Predicting mine water inrush accidents based on water level anomalies of borehole groups using long short-term memory and isolation forest

Journal of Hydrology, Год журнала: 2022, Номер 616, С. 128813 - 128813

Опубликована: Ноя. 26, 2022

Язык: Английский

---

Unraveling Overlying Rock Fracturing Evolvement for Mining Water Inflow Channel Prediction: A Spatiotemporal Analysis Using ConvLSTM Image Reconstruction

IEEE Transactions on Geoscience and Remote Sensing, Год журнала: 2024, Номер 62, С. 1 - 17

Опубликована: Янв. 1, 2024

Язык: Английский

---

Predicting compressive strength of eco-friendly plastic sand paver blocks using gene expression and artificial intelligence programming

Scientific Reports, Год журнала: 2023, Номер 13(1)

Опубликована: Июль 27, 2023

Plastic sand paver blocks provide a sustainable alternative by using plastic waste and reducing the need for cement. This innovative approach leads to more construction sector promoting environmental preservation. No model or Equation has been devised that can predict compressive strength of these blocks. study utilized gene expression programming (GEP) multi-expression (MEP) develop empirical models forecast (PSPB) comprised plastic, sand, fibre in an effort advance field. The database contains 135 results with seven input parameters. R2 values 0.87 GEP 0.91 MEP reveal relatively significant relationship between predicted actual values. outperformed displaying higher lower statistical evaluations. In addition, sensitivity analysis was conducted, which revealed grain size percentage fibres play essential part strength. It estimated they contributed almost 50% total. outcomes this research have potential promote reuse PSPB building green environments, hence boosting protection economic advantage.

Язык: Английский

---

Sustainable concrete Production: Incorporating recycled wastewater as a green building material

Construction and Building Materials, Год журнала: 2023, Номер 407, С. 133522 - 133522

Опубликована: Сен. 29, 2023

Язык: Английский

---

A Deep Learning-Based Data-Driven Approach for Predicting Mining Water Inrush From Coal Seam Floor Using Microseismic Monitoring Data

IEEE Transactions on Geoscience and Remote Sensing, Год журнала: 2023, Номер 61, С. 1 - 15

Опубликована: Янв. 1, 2023

Micro-seismic monitoring during mining operations generates spatiotemporal data that could indicate strata fractures and deformations leading to water inrush anomalies. However, current prediction methods face challenges from the non-stationarity multi-dimensionality, resulting in low precision effectiveness. This study proposes an innovative data-driven approach for predicting using field 3D micro-seismic data. The couples machine learning deep models analyze events, pre-processed Density-Based Spatial Clustering of Applications with Noise (DBSCAN) Random Sample Consensus (RANSAC) algorithms both denoising risk locating. Weighting periods are analyzed periodic variations event attributes fast Fourier transform (FFT), continuous wavelet (CWT), empirical mode decomposition (EMD), seasonal trend Loess (STL) methods. Anomalies detected long short-time memory (LSTM)+absolute error (AE), isolation forest (iForest) LSTM+iForest models. is conducted a dataset acquired intermittent inflow anomalies Xingdong coal mine China. accurately predicts major incident hours prior its occurrence merging obtained weighting periods, which also used model calibration. Future studies focus on performance evaluation calibration datasets different operations, expanding approach's scope by incorporating other geophysical exploration technologies like electrical further presence movement mines improving safety.

