A novel method for dry density forecasting of high-speed railway graded aggregate fillers based on the PSO-ANN model

Results in Engineering, Год журнала: 2023, Номер 19, С. 101361 - 101361

Опубликована: Авг. 18, 2023

To address the lack of intelligent prediction research on dry density, this paper proposes a machine learning-based method for compaction quality high-speed railway graded aggregate (HRGA) fillers. Firstly, to reveal main control characteristics between HRGA fillers performance and 80 sets vibration tests were carried out. The Grey Relation Analysis (GRA) algorithm was used quantify degree influence properties particle gradation, shape fragmentation, density. Then, stable database density established through 300 lab vibratory tests. Finally, three Hybrid Machine Learning (ML) models (PSO-ANN, PSO-SVR, PSO-RF) fit nonlinear relationship mian results show that gradation parameter b, maximum size dmax, m are strongly correlated features, with mean value correlation reaching 0.77. These features considered factors affecting as input subsequent ML model. Moreover, PSO-ANN model is shown have higher accuracy R2 = 0.9634 in test set lower uncertainty U95= 0.1413, Tstat= 0.533, SI 0.2306 predictions. finding can provide significant reference compaction.

Язык: Английский

---

Macro- and micro- deterioration mechanism of high-speed railway graded gravel filler during vibratory compaction

Construction and Building Materials, Год журнала: 2023, Номер 409, С. 134043 - 134043

Опубликована: Ноя. 3, 2023

Язык: Английский

---

A Framework for Determining the Optimal Vibratory Frequency of Graded Gravel Fillers Using Hammering Modal Approach and ANN

Sensors, Год журнала: 2024, Номер 24(2), С. 689 - 689

Опубликована: Янв. 22, 2024

To address the uncertainty of optimal vibratory frequency fov high-speed railway graded gravel (HRGG) and achieve high-precision prediction fov, following research was conducted. Firstly, commencing with compaction experiments hammering modal analysis method, resonance f0 HRGG fillers, varying in compactness K, initially determined. The correlation between revealed through conducted at different frequencies. This established based on physical–mechanical properties encompassing maximum dry density ρdmax, stiffness Krd, bearing capacity coefficient K20. Secondly, gray relational algorithm used to determine key feature influencing quantified relationship filler fov. Finally, features were as input parameters establish artificial neural network model (ANN-PM) for predictive performance ANN-PM evaluated from ablation study, accuracy, error. results showed that K20 all obtained states when set gradation fillers. Furthermore, it found determined be particle diameter dmax, b m, flat elongated particles coarse aggregate Qe, Los Angeles abrasion LAA. Among them, influence dmax most significant. On training testing sets, goodness-of-fit R2 exceeded 0.95, errors small, which indicated accuracy predictions relatively high. In addition, clear exhibited excellent robust performance. provide a novel method determining subgrade fillers theoretical guidance intelligent construction subgrades.

Язык: Английский

---

Microstructure evolution of coarse-grained soil fillers during subgrade compaction-operation period based on CT technology

Transportation Geotechnics, Год журнала: 2024, Номер 47, С. 101279 - 101279

Опубликована: Май 29, 2024

Язык: Английский

---

ABAQUS-based research on the parameters of highway subgrade vibratory compaction and vibration wave propagation laws

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

Опубликована: Ноя. 28, 2024

Vibratory rollers are generally used in the process of highway subgrade compaction. In paper, vibratory roller—subgrade finite element model was established to simulate field construction by using ABAQUS. We Hilbert–Huang Transform analyze compaction test from time–frequency domain. By changing parameters roller and filler, comprehensive influence such as elastic modulus rolling speed, excitation force, vibration frequency thickness filler on quality investigated. studied propagation pattern waves three-dimensional space. The study shows that signals different bands Hilbert spectrum represent degrees fillers zones. peak acceleration decreases with increase horizontal vertical distance, but there is an at boundary field. There optimal combination filler. propagate form ellipsoid three dimensions, amplitude distance.

Язык: Английский

---

Investigation on the Confined Breakage Characteristics of Calcareous Sand in the South China Sea Integrated Using Relative Breakage Ratio and Fractal Dimension

Sustainability, Год журнала: 2024, Номер 16(5), С. 2190 - 2190

Опубликована: Март 6, 2024

Calcareous sand, ubiquitous in the geotechnical makeup of South China Sea, exhibits both compressibility and vulnerability to fragmentation when subjected external loading, spanning a spectrum from typical extreme conditions. This investigation aims quantitatively assess compression particle breakage characteristics calcareous sand under varied parameters, including relative density, saturation, applied loads, loading paths, specifically focusing on sustainable methodologies. Through series confined tests, this evaluation employed ratio fractal dimension as key evaluative metrics. The results indicated that employing integrated approach offered more comprehensive understanding breakdown mechanisms than relying singular index. Furthermore, an increase density can induce transition contact behavior, shifting point-to-point interactions face-to-face contact, thereby reducing inter-particle stress minimizing grain breakage, particularly loads below 200 kPa. Increasing exacerbated with finer particles predominantly initiating process. During reloading, pore ratios across various load levels surpass those observed during initial except at 1600 kPa, where decline was noted, coinciding water extrusion onset new fracturing. lubricating effect reduces friction, enhancing concentration edges localized increasing presence without significantly altering overall structure. Notably, influence pressure is evident reloading phase. These findings contribute refined theoretical framework for predicting coastal erosion risks devising effective environmental protection strategies engineering practices.

Язык: Английский

---

Deterioration micro-mechanism of graded aggregates with different gradations under vibratory compaction using X-CT testing

Measurement, Год журнала: 2024, Номер 235, С. 114938 - 114938

Опубликована: Май 18, 2024

Язык: Английский

---

Research on Safety Risk Prediction Model of High-Speed Railway Based on Chaotic RBF Neural Network

Опубликована: Дек. 16, 2023

Язык: Английский

---