| China is one of the largest coal producers and consumers in the world,and coal will still be China’s main source of energy for a long period of time in the future.With the continuous reduction in shallow coal resources,the depth and intensity of coal mining have increased.Stress and coal gas pressure are also increasing,causing dynamic disasters,such as rock burst,coal and gas outbursts,to become more severe and complex.Effective monitoring to the physical signals emitted from coal and rock during the evolution process of these dynamic disasters is one of the most important way for disaster prediction and prevention.Acoustic emission technology,as an important means of monitoring and warning coal and rock dynamic disasters,has been widely studied by scholars.However,currently,research on the Acoustic emission effects of rock damage under load is less related to water-bearing coal.Hydraulic measures such as water injection into rock mass are widely used in the prevention and control of coal mine rock burst,which changes the original stress state of the rock mass and has a significant impact on the microstructure of coal.Therefore,it is of great significance to fully consider the water rock interaction and study the mechanical behavior and Acoustic emission signal characteristics of water bearing rock under load for the prevention and monitoring of dynamic disasters such as mine rock burst.In addition,this is also related to the monitoring of the stability of water-bearing rock masses in underground engineering and the warning of other related geological disasters.To study the mechanical behaviors and acoustic emission characteristics of water-saturated coal in the deformation and failure process,uniaxial and impact loading experiments on water-saturated coal with different content were conducted and the AE signals were collected synchronously.The mechanical behaviors are studied,and the AEs characteristics are analyzed using fractal theory and critical slowing down theory(CSDT)for early warning.Acoustic emission parameters were used for predicting dilatancy point and to establish an innovative statistical model.and for impact dynamic analysis SHPB was used.The following conclusions are drawn from our research:1.The mechanical properties,AE characteristics,and fractal characteristics of coal with different water content were studied under uniaxial loading.The results showed that the mechanical properties of saturated coal samples,such as uniaxial compressive strength,peak stress,dissipation energy,and elastic modulus,were considerably reduced.A natural coal sample has a peak AE value that is 40%higher than a fully saturated coal sample.According to stress-strain curves,the deformation of coal during loading were found to have five distinct stages:compaction,linear elastic,crack stable propagation,crack accelerating propagation,and post-peak residual stages.Using a novel Grassberger Procaccia(GP)algorithm based on phase-space theory,the AE fractal characteristics of coal samples in different stages were determined.It was observed that AE energy did not show any fractal characteristics in the first or second stages.However,obvious fractal characteristics were observed in the third stage,indicating the initiation of complex microcracks in coal.In the fourth stage,the fractal dimension rapidly declined as the strength reached its limit,suggesting the occurrence of macrocracks.However,the fractal dimensions continued to decrease or increased slightly in the fifth stage,indicating that the coal was beginning to collapse,which could potentially lead to disaster and failure.Based on the findings,it is possible to accurately predict coal and rock dynamic failures by observing the subsequent sudden drop in the correlation dimension of the AE signals in response to different stages of loading.Therefore,fractal analysis of AE time sequences can be a useful precursor tool for assessing the stability and safety of coal and rock.2.Stress waves of the incident,reflected,and transmitted waves were measured using the strain data acquisition system in an impact dynamics experiment.The results of the experiment indicated that reflected and transmitted waves had lower amplitudes than incident waves,which were affected by water content and impact load.These findings were used to generate a stress-strain curve for water-saturated coal,which showed three distinct stages:linear elastic,yielding,and failure.Compaction stage is Regarding the dynamic properties of the coal samples,it was observed that with different water content and impact loading,the elastic modulus and peak stress decrease linear and the linear coefficient(R~2)is more than 0.90.The sample with water content more than 1.5%and impact velocity greater than 8m/s have lower peak than sample subject to 7m/s.These results demonstrated that the water content has a significant influence on the dynamic characteristics of the coal samples under varying loading conditions.3.AE counts were analyzed using critical slowing down theory(CSDT)based on variance and autocorrelation for coal samples with varying water content in order to enhance coal failure prediction.As a result of the study’s findings,it has been concluded that an increase in both variance and autocorrelation coefficient of AE counts may serve as an early warning sign of coal pillar instability.There were similar variations in AE counts and accumulative counts between the samples,but the rise in AE counts differed according to the content of water in the samples.For fully saturated samples,the ratio of early warning point(inflection point)to peak stress and the time needed for AE were more than 50%and the ratio increased with decreasing water content.It was found that window length had a minimal influence on variance and autocorrelation,but it had an impact on the amplitude of fluctuation.There was an inverse relationship between variance amplitude and window length,while there was a direct relationship between autocorrelation and window length.In the variance curves,the precursory characteristics of AE counts appear to be more apparent than other AE parameters.This finding provides an effective guideline for predicting coal pillar instability,particularly in situations where water is a problem.4.By analyzing acoustic emission data,an innovative and unique statistical model of coal damage has been developed under stress applied uniaxially.A model based on this technique was found to be superior to those based on lognormal and Weibull distributions,because it takes into account the compaction stage.Compared to the experimental curve,the piecewise constitutive model exhibits a correlation coefficient of greater than 0.956.Statistical damage constitutive models for coal are compatible with this model.Additionally,the model is capable of precisely forecasting the stress associated with both natural and saturated coal.Water significantly affects the micro fracture mode of coal.Intergranular fractures are more likely to occur in coal with a high-water content,whereas transgranular fractures are less likely to occur in coal with a high water content.In response to water’s action,the cementitious material between coal particles will fracture,resulting in a more uniform micro fracture pattern.The decrease in strength and the increase in deformation can be attributed to this microscopic phenomenon.5.An investigation was conducted to determine how water content affects the dilatancy point of coal under loading in the presence of acoustic emission.AE data were used to predict the dilatancy(crack damage)point index using three computational techniques,including artificial neural networks(ANN),rain forest regression(RFR),and k-Nearest Neighbor(KNN).Based on the performance coefficient(R~2)and root mean square error(RMSE),these technologies were compared.A decrease in elastic deformation and a stable crack growth stage were observed as a result of the effect of water content on different stages of the stress-strain curve.The AE rate at crack initiation points and dilatancy points increases,but as the water content increases,the corresponding stress decreases.The dilatancy point index was derived from the absolute strain energy rate.Dilatancy points were predicted using artificial neural networks,random forests,and k-nearest neighbors.The study’s findings could aid in predicting early rock failure and serve as a precursor indicator of rock failure.The dilatancy point index was predicted using different models,and their evaluation indicators were compared.A neural network was found to be more effective at predicting events than other models based on the results of the studyThe research results have theoretical significance and laid the technical foundation for water-saturated coal,revealing the mechanism of complex disasters in water-saturated coal,and also have application value for promoting the prevention and control of dynamic disasters in water-saturated coal.In addition,research results can provide a new idea and a new method for monitoring and early warning of water-saturated coal and rock dynamic disasters.It is of great theoretical significance and practical value for coal stress monitoring,stability evaluation and for coal and rock dynamic disaster prediction.A total of 101 graphs,12 tables and 265 references are included in the dissertation. |
