疲劳计算的破坏度法

本文应用概率统计的观点 ,通过对应力状态、载荷块、应力幅分布函数的讨论 ,确定出安全系数、偏差系数和正态分布分位点 。

Simulation analysis on three-dimensional slope failure under different conditions

The failure mechanism of two-dimensional（2D） and three-dimensional（3D） slopes were investigated by using the strength reduction method.An extensive study of 3D effect was conducted with respect to boundary conditions,shear strength and concentrated surcharge load.The results obtained by 2D and 3D analyses were compared and the applicable scope of 2D and 3D method was analyzed.The results of the numerical simulation show that 3D effect is sensitive to the width of slip surface.As for slopes with specific geometry,3D effect is influenced by dimensionless parameter c/（γHtanφ）.For those infinite slopes with local loading,external load has the major impact on failure mode.For those slopes with local loading and geometric constraints,the failure mode is influenced by both factors.With the increase of loading length,boundary condition exerts a more significant impact on the failure mode,and then 2D and 3D stability charts are developed,which provides a rapid and reliable way to calculate 2D and 3D factor of safety without iteration.Finally,a simple and practical calculation procedure based on the study of 3D effect and stability charts is proposed to recognize the right time to apply 2D or 3D method.

Feasibility of stochastic gradient boosting approach for predicting rockburst damage in burst-prone mines

The database of 254 rockburst events was examined for rockburst damage classification using stochastic gradient boosting （SGB） methods. Five potentially relevant indicators including the stress condition factor, the ground support system capacity, the excavation span, the geological structure and the peak particle velocity of rockburst sites were analyzed. The performance of the model was evaluated using a 10 folds cross-validation （CV） procedure with 80%of original data during modeling, and an external testing set （20%） was employed to validate the prediction performance of the SGB model. Two accuracy measures for multi-class problems were employed： classification accuracy rate and Cohen’s Kappa. The accuracy analysis together with Kappa for the rockburst damage dataset reveals that the SGB model for the prediction of rockburst damage is acceptable.

Analysis of floor failure depth by using electric profiling method in longwall gangue backfill mining

In underground mining, floor failure depth accompanying mining phases usually results from changes in the advance abutment pressure in the coal mass, and changes in stress redistribution in the areas that have already been mined. Although a variety of techniques have been applied to determine the failure depth, and a number of studies have provided the evidence for the decreasing of failure depth under backfilling, these methods and interactions have not been unequivocally identified. Based on the premise of one possible relation between the failure depth and filling body, which is that the filling materials （gangue） in the gob area can not only restrain the movement of the overlying strata effectively, but also can help to decrease failure depth of the floor in the coal mine. The failure depth in a specific longwall gangue backfilling mine was measured using the mine electricity profiling method. These electrode cables are arranged in a crossheading order to measure the depth and position of the destroyed floor using the DC method. After this, several different methods were used to interpret the recorded data from the field study for gaining failure depth, and the results were compared to the theoretical calculation values. And finally, the authors analyzed the reasons for failure depth form values recorded not indicating a large decrease trend when compared to the theoretical calculation. In this area, it is found that： ① The results using the mine electricity profiling method turns out to be robust and can be used in predicting floor failure depth, and the horizontal position of the maximum destroyed in working face of longwall backfilling. The maximum destroyed position and failure space of the floor can be identified by using this method. ②There is a time-delay processing between the advance of the working face and the failure of floor strata in the mining processing. ③Additionally, based on the data collected from field measurements, which includes three different test electrode spacing approaches （single, double and triple electrode spacing）, and the theoretical value from theoretical calculations. The premise mentioned above cannot be supported during the specific field test, and the role of the filling body in the mined area cannot decrease the floor failure depth effectively in comparison to the theory predictions. Basically, the failure depths in the two different methods have similar results and it is possible that there will not be a direct correlation between the filling body and failure depth. ④Although the failure depth cannot decrease effectiveness when using gangue backfilling in the field testing, due to gob gangue, filling materials being able to deliver the abutment pressure from the overburden in most cases, once they were compacted and rammed by the overburden pressure, it still can make the fracture of the gob area clog and be further consolidated. In this way, it is assumed that water-bursting accidents can be prevented effectively under backfill mining. For this reason, gangue backfilling may make a significant contribution to safety mining.

A random medium model for simulation of concrete failure

A random medium model is developed to describe damage and failure of concrete.In the first place,to simulate the evolving cracks in a mesoscale,the concrete is randomly discretized as irregular finite elements.Moreover,the cohesive elements are inserted into the adjacency of finite elements as the possible cracking paths.The spatial variation of the material properties is considered using a 2-D random field,and the stochastic harmonic function method is adopted to simulate the sample of the fracture energy random field in the analysis.Then,the simulations of concrete specimens are given to describe the different failure modes of concrete under tension.Finally,based on the simulating results,the probability density distributions of the stress-strain curves are solved by the probability density evolution methods.Thus,the accuracy and efficiency of the proposed model are verified in both the sample level and collection level.