操纵舒适性是人机系统研究的重要内容之一。针对操纵舒适性评价的不确定性和模糊性,构建基于贝叶斯的多核回声状态网络(Echo state network,ESN)模型,对站立姿态下的操纵舒适性进行评价。通过实验,收集不同操纵位置对用户的舒适性影响...操纵舒适性是人机系统研究的重要内容之一。针对操纵舒适性评价的不确定性和模糊性,构建基于贝叶斯的多核回声状态网络(Echo state network,ESN)模型,对站立姿态下的操纵舒适性进行评价。通过实验,收集不同操纵位置对用户的舒适性影响。15名被试者参与了本次实验,每个被试者需要完成100个操纵任务,实验过程中被试者的操纵姿势、目标位置、脚底压力、被试人体尺寸和主观舒适性的数据将被记录下来。选取了10%的实验数据对所提出的方法进行了验证,并与BP神经网络预测模型进行比较,结果表明T-S FNN模型具有较小的均方根误差。最后随机选取了10组操纵任务与快速上肢评估方法(RULA)进行比较,结果表明该方法和实际值相关性系数为0.97(p<0.05),与RULA结果的相关性为0.66(p<0.05),说明该方法能够较好地反应真实结果。展开更多
Remanufacturing,as one of the optimal disposals of end-of-life products,can bring tremendous economic and ecological benefits.Remanufacturing process planning is facing an immense challenge due to uncertainties and fu...Remanufacturing,as one of the optimal disposals of end-of-life products,can bring tremendous economic and ecological benefits.Remanufacturing process planning is facing an immense challenge due to uncertainties and fuzziness of recoverable products in damage conditions and remanufacturing quality requirements.Although researchers have studied the influence of uncertainties on remanufacturing process planning,very few of them comprehensively studied the interactions among damage conditions and quality requirements that involve uncertain,fuzzy information.Hence,this challenge in the context of uncertain,fuzzy information is undertaken in this paper,and a method for remanufacturing process planning is presented to maximize remanufacturing efficiency and minimize cost.In particular,the characteristics of uncertainties and fuzziness involved in the remanufacturing processes are explicitly analyzed.An optimization model is then developed to minimize remanufacturing time and cost.The solution is provided through an improved Takagi-Sugeno fuzzy neural network(T-S FNN)method.The effectiveness of the proposed approach is exemplified and elucidated by a case study.Results show that the training speed and accuracy of the improved T-S FNN method are 23.5%and 82.5%higher on average than those of the original method,respectively.展开更多
文摘操纵舒适性是人机系统研究的重要内容之一。针对操纵舒适性评价的不确定性和模糊性,构建基于贝叶斯的多核回声状态网络(Echo state network,ESN)模型,对站立姿态下的操纵舒适性进行评价。通过实验,收集不同操纵位置对用户的舒适性影响。15名被试者参与了本次实验,每个被试者需要完成100个操纵任务,实验过程中被试者的操纵姿势、目标位置、脚底压力、被试人体尺寸和主观舒适性的数据将被记录下来。选取了10%的实验数据对所提出的方法进行了验证,并与BP神经网络预测模型进行比较,结果表明T-S FNN模型具有较小的均方根误差。最后随机选取了10组操纵任务与快速上肢评估方法(RULA)进行比较,结果表明该方法和实际值相关性系数为0.97(p<0.05),与RULA结果的相关性为0.66(p<0.05),说明该方法能够较好地反应真实结果。
基金This work was supported in part by the National Natural Science Foundation of China(Grant No.51975075)the National Major Scientific and Technological Special Project,China(Grant No.2019ZX04005-001)the Chongqing Technology Innovation and Application Program,China(Grant No.cstc2020jscx-msxmX0221).
文摘Remanufacturing,as one of the optimal disposals of end-of-life products,can bring tremendous economic and ecological benefits.Remanufacturing process planning is facing an immense challenge due to uncertainties and fuzziness of recoverable products in damage conditions and remanufacturing quality requirements.Although researchers have studied the influence of uncertainties on remanufacturing process planning,very few of them comprehensively studied the interactions among damage conditions and quality requirements that involve uncertain,fuzzy information.Hence,this challenge in the context of uncertain,fuzzy information is undertaken in this paper,and a method for remanufacturing process planning is presented to maximize remanufacturing efficiency and minimize cost.In particular,the characteristics of uncertainties and fuzziness involved in the remanufacturing processes are explicitly analyzed.An optimization model is then developed to minimize remanufacturing time and cost.The solution is provided through an improved Takagi-Sugeno fuzzy neural network(T-S FNN)method.The effectiveness of the proposed approach is exemplified and elucidated by a case study.Results show that the training speed and accuracy of the improved T-S FNN method are 23.5%and 82.5%higher on average than those of the original method,respectively.
