Although state of the art in many typical machine learning tasks, deep learning
algorithmsareverycostly interms ofenergyconsumption,duetotheirlargeamount
of required computations and huge model sizes. Because of this, deep learning
applications on battery-constrained wearables have only been possible through
wireless connections with a resourceful cloud. This setup has several drawbacks.
First, there are privacy concerns. Cloud computing requires users to share their raw
data—images, video, locations, speech—with a remote system. Most users are not
willing to do this. Second, the cloud-setup requires users to be connected all the
time, which is unfeasible given current cellular coverage. Furthermore, real-time
applications require low latency connections, which cannot be guaranteed using
the current communication infrastructure. Finally, wireless connections are very
inefficient—requiringtoo much energyper transferredbit for real-time data transfer
on energy-constrained platforms.
標簽：
Embedded_Deep_Learning
Algorithms
上傳時間：
2020-06-10
上傳用戶：shancjb
目前電動汽車主要以鋰電池作為動力來源,為了提高鋰電池的使用時間和安全性,為鋰電池提供安全良好的運行環境,電池管理系統應運而生。BMS主控單元基于S32K144汽車級單片機,通過主從式網絡控制結構能夠對鋰電池的各個參數進行采集與分析。采用擴展卡爾曼濾波對電池的荷電狀態(SOC)進行估算,克服普通估算方法無法避免電池內阻誤差的缺點,通過Matlab/Simulink軟件仿真驗證可使估算誤差達到2%以內。At present,electric vehicles mainly use lithium batteries as the power source.In order to improve the running time and safety of lithium batteries,a safe and good operating environment for power batteries is provided,and a battery management system(BMS) has emerged.The BMS main control unit is based on the S32K144 automotive-grade control chip.Through the master-slave network control structure,it can collect and analyze the various parameters of the lithium battery.The Extended Kalman Filter(EKF) is used to estimate the state of charge(SOC) of the battery,which overcomes the shortcomings of the internal estimation method that cannot overcome the internal resistance error of the battery.It can be verified by Matlab/Simulink software simulation.The estimation error is within 2%.
標簽：
s32k144
bms
上傳時間：
2022-03-26
上傳用戶：XuVshu
實驗教學一直是工科教學中不可或缺的組成部分,對培養學生的動手能力,獨立思考能力,創新思維與發散思維具有重要的作用。針對目前電路教學實驗中電路仿真實驗與實物電路實驗各自獨立,無法統一問題,提出將仿真電路實驗與實物電路實驗有機的結合同步操作,并使用Web發布實現遠程實驗操作。采用Multisim作為電路實驗仿真平臺,NI Eiviss II作為實物電路實驗硬件平臺,運用LabVIEW整合Multisim電路仿真實驗與實物電路實驗,實現仿真與實物實驗有機結合,兩種實驗可同步進行。學生在仿真實驗中先可探索實驗,然后做實物實驗。同時運用LabVIEW開發出實驗過程人機交互操作接口界面,使用過程中效果良好。Experimental teaching has always been an indispensable part of engineering education.And it always plays an important role in cultivating students'practical ability,independent thinking ability,innovative thinking and divergent thinking.But simulation experiment and physical experiment cannot be unified in the circuit teaching experiment at present.In order to solve this problem,this paper proposes to combine organically the simulation circuit experiment with physical circuit experiment,and synchronously operate them.This paper uses the WEB publishing to achieve remote experimental operation.Multisim is used as the circuit simulation platform,and NI Eiviss II is used as the physical circuit hardware platform.Multisim circuit simulation experiment and physical circuit experiment are implemented by LabVIEW to realize the combination of simulation experiment and physical experiment.Students do explore experiments in simulation experiment firstly,and then do physical experiment.And this paper uses LabVIEW to develop the experimental man-machine interface.
標簽：
multisim
labview
上傳時間：
2022-04-05
上傳用戶：
