brighter and the element distribution was more uniform. Further analysis of SEM-EDX mapping and XRF images revealed that the difference in particle transport in different environments is the key factor leading to the difference in the structural and strength characteristics of loess. In the open environment, cyclic humidification and dehumidification greatly reduced the effective cementation among the skeleton particles, and the structure of the skeleton was significantly reduced. As a result of particle migration in a closed environment, on the one hand, the ‘viscous bridge’ effect of cementation between the skeleton particles of loess is strengthened; on the other hand, the surface of the skeleton mineral crystals that constitute the main part of the pore wall is smoother, and the viscous force of seepage through the pores is weakened, thus resulting in a strong structure.

The ceramic tiles industry is one of the largest industries globally and is expected to grow remarkably in the current decade. The ceramic tiles are a major source of decorations for building facades, interior design, etc. The manufacturing process of ceramic tiles is heavily dependent on the natural raw materials; hence a lot of research has been carried out to integrate various industrial wastes and coal ashes in its manufacturing. In this research project, an effort has been made to optimize the manufacturing process of ceramic tiles developed with the incorporation of industrial wastes and coal ashes, by predicting the flexure strength of ceramic tiles using different machine learning models. In this study, a comprehensive dataset was developed by combining the findings of 13 journal articles. The dataset consisted of 7 input variables, while the flexure strength of ceramic tiles was taken as an output variable. Three machine learning models were developed in this research project and their performance comparison was made. The developed machine learning models included artificial neural networks ANN, K-Nearest Neighbors and Extreme Gradient Boosting. In the later part of studies, K-Fold Cross Validation technique was applied to the best performing machine learning model to cross check the performance of the developed model. It was revealed that the ANN model performed the best among all the three machine learning models with an R2 value of 0.9795, which was authenticated by K-Fold cross Validation technique.

The Mafube processing plant encountered significant discrepancies between the actual yield of middling product compared to the yield predicted by the geological model. This triggered a review of the performance of the coarse-fines circuit given its significant contribution to the middlings product. This paper presents the results of performance measurements conducted on the coarse-fines circuit performance to understand existing performance and improvements required. The Mafube coarse-fines circuit beneficiates coarse-fines within the particle size range from -0.50mm to 0.150mm and consist of the low-cut density three product (LC3) spirals and the teetered bed separator (TBS). The LC3 spirals operate as the primary beneficiation circuit of the coarse-fines whereas the TBS serves as a secondary upgrade stage, to achieve a target overall coarse-fines calorific value (CV) upgrade of 5 MJ/kg. The LC3 spirals results indicated an efficient performance in the primary beneficiation circuit with a CV upgrade of 3.2 MJ/kg with an ash yield of 51%. The secondary beneficiation stage in TBS indicated a further coarse-fines upgrade CV of 1.55 MJ/kg, thus yielding an overall coarse-fines CV upgrade of 4.82 MJ/kg from the Mafube coarse-fines circuit. Notably, the TBS discard stream showed a CV quality of 21.90 MJ/kg which still meets the Mafube middlings product specifications, thus indicating additional potential to capture the Mafube coarse-fines circuit's discard stream as middlings product. A further feasibility study must be conducted to evaluate the potential gains of capturing the Mafube coarse-fines product for the RB2 export stream and the coarse-fines discard for the RB4 middlings stream.

The coal preparation industry faces increasing challenges, marked by escalating capital and operational costs since 2019, alongside stricter environmental, social, and governance regulations. To navigate these obstacles, there is a growing emphasis on implementing advanced processing and blending strategies, necessitating the use of data science and analytics. Operations are increasingly investing in dedicated remote operations centres and individual site analytics to optimize short-term and long-term operational efficiency, enhancing profitability across the entire mine site. This evolution includes the development of geometallurgical modelling software designed to maximize returns from coal resources by analysing resource databases and financial models to assess various processing strategies. Enhanced capabilities using R Shiny coding and Power BI visualizations enable swift evaluations of infrastructure investments and production strategies over the mine's lifespan. The tool's application extends to individual equipment strategies, focusing on efficiency metrics to determine optimal feed rates and necessary upgrades. A series of case studies illustrate the practical use of this technology for scheduling capital improvements and optimizing supply chain dynamics for ROM and product stockpiling across multi-seam operations, ensuring resource management aligns with financial and operational goals.