http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Performance Comparison of Data Driven