کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
7728289 1497917 2016 10 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Probability based remaining capacity estimation using data-driven and neural network model
ترجمه فارسی عنوان
تخمین ظرفیت باقیمانده احتمالی با استفاده از مدل داده و مدل شبکه عصبی
کلمات کلیدی
خودرو الکتریکی، برآورد وضعیت باتری، رویکرد مبتنی بر داده ها، مدل شبکه عصبی،
موضوعات مرتبط
مهندسی و علوم پایه شیمی الکتروشیمی
چکیده انگلیسی
Since large numbers of lithium-ion batteries are composed in pack and the batteries are complex electrochemical devices, their monitoring and safety concerns are key issues for the applications of battery technology. An accurate estimation of battery remaining capacity is crucial for optimization of the vehicle control, preventing battery from over-charging and over-discharging and ensuring the safety during its service life. The remaining capacity estimation of a battery includes the estimation of state-of-charge (SOC) and state-of-energy (SOE). In this work, a probability based adaptive estimator is presented to obtain accurate and reliable estimation results for both SOC and SOE. For the SOC estimation, an n ordered RC equivalent circuit model is employed by combining an electrochemical model to obtain more accurate voltage prediction results. For the SOE estimation, a sliding window neural network model is proposed to investigate the relationship between the terminal voltage and the model inputs. To verify the accuracy and robustness of the proposed model and estimation algorithm, experiments under different dynamic operation current profiles are performed on the commercial 1665130-type lithium-ion batteries. The results illustrate that accurate and robust estimation can be obtained by the proposed method.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Journal of Power Sources - Volume 315, 31 May 2016, Pages 199-208
نویسندگان
, , , ,