Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can h...Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can handle only one will result in prediction bias in reality.Existing disk failure prediction methods simply fuse various models,lacking discussion of training data preparation and learning patterns when facing multiple issues,although the solutions to different issues often conflict with each other.As a result,we first explore the training data preparation for multiple issues via a data partitioning pattern,i.e.,our proposed multi-property data partitioning(MDP).Then,we consider learning with the partitioned data for multiple issues as learning multiple tasks,and introduce the model-agnostic meta-learning(MAML)framework to achieve the learning.Based on these improvements,we propose a novel disk failure prediction model named MDP-MAML.MDP addresses the challenges of uneven partitioning and difficulty in partitioning by time,and MAML addresses the challenge of learning with multiple domains and minor samples for multiple issues.In addition,MDP-MAML can assimilate emerging issues for learning and prediction.On the datasets reported by two real-world data centers,compared to state-of-the-art methods,MDP-MAML can improve the area under the curve(AUC)and false detection rate(FDR)from 0.85 to0.89 and from 0.85 to 0.91,respectively,while reducing false alarm rate(FAR)from 4.88%to 2.85%.展开更多
To investigate the ballistic resistance and failure pattern of aeroengine casing following the impact of disk fragments, and to determine the optimum case structure, the phenomena of a 1/3rd disk fragment impact on si...To investigate the ballistic resistance and failure pattern of aeroengine casing following the impact of disk fragments, and to determine the optimum case structure, the phenomena of a 1/3rd disk fragment impact on single and double-layered thin plate targets were simulated using nonlinear dynamical analysis software MSC.Dytran. Strain rate effect was introduced in a Johnson-Cook (JC) material model for the disk fragment and the plate. Impact modeling was based on the Arbitrary Lagrange-Eulerian method, and simulated using explicit finite element method (FEM). Simulation results showed that the major failure pattern of the plate is shearing and tensile fracture with large plastic deformation. It was also concluded that the ballistic limit velocity increases with the standoff distance when it is beyond a certain value, and that greater resistance is obtained when the front plate has either a proportionately low or high thickness. The impact resistance of a double-layered plate may exceed that of a single plate if the thicknesses and standoff distance of the two plates are set appropriately.展开更多
基金Project supported by the National Natural Science Foundation of China(No.61902135)the Shandong Provincial Natural Science Foundation,China(No.ZR2019LZH003)。
文摘Disk failure prediction methods have been useful in handing a single issue,e.g.,heterogeneous disks,model aging,and minority samples.However,because these issues often exist simultaneously,prediction models that can handle only one will result in prediction bias in reality.Existing disk failure prediction methods simply fuse various models,lacking discussion of training data preparation and learning patterns when facing multiple issues,although the solutions to different issues often conflict with each other.As a result,we first explore the training data preparation for multiple issues via a data partitioning pattern,i.e.,our proposed multi-property data partitioning(MDP).Then,we consider learning with the partitioned data for multiple issues as learning multiple tasks,and introduce the model-agnostic meta-learning(MAML)framework to achieve the learning.Based on these improvements,we propose a novel disk failure prediction model named MDP-MAML.MDP addresses the challenges of uneven partitioning and difficulty in partitioning by time,and MAML addresses the challenge of learning with multiple domains and minor samples for multiple issues.In addition,MDP-MAML can assimilate emerging issues for learning and prediction.On the datasets reported by two real-world data centers,compared to state-of-the-art methods,MDP-MAML can improve the area under the curve(AUC)and false detection rate(FDR)from 0.85 to0.89 and from 0.85 to 0.91,respectively,while reducing false alarm rate(FAR)from 4.88%to 2.85%.
基金Project (No. 1104-03) supported by the Aviation Propulsion Technology Development Program, China
文摘To investigate the ballistic resistance and failure pattern of aeroengine casing following the impact of disk fragments, and to determine the optimum case structure, the phenomena of a 1/3rd disk fragment impact on single and double-layered thin plate targets were simulated using nonlinear dynamical analysis software MSC.Dytran. Strain rate effect was introduced in a Johnson-Cook (JC) material model for the disk fragment and the plate. Impact modeling was based on the Arbitrary Lagrange-Eulerian method, and simulated using explicit finite element method (FEM). Simulation results showed that the major failure pattern of the plate is shearing and tensile fracture with large plastic deformation. It was also concluded that the ballistic limit velocity increases with the standoff distance when it is beyond a certain value, and that greater resistance is obtained when the front plate has either a proportionately low or high thickness. The impact resistance of a double-layered plate may exceed that of a single plate if the thicknesses and standoff distance of the two plates are set appropriately.
