Determination of similar material proportions based on orthogonal experiments and neural network optimization in the goaf area

Frontiers in Earth Science, Journal Year: 2025, Volume and Issue: 13

Published: April 10, 2025

The stability of the surrounding rock in goaf mine is poor, which can easily cause collapse disasters mining area. This paper used orthogonal experiments and multi factor optimization methods to study optimal mixing ratio similar materials for limestone a certain iron Yunnan, introduced new material combination (quartz, cement, gypsum, barite, glycerol), utilized alongside machine learning techniques predictive analysis. Systematic testing samples with 25 different ratios yielded extensive data on various physical mechanical parameters, were then reveal complex interactions among influencing factors.The cement-gypsum-ratio significantly influenced uniaxial compressive strength elastic modulus, while barite content density, glycerol impacted internal friction angle. Furthermore, this proposed novel learning-based prediction model that utilizes PSO-BP neural network regress predict experimental data. Compared traditional BP network, results show has higher correlation coefficient R 2 than model, root mean square error (RMSE) absolute (MAE) are lower indicating better more stable performance.So accurately materials. effectiveness was verified through practical engineering case studies. provides theoretical foundations technical support analysis goafs, demonstrating significant application value.

Language: Английский

Mechanical testing and numerical simulation analysis of transverse landslides in mountainous pipelines

Journal of Physics Conference Series, Journal Year: 2025, Volume and Issue: 3006(1), P. 012006 - 012006

Published: May 1, 2025

Abstract Mountain pipelines face the threat of landslide disasters in complex geological environments, and this study focuses on mechanical mechanisms landslides mountain pipelines. Through carefully designed indoor tests, stress conditions under action are simulated, key data such as pipeline strain displacement obtained. At same time, advanced numerical simulation methods used to construct a model interaction between bodies, simulate reproduce experimental conditions, expand analysis further explore transmission, deformation coordination mechanism, possible failure modes bodies. The research results indicate that combination experiments simulations can effectively reveal complexity behavior landslides, providing scientific basis for protection safe operation It is great significance improving reliability disasters. These achievements contribute deeper understanding provide safety assessment energy transmission natural gas They strong support technical guidance optimizing design, construction, maintenance strategies, ensuring continuous reliable supply.

Language: Английский

Study of the evolution of characteristic parameters and damage mechanism of sandstone under the synergistic effects of F-T and stepped cyclic loading-unloading via real-time CT scanning and AE monitoring

Measurement, Journal Year: 2025, Volume and Issue: unknown, P. 118078 - 118078

Published: June 1, 2025

Language: Английский