This study aims to establish an integrated sensitivity analysis framework for optimization and design of the dynamic performance of mechanical systems such as tracked vehicles,by combining the direct differentiation m...This study aims to establish an integrated sensitivity analysis framework for optimization and design of the dynamic performance of mechanical systems such as tracked vehicles,by combining the direct differentiation method(DDM)with the linear multibody system transfer matrix method(linear MSTMM).The rigid-flexible coupled multibody system dynamics model of a tracked vehicle is established using the linear MSTMM and validated through the modal test.Building upon the existing DDM-based eigenvalue sensitivity analysis method within the linear MSTMM,the DDM is embedded into it to enable programmable and efficient computation of dynamic response sensitivities for mechanical systems.The proposed approach is used to quantitatively evaluate the sensitivities of both natural vibration characteristics(e.g.,natural frequencies and mode shapes)and transient dynamic responses of the tracked vehicle with respect to system parameters,successfully identifying critical structural parameters.Compared to conventional finite difference methods,the developed methodology eliminates sensitivity to perturbation step sizes.The contributions of this work lie in establishing a unified theoretical foundation and analysis framework for guiding dynamics optimization and design of mechanical systems,and extending the applicability of the linear MSTMM to sensitivity analysis of transient dynamic responses.展开更多
Amphibious vehicles are more prone to attitude instability compared to ships,making it crucial to develop effective methods for monitoring instability risks.However,large inclination events,which can lead to instabili...Amphibious vehicles are more prone to attitude instability compared to ships,making it crucial to develop effective methods for monitoring instability risks.However,large inclination events,which can lead to instability,occur frequently in both experimental and operational data.This infrequency causes events to be overlooked by existing prediction models,which lack the precision to accurately predict inclination attitudes in amphibious vehicles.To address this gap in predicting attitudes near extreme inclination points,this study introduces a novel loss function,termed generalized extreme value loss.Subsequently,a deep learning model for improved waterborne attitude prediction,termed iInformer,was developed using a Transformer-based approach.During the embedding phase,a text prototype is created based on the vehicle’s operation log data is constructed to help the model better understand the vehicle’s operating environment.Data segmentation techniques are used to highlight local data variation features.Furthermore,to mitigate issues related to poor convergence and slow training speeds caused by the extreme value loss function,a teacher forcing mechanism is integrated into the model,enhancing its convergence capabilities.Experimental results validate the effectiveness of the proposed method,demonstrating its ability to handle data imbalance challenges.Specifically,the model achieves over a 60%improvement in root mean square error under extreme value conditions,with significant improvements observed across additional metrics.展开更多
基金supported by the Natural Science Foundation of Jiangsu Province,China(Grant No.BK20241443)the Jiangsu Funding Program for Excellent Postdoctoral Talent(Grant No.2024ZB072)the National Natural Science Foundation of China(Grant No.92266201).
文摘This study aims to establish an integrated sensitivity analysis framework for optimization and design of the dynamic performance of mechanical systems such as tracked vehicles,by combining the direct differentiation method(DDM)with the linear multibody system transfer matrix method(linear MSTMM).The rigid-flexible coupled multibody system dynamics model of a tracked vehicle is established using the linear MSTMM and validated through the modal test.Building upon the existing DDM-based eigenvalue sensitivity analysis method within the linear MSTMM,the DDM is embedded into it to enable programmable and efficient computation of dynamic response sensitivities for mechanical systems.The proposed approach is used to quantitatively evaluate the sensitivities of both natural vibration characteristics(e.g.,natural frequencies and mode shapes)and transient dynamic responses of the tracked vehicle with respect to system parameters,successfully identifying critical structural parameters.Compared to conventional finite difference methods,the developed methodology eliminates sensitivity to perturbation step sizes.The contributions of this work lie in establishing a unified theoretical foundation and analysis framework for guiding dynamics optimization and design of mechanical systems,and extending the applicability of the linear MSTMM to sensitivity analysis of transient dynamic responses.
基金Supported by the National Defense Basic Scientific Research Program of China.
文摘Amphibious vehicles are more prone to attitude instability compared to ships,making it crucial to develop effective methods for monitoring instability risks.However,large inclination events,which can lead to instability,occur frequently in both experimental and operational data.This infrequency causes events to be overlooked by existing prediction models,which lack the precision to accurately predict inclination attitudes in amphibious vehicles.To address this gap in predicting attitudes near extreme inclination points,this study introduces a novel loss function,termed generalized extreme value loss.Subsequently,a deep learning model for improved waterborne attitude prediction,termed iInformer,was developed using a Transformer-based approach.During the embedding phase,a text prototype is created based on the vehicle’s operation log data is constructed to help the model better understand the vehicle’s operating environment.Data segmentation techniques are used to highlight local data variation features.Furthermore,to mitigate issues related to poor convergence and slow training speeds caused by the extreme value loss function,a teacher forcing mechanism is integrated into the model,enhancing its convergence capabilities.Experimental results validate the effectiveness of the proposed method,demonstrating its ability to handle data imbalance challenges.Specifically,the model achieves over a 60%improvement in root mean square error under extreme value conditions,with significant improvements observed across additional metrics.
