Roof pre-fracture poses a considerable risk during the re-mining of residual coal above abandoned roadways,threatening the safety of the mining faces.This study employs a Winkler foundation beam model for mechanical a...Roof pre-fracture poses a considerable risk during the re-mining of residual coal above abandoned roadways,threatening the safety of the mining faces.This study employs a Winkler foundation beam model for mechanical analysis of roof structures and adopts a multivariate nonlinear analysis approach to explore the synergistic load-bearing effects within the'coal pillar-support-backfill body'system during the fill and re-mining processes above these roadways.The findings demonstrate that backfill mining significantly reduces stress concentrations in coal pillars and reduces excessive bending moments in roofs near abandoned roadways.The roof deflection equation incorporates three critical factors affecting stability during backfill mining:the width of the coal pillar(L_(3)),the working resistance of the support(q_(z)),and the elastic foundation coefficient of the backfill material(kcÞ.Under single-factor conditions,the impact sequence on roof stability in the coal pillar zone is·k_(c)>L_(3)>q_(z).Further,multivariate nonlinear analysis reveals the interactions within the'coal-support-backfill'structure,indicating that in terms of roof control,the interaction terms are ordered as L_(3)·k_(c)>q_(z)·k_(c)>L_(3)q_(z).Therefore,priority should be given to adjusting the coal pillar width and backfill strength,followed by modifications to the support resistance and backfill strength during the recovery of abandoned roadways.An improved understanding of these interactions will help optimize strategies for the recovery of residual coal through abandoned roadways,thereby enhancing the stability and safety of mining operations under complex geological conditions.展开更多
RISC-V作为一种新兴的开源精简指令集架构,是后摩尔时代处理器技术发展与创新的关键之一.浮点求和与点积运算是数值运算的基础组成部分,在众多领域应用广泛.目前RISC-V架构尚未适配兼具高精度和高效率的求和与点积运算算法,这是因为现...RISC-V作为一种新兴的开源精简指令集架构,是后摩尔时代处理器技术发展与创新的关键之一.浮点求和与点积运算是数值运算的基础组成部分,在众多领域应用广泛.目前RISC-V架构尚未适配兼具高精度和高效率的求和与点积运算算法,这是因为现有优化方案难以良好地平衡运算精度和效率,要么侧重于低精度算法效率,要么通过牺牲效率实现高精度运算.本文利用RVV(RISC-V Vector instruction set extension,RVV)矢量扩展指令,设计并实现了一种基于无误差变换技术的高效、高精度求和与点积算法.首先避免使用规约指令以防止运算精度降低,实现并优化两类运算基于RVV的向量化算法;其次根据算法中的数据依赖关系,对寄存器配置参数进行优化.最后针对算法核心步骤进行汇编优化,增加指令级并行度,提高流水线利用率.实验结果表明,与两类运算操作的原始算法相比,优化后的算法运算效率分别提高了4.4和4.2倍.优化后的算法与多精度库MPFR中的四精度算法有相同精度,但其运算效率明显优于后者,其计算速度与OpenBLAS的双精度计算速度相当.展开更多
基金support from National Natural Science Foundation of China(Grant No.52474142)The National Science Fund for Distinguished Young Scholars(No.51925402),Chinathe China Postdoctoral Science Foundation(Grant No.2021M702049).
文摘Roof pre-fracture poses a considerable risk during the re-mining of residual coal above abandoned roadways,threatening the safety of the mining faces.This study employs a Winkler foundation beam model for mechanical analysis of roof structures and adopts a multivariate nonlinear analysis approach to explore the synergistic load-bearing effects within the'coal pillar-support-backfill body'system during the fill and re-mining processes above these roadways.The findings demonstrate that backfill mining significantly reduces stress concentrations in coal pillars and reduces excessive bending moments in roofs near abandoned roadways.The roof deflection equation incorporates three critical factors affecting stability during backfill mining:the width of the coal pillar(L_(3)),the working resistance of the support(q_(z)),and the elastic foundation coefficient of the backfill material(kcÞ.Under single-factor conditions,the impact sequence on roof stability in the coal pillar zone is·k_(c)>L_(3)>q_(z).Further,multivariate nonlinear analysis reveals the interactions within the'coal-support-backfill'structure,indicating that in terms of roof control,the interaction terms are ordered as L_(3)·k_(c)>q_(z)·k_(c)>L_(3)q_(z).Therefore,priority should be given to adjusting the coal pillar width and backfill strength,followed by modifications to the support resistance and backfill strength during the recovery of abandoned roadways.An improved understanding of these interactions will help optimize strategies for the recovery of residual coal through abandoned roadways,thereby enhancing the stability and safety of mining operations under complex geological conditions.
文摘RISC-V作为一种新兴的开源精简指令集架构,是后摩尔时代处理器技术发展与创新的关键之一.浮点求和与点积运算是数值运算的基础组成部分,在众多领域应用广泛.目前RISC-V架构尚未适配兼具高精度和高效率的求和与点积运算算法,这是因为现有优化方案难以良好地平衡运算精度和效率,要么侧重于低精度算法效率,要么通过牺牲效率实现高精度运算.本文利用RVV(RISC-V Vector instruction set extension,RVV)矢量扩展指令,设计并实现了一种基于无误差变换技术的高效、高精度求和与点积算法.首先避免使用规约指令以防止运算精度降低,实现并优化两类运算基于RVV的向量化算法;其次根据算法中的数据依赖关系,对寄存器配置参数进行优化.最后针对算法核心步骤进行汇编优化,增加指令级并行度,提高流水线利用率.实验结果表明,与两类运算操作的原始算法相比,优化后的算法运算效率分别提高了4.4和4.2倍.优化后的算法与多精度库MPFR中的四精度算法有相同精度,但其运算效率明显优于后者,其计算速度与OpenBLAS的双精度计算速度相当.
