Predicting CO2 trapping in deep saline aquifers using optimized long short-term memory

CO2 trapping dynamics in tight sandstone: Insights into trapping mechanisms in Mae Moh's reservoir

The reliance on fossil fuels is a major contributor to increased anthropogenic CO2 emissions, driving global challenges such as climate change through the greenhouse effect. Carbon capture and storage (CCS) is a promising interdisciplinary technology aimed at mitigating these emissions by securely sequestering gigatons of CO2. This study focuses on the feasibility of storing point-source CO2 emissions in saline formations, with a particular emphasis on the Mae Moh coal-fired power plant in Lampang, Thailand, which is located near its associated coal mine. The region presents challenges due to tight sandstone reservoirs buried over 2000 m deep. With reservoir simulation, this study evaluates the impact of various factors on CO2 containment and trapping in these geological settings. Results show that elevated temperatures decrease structural trapping of 43.0%-28.9% and increase solubility trapping of 28.55%-46.5%, at 40 °C and 80 °C respectively. Hysteresis is found to enhance residual trapping by immobilizing up to 31.1% of CO2 within pore spaces at 0.5. Permeability heterogeneity has a minimal impact on overall trapping efficiency due to the less heterogeneity of the tight sandstone. However, the kV/kH ratio significantly influences vertical CO2 migration which resulted in residual trapping at its highest at the ratio of 0.1, while lower ratios support lateral dispersion. Moderate rock compressibility values are identified as optimal for structural and residual trapping, while extreme compressibility enhances solubility trapping by up to 30%. These findings emphasize the complexity of CO2 trapping mechanisms in tight sandstone formations, emphasizing the need for careful consideration of key factors in CCS projects.

Predicting mine water inflow volumes using a decomposition-optimization algorithm-machine learning approach

Disasters caused by mine water inflows significantly threaten the safety of coal mining operations. Deep mining complicates the acquisition of hydrogeological parameters, the mechanics of water inrush, and the prediction of sudden changes in mine water inflow. Traditional models and singular machine learning approaches often fail to accurately forecast abrupt shifts in mine water inflows. This study introduces a novel coupled decomposition-optimization-deep learning model that integrates Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Northern Goshawk Optimization (NGO), and Long Short-Term Memory (LSTM) networks. We evaluate three types of mine water inflow forecasting methods: a singular time series prediction model, a decomposition-prediction coupled model, and a decomposition-optimization-prediction coupled model, assessing their ability to capture sudden changes in data trends and their prediction accuracy. Results show that the singular prediction model is optimal with a sliding input step of 3 and a maximum of 400 epochs. Compared to the CEEMDAN-LSTM model, the CEEMDAN-NGO-LSTM model demonstrates superior performance in predicting local extreme shifts in mine water inflow volumes. Specifically, the CEEMDAN-NGO-LSTM model achieves scores of 96.578 in MAE, 1.471% in MAPE, 122.143 in RMSE, and 0.958 in NSE, representing average performance improvements of 44.950% and 19.400% over the LSTM model and CEEMDAN-LSTM model, respectively. Additionally, this model provides the most accurate predictions of mine water inflow volumes over the next five days. Therefore, the decomposition-optimization-prediction coupled model presents a novel technical solution for the safety monitoring of smart mines, offering significant theoretical and practical value for ensuring safe mining operations.

Toward Estimating CO2 Solubility in Pure Water and Brine Using Cascade Forward Neural Network and Generalized Regression Neural Network: Application to CO2 Dissolution Trapping in Saline Aquifers

Predicting carbon dioxide (CO2) solubility in water and brine is crucial for understanding carbon capture and storage (CCS) processes. Accurate solubility predictions inform the feasibility and effectiveness of CO2 dissolution trapping, a key mechanism in carbon sequestration in saline aquifers. In this work, a comprehensive data set comprising 1278 experimental solubility data points for CO2-brine systems was assembled, encompassing diverse operating conditions. These data encompassed brines containing six different salts: NaCl, KCl, NaHCO3, CaCl2, MgCl2, and Na2SO4. Also, this databank encompassed temperature spanning from 273.15 to 453.15 K and a pressure range spanning 0.06-100 MPa. To model this solubility databank, cascade forward neural network (CFNN) and generalized regression neural network (GRNN) were employed. Furthermore, three optimization algorithms, namely, Bayesian Regularization (BR), Broyden-Fletcher-Goldfarb-Shanno (BFGS) quasi-Newton, and Levenberg-Marquardt (LM), were applied to enhance the performance of the CFNN models. The CFNN-LM model showcased average absolute percent relative error (AAPRE) values of 5.37% for the overall data set, 5.26% for the training subset, and 5.85% for the testing subset. Overall, the CFNN-LM model stands out as the most accurate among the models crafted in this study, boasting the highest overall R2 value of 0.9949 among the other models. Based on sensitivity analysis, pressure exerts the most significant influence and stands as the sole parameter with a positive impact on CO2 solubility in brine. Conversely, temperature and the concentration of all six salts considered in the model exhibited a negative impact. All salts exert a negative impact on CO2 solubility due to their salting-out effect, with varying degrees of influence. The salting-out effects of the salts can be ranked as follows: from the most pronounced to the least: MgCl2 > CaCl2 > NaCl > KCl > NaHCO3 > Na2SO4. By employing the leverage approach, only a few instances of potential suspected and out-of-leverage data were found. The relatively low count of identified potential suspected and out-of-leverage data, given the expansive solubility database, underscores the reliability and accuracy of both the data set and the CFNN-LM model's performance in this survey.

