Height prediction of water-flowing fracture zone with a geneticalgorithm support-vector-machine method

Prediction of the height of a water-flowing fracture zone(WFFZ)is the foundation for evaluating water bursting conditions on roof coal.By taking the Binchang mining area as the study area and conducting an in-depth study of the influence of coal seam thickness,burial depth,working face length,and roof category on the height of a WFFZ,we proposed that the proportion of hard rock in different roof ranges should be used to characterise the influence of roof category on WFFZ height.Based on data of WFFZ height and its influence index obtained from field observations,a prediction model is established for WFFZ height using a combination of a genetic algorithm and a support-vector machine.The reliability and superiority of the prediction model were verified by a comparative study and an engineering application.The results show that the main factors affecting WFFZ height in the study area are coal seam thickness,burial depth,working face length,and roof category.Compared with multiple-linear-regression and back-propagation neural-network approaches,the height-prediction model of the WFFZ based on a genetic-algorithm support-vector-machine method has higher training and prediction accuracy and is more suitable for WFFZ prediction in the mining area.

A method for predicting the water-flowing fractured zone height based on an improved key stratum theory

In the process of using the original key stratum theory to predict the height of a water-flowing fractured zone(WFZ),the influence of rock strata outside the calculation range on the rock strata within the calculation range as well as the fact that the shape of the overburden deformation area will change with the excavation length are ignored.In this paper,an improved key stratum theory(IKS theory)was proposed by fixing these two shortcomings.Then,a WFZ height prediction method based on IKS theory was established and applied.First,the range of overburden involved in the analysis was determined according to the tensile stress distribution range above the goaf.Second,the key stratum in the overburden involved in the analysis was identified through IKS theory.Finally,the tendency of the WFZ to develop upward was determined by judging whether or not the identified key stratum will break.The proposed method was applied and verified in a mining case study,and the reasons for the differences in the development patterns between the WFZs in coalfields in Northwest and East China were also fully explained by this method.

Predicting the height of water-flow fractured zone during coal mining under the Xiaolangdi Reservoir

It is very important to determine the extent of the fractured zone through which water can flow before coal mining under the water bodies.This paper deals with methods to obtain information about overburden rock failure and the development of the fractured zone while coal mining in Xin'an Coal Mine.The risk of water inrush in this mine is great because 40%of the mining area is under the Xiaolangdi reservoir.Numerical simulations combined with geophysical methods were used in this paper to obtain the development law of the fractured zone under different mining conditions.The comprehensive geophysical method described in this paper has been demonstrated to accurately predict the height of the water-flow fractured zone.Results from the new model, which created from the results of numerical simulations and field measurements,were successfully used for making decisions in the Xin'an Coal Mine when mining under the Xiaolangdi Reservoir.Industrial scale experiments at the number 11201,14141 and 14191 working faces were safely carried out.These achievements provide a successful background for the evaluation and application of coal mining under large reservoirs.

Mechanical mechanism of overlying strata breaking and development of fractured zone during close-distance coal seam group mining

This study mainly investigates the mechanical mechanism of overlying strata breaking and the development of fractured zones during close-distance coal seam group mining in the Gaojialiang coal mine.First,a mechanical model for the second"activation"of broken overlying strata is established,and the related mechanical"activation"conditions are obtained.A recursive formula for calculating the separation distance of overlying strata is deduced.Second,a height determining method for predicting the height of fractured zones during close-distance coal seam group mining is proposed based on two values,namely,the separation distance and ultimate subsidence value of overlying strata.This method is applied to calculate the fractured zone heights in nos.20107 and 20307 mining faces.The calculated results are almost equal to the field observation results.Third,a modified formula for calculating the height of a waterflowing fractured zone is proposed.A comparison of the calculated and observed results shows that the errors are small.The height determining method and modified formula not only build a theoretical foundation for water conservation mining at the Gaojialiang coal mine,but also provide a reference for estimating the height of water-flowing fractured zones in other coal mines with similar conditions.

Solution to reinforcement learning problems with artificial potential field

A novel method was designed to solve reinforcement learning problems with artificial potential field.Firstly a reinforcement learning problem was transferred to a path planning problem by using artificial potential field(APF),which was a very appropriate method to model a reinforcement learning problem.Secondly,a new APF algorithm was proposed to overcome the local minimum problem in the potential field methods with a virtual water-flow concept.The performance of this new method was tested by a gridworld problem named as key and door maze.The experimental results show that within 45 trials,good and deterministic policies are found in almost all simulations.In comparison with WIERING's HQ-learning system which needs 20 000 trials for stable solution,the proposed new method can obtain optimal and stable policy far more quickly than HQ-learning.Therefore,the new method is simple and effective to give an optimal solution to the reinforcement learning problem.

Prediction of Turbulent Flow Structure in a Fully Developed Rib-Roughened Narrow Rectangular Channel

Fully developed turbulent water-flow structure over one-side repeated-ribs in narrow two-dimensional rectangular channels was investigated experimentally by Particle Image Velocimetry （PIV） and analytically by the standard k-ε and nonlinear k-ε turbulent models. Two rib-pitch to height ratios （p/k） of 10 and 20 were investigated while the rib height was held constant at 4 mm. The rib height-to-channel equivalent diameter ratio （k/De） was 0.1. The streamwise mean velocity and turbulent kinetic energy distributions at six selected axial stations from the center rib for the two Reynolds number （Re） of 7,000 and 20,000 were obtained and compared with the predicted one. The performance ability in predicting separating and reattaching turbulent water-flow between the standard K-ε and nonlinear k-ε models had yielded no clear conclusion. A large-scale turbulent eddy was generated by the rib promoter and then propagated into the mainstream flow, which led to the deformation of the velocity profile. The turbulent kinetic energy was increased about two times higher at p/k = 20 than that at p/k =10 under the two Reynolds numbers. The effect of the p/k value and the Reynolds number （Re） on reattachment length （XR） was investigated and showed that the p/k and Re had no significant effect on the reattachment length beyond a critical value of Re = 15,000 where XR was found to be approximately 4 times of the rib height under water-flow condition.