Язык: Английский

---

Assessing the effect of lime-zeolite on geotechnical properties and microstructure of reconstituted clay used as a subgrade soil

Physics and Chemistry of the Earth Parts A/B/C, Год журнала: 2023, Номер 132, С. 103501 - 103501

Опубликована: Ноя. 1, 2023

Язык: Английский

---

Estimation of compressive strength of waste concrete utilizing fly ash/slag in concrete with interpretable approaches: optimization and graphical user interface (GUI)

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Фев. 26, 2024

Abstract Geo-polymer concrete has a significant influence on the environmental condition and thus its use in civil industry leads to decrease carbon dioxide (CO 2 ) emission. However, problems lie with mixed design casting field. This study utilizes supervised artificial-based machine learning algorithms (MLAs) anticipate mechanical characteristic of fly ash/slag-based geopolymer (FASBGPC) by utilizing AdaBoost Bagging MLPNN make an ensemble model 156 data points. The consist GGBS (kg/m 3 ), Alkaline activator Fly ash SP dosage NaOH Molarity, Aggregate Temperature (°C) compressive strength as output parameter. Python programming is utilized Anaconda Navigator using Spyder version 5.0 predict response. Statistical measures validation are done splitting dataset into 80/20 percent K-Fold CV employed check accurateness MAE, RMSE, R . analysis relies errors, tests against external indicators help determine how well models function terms robustness. most important factor measurements examined permutation characteristics. result reveals that ANN outclassed giving maximum enhancement = 0.914 shows least error statistical validations. Shapley GGBS, temperature influential parameter content making FASBGPC. Thus, methods suitable for constructing prediction because their strong reliable performance. Furthermore, graphical user interface (GUI) generated through process training forecasts desired outcome values when corresponding inputs provided. It streamlines provides useful tool applying model's abilities field engineering.

Язык: Английский

---

Comparison of machine learning and statistical approaches to estimate rock tensile strength

Case Studies in Construction Materials, Год журнала: 2024, Номер 20, С. e02890 - e02890

Опубликована: Янв. 21, 2024

Tensile strength is very important in drilling operations. The main objective of this study was to assess petrography, physical, and mechanical properties predict the Brazilian tensile sedimentary rocks by leveraging key parameters such as Schmidt hardness, compressional wave velocity, density, porosity. A diverse array predictive models employed, encompassing simple regression, multivariate linear nonlinear backpropagation artificial neural network, gaussian process classification regression tree, K-nearest neighbor, random forest, support vector regression. Based on thin section analysis X-ray diffraction, samples were identified. sandstone meticulously categorized into two distinct groups: arenite litharenite. Additionally, limestone stratified categories packstone mudstone based texture. highest failure mode frequency under test identified central fracturing. Upon meticulous examination, it discerned that velocity exerted most substantial influence estimates, while density exhibited least impact. Comparing outcomes derived from modeling techniques, unequivocally established model showcased level performance for forecasting strength. This evidenced remarkable coefficient determination 0.99 along with an impressively low root mean square error 0.03.

Язык: Английский

---

Quantitative identification of nitrate and sulfate sources of a multiple land-use area impacted by mine drainage

Journal of Environmental Management, Год журнала: 2022, Номер 325, С. 116551 - 116551

Опубликована: Окт. 22, 2022

Язык: Английский

---

Investigation on mechanical properties of novel natural fiber-epoxy resin hybrid composites for engineering structural applications

Case Studies in Construction Materials, Год журнала: 2023, Номер 19, С. e02356 - e02356

Опубликована: Июль 29, 2023

Polymers and fibers are the main components of cementitious composite materials to reinforce concrete. The synthetic used in concrete composites typically weigh also, they easily subjected thermal degradation. To tackle above issues, researchers developed various natural fiber-based composites. In paper, mechanical characteristics hybrid were examined. using Madar, Gongura, Hibiscus cannabinus fibers. polyester resin was matrix. treated chemically a 5% sodium hydroxide (NaOH) solution. then carefully cleaned with distilled water twice baked for 70 minutes at 60°C. For evaluation tensile, flexural, impact properties composites, specimens made accordance ASTM standards. sample (S3) specimen exhibits tensile strength (TS) approximately 34.720 N/mm2, flexural (FS) 77.957 MPa, modulus (FM) 1548.588 GPa. average absorption is only 2.45%. hardness (IS) samples also superior those several other studied literature. Hence, proposed could be potential material reinforcing provide higher service rate greater durability structures.

Язык: Английский

---