Predicting adsorption of organic compounds onto graphene and black phosphorus by molecular dynamics and machine learning

With an increase in production and application of various engineering nanomaterials (ENMs), they will inevitably be released into the environment. Adsorption of various organic chemicals onto ENMs will impact on their environmental behavior and toxicology. It is unrealistic to experimentally determine adsorption equilibrium constants (K) for the vast number of organics and ENMs due to high cost in expenditure and time. Herein, appropriate molecular dynamics (MD) methods were evaluated and selected by comparing experimental K values of seven organics adsorbed onto graphene with the MD-calculated ones. Machine learning (ML) models on K of organics adsorption onto graphene and black phosphorus nanomaterials were constructed based on a benchmark data set from the MD simulations. Lasso models based on Mordred descriptors outperformed ML models built by support vector machine, random forest, k-nearest neighbor, and gradient boosting decision tree, in terms of cross-validation coefficients (Q2 > 0.90). The Lasso models also outperformed conventional poly-parameter linear free energy relationship models for predicting logK. Compared with previous models, the Lasso models considered more compounds with different functional groups and thus have broader applicability domains. This study provides a promising way to fill the data gap in logK for chemicals adsorbed onto the ENMs.

Exploring influential factors of CO2 emissions in China's cities using machine learning techniques

The precise and exhaustive discernment of factors influencing CO₂ emissions underpins the advancement toward sustainable, low-carbon development. Although numerous studies have probed the correlation between predetermined proxy variables and carbon emissions, methodological constraints have often led to an inability to effectively discern carbon emission determinants among numerous potential variables or unravel complex, non-linear relationships, and interaction effects. To redress these research gaps, this research utilized machine learning models to correlate urban CO₂ emissions with socioeconomic indicators. The model outputs were then visualized and interpreted using explainable methods. The findings indicated that the model successfully identified a comprehensive array of dominant influences on urban CO₂ emissions, principally associated with local fiscal policies, land use, energy consumption, industrial development, and urban transportation. The findings further revealed a complex non-linear association between these factors and urban CO₂ emissions; however, the majority of these variables displayed a prevalent propensity to intensify carbon emissions in correspondence with an increase in sample value. Additionally, these factors exhibited a complex interactive influence on urban CO₂ emissions, with distinct pairings producing a suppressive effect exclusively at specific combination of sample values. Consequently, this research posited that a robust correlation between urban socioeconomic development and CO₂ emissions in China remains to be established. Given the varied impacts of these influencing factors across different cities, a differentiated approach to development should be adopted when charting low-carbon trajectories.

Child-Sum EATree-LSTMs: enhanced attentive Child-Sum Tree-LSTMs for biomedical event extraction

BACKGROUND

Tree-structured neural networks have shown promise in extracting lexical representations of sentence syntactic structures, particularly in the detection of event triggers using recursive neural networks.

METHODS

In this study, we introduce an attention mechanism into Child-Sum Tree-LSTMs for the detection of biomedical event triggers. We incorporate previous researches on assigning attention weights to adjacent nodes and integrate this mechanism into Child-Sum Tree-LSTMs to improve the detection of event trigger words. We also address a limitation of shallow syntactic dependencies in Child-Sum Tree-LSTMs by integrating deep syntactic dependencies to enhance the effect of the attention mechanism.

RESULTS

Our proposed model, which integrates an enhanced attention mechanism into Tree-LSTM, shows the best performance for the MLEE and BioNLP'09 datasets. Moreover, our model outperforms almost all complex event categories for the BioNLP'09/11/13 test set.

CONCLUSION

We evaluate the performance of our proposed model with the MLEE and BioNLP datasets and demonstrate the advantage of an enhanced attention mechanism in detecting biomedical event trigger words.

A Comprehensive Survey on Aquila Optimizer

Aquila Optimizer (AO) is a well-known nature-inspired optimization algorithm (NIOA) that was created in 2021 based on the prey grabbing behavior of Aquila. AO is a population-based NIOA that has demonstrated its effectiveness in the field of complex and nonlinear optimization in a short period of time. As a result, the purpose of this study is to provide an updated survey on the topic. This survey accurately reports on the designed enhanced AO variations and their applications. In order to properly assess AO, a rigorous comparison between AO and its peer NIOAs is conducted over mathematical benchmark functions. The experimental results show the AO provides competitive outcomes.

Boosted Reptile Search Algorithm for Engineering and Optimization Problems

Recently, various metaheuristic (MH) optimization algorithms have been presented and applied to solve complex engineering and optimization problems. One main category of MH algorithms is the naturally inspired swarm intelligence (SI) algorithms. SI methods have shown great performance on different problems. However, individual MH and SI methods face some shortcomings, such as trapping at local optima. To solve this issue, hybrid SI methods can perform better than individual ones. In this study, we developed a boosted version of the reptile search algorithm (RSA) to be employed for different complex problems, such as intrusion detection systems (IDSs) in cloud–IoT environments, as well as different optimization and engineering problems. This modification was performed by employing the operators of the red fox algorithm (RFO) and triangular mutation operator (TMO). The aim of using the RFO was to boost the exploration of the RSA, whereas the TMO was used for enhancing the exploitation stage of the RSA. To assess the developed approach, called RSRFT, a set of six constrained engineering benchmarks was used. The experimental results illustrated the ability of RSRFT to find the solution to those tested engineering problems. In addition, it outperformed the other well-known optimization techniques that have been used to handle these problems.