Deformation response of roof in solid backfilling coal mining based on viscoelastic properties of waste gangue

Solid backfill mining(SBM)is a form of green mining,the core of which is to control and minimize the deformation and movement of strata above longwall coal mines.Establishing a mechanical model that can reliably describe roof deformation by considering the viscoelastic properties of waste gangue is important as it assists in improving mine designs and reducing the environmental impact on the surface.In this paper,the time-dependent deformation characteristics of gangue under different stress levels were obtained by using lateral confinement compression,that reliably represents the compaction of goaf.The viscoelastic foundation model for gangue mechanical response is different from the traditionally used elastic foundation model,as it considers the time factor and viscoelasticity.A mechanical model using a thin plate on a fractional viscoelastic foundation was established,and the roof deflection,bending moment,time-dependent,viscous and other characteristics of SBM were included and analyzed.Compared with the existing elastic foundation model,the proposed fractional order viscoelastic foundation model has higher accuracy with laboratory data.The plate deflection increases by 50.9%and the bending moment increases by 37.9%after 100 days,which the elastic model would not have been able to predict.

Predicting the severity of traffic accidents on mountain freeways with dynamic traffic and weather data

Traffic accident severity prediction is essential for dynamic traffic safety management.To explore the factors influencing the severity of traffic accidents on mountain freeways and to predict the severity of traffic accidents,four models based on machine learning algorithms are constructed using support vector machine(SVM),decision tree classifier(DTC),Ada_SVM and Ada_DTC.In addition,random forest(RF)is used to calculate the importance degree of variables and the accident severity influences with high importance levels form the RF dataset.The results show that rainfall intensity,collision type,number of vehicles involved in the accident and toad section type are important variables influencing accident severity.The RF feature selection method improves the classification performance of four machine leaming algorithms,resulting in a 9.3%,5.5%,7.2% and 3.6% improvement in prediction accuracy for SVM,DTC,Ada_SVM and Ada_DTC,respectively.The combination of the Ada_SVM integrated algorithm and RF feature selection method has the best prediction performance,and it achieves 78.9% and 88.4% prediction precision and accuracy,respectively.

Steady-state solution of traffic flow on a simple road network

The steady-state traffic flow on a simply circled road network is analytically studied using the Lighthill-Witham-Richards(LWR)model.The network is typically composed of a diverging and a merging junction together with three connected road sections.At the diverging junction,traffic flow is assigned to satisfy the user-equilibrium condition.At the merging junction,queuing or shock structures due to the bottleneck effect is taken into account.We indicate that the solution depends on the total number of vehicles on the road network,and that the bottleneck effect gives rise to not only capacity drop but inefficient utilization of the two road sections upstream the merging junction.To further validate the derived steady-state solution,a first-order Godunov scheme of the LWR model is adopted for simulation of traffic flow in each road section and the demand-supply concept is applied to provide boundary values at the junctions for the scheme.By varying the total number of vehicles from zero to the maximum,the simulation shows that a randomly distributed state of traffic flow is bound to evolve into a steady state,which is completely in agreement with the analytical solution.