This paper offers systematic analysis and in-depth research on the surface rust problem of cold-rolled bell-type annealing strip products.The defect characteristics,occurrence rules,generation mechanism,and influencin...This paper offers systematic analysis and in-depth research on the surface rust problem of cold-rolled bell-type annealing strip products.The defect characteristics,occurrence rules,generation mechanism,and influencing factors of surface rust are presented.This research employed a cold-rolled bell-type furnace annealing unit in a coastal steel factory as an example and conducted production process tests,on-site production tests,scanning electron microscopy,energy spectrum analysis,MINITAB statistical analysis,etc.Moreover,six significant influencing factors and their rules were studied:the cooling time of the final cooling table,the storage time of the intermediate storage,the temperature setting of the intermediate storage,the temperature of leveling liquid,the purging pressure of leveling machine,and the dust in the environment.展开更多
Recent years have witnessed a spurt of progress in federated learning,which can coordinate multi-participation model training while protecting the data privacy of participants.However,low communication efficiency is a...Recent years have witnessed a spurt of progress in federated learning,which can coordinate multi-participation model training while protecting the data privacy of participants.However,low communication efficiency is a bottleneck when deploying federated learning to edge computing and IoT devices due to the need to transmit a huge number of parameters during co-training.In this paper,we verify that the outputs of the last hidden layer can record the characteristics of training data.Accordingly,we propose a communication-efficient strategy based on model split and representation aggregate.Specifically,we make the client upload the outputs of the last hidden layer instead of all model parameters when participating in the aggregation,and the server distributes gradients according to the global information to revise local models.Empirical evidence from experiments verifies that our method can complete training by uploading less than one-tenth of model parameters,while preserving the usability of the model.展开更多
文摘This paper offers systematic analysis and in-depth research on the surface rust problem of cold-rolled bell-type annealing strip products.The defect characteristics,occurrence rules,generation mechanism,and influencing factors of surface rust are presented.This research employed a cold-rolled bell-type furnace annealing unit in a coastal steel factory as an example and conducted production process tests,on-site production tests,scanning electron microscopy,energy spectrum analysis,MINITAB statistical analysis,etc.Moreover,six significant influencing factors and their rules were studied:the cooling time of the final cooling table,the storage time of the intermediate storage,the temperature setting of the intermediate storage,the temperature of leveling liquid,the purging pressure of leveling machine,and the dust in the environment.
基金supported by Shenzhen Basic Research (General Project)under Grant No.JCYJ20190806142601687Shenzhen Stable Supporting Program (General Project) under Grant No.GXWD20201230155427003-20200821160539001+1 种基金Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies under Grant No.2022B1212010005Shenzhen Basic Research (Key Project) under Grant No.JCYJ20200109113405927。
文摘Recent years have witnessed a spurt of progress in federated learning,which can coordinate multi-participation model training while protecting the data privacy of participants.However,low communication efficiency is a bottleneck when deploying federated learning to edge computing and IoT devices due to the need to transmit a huge number of parameters during co-training.In this paper,we verify that the outputs of the last hidden layer can record the characteristics of training data.Accordingly,we propose a communication-efficient strategy based on model split and representation aggregate.Specifically,we make the client upload the outputs of the last hidden layer instead of all model parameters when participating in the aggregation,and the server distributes gradients according to the global information to revise local models.Empirical evidence from experiments verifies that our method can complete training by uploading less than one-tenth of model parameters,while preserving the usability of the model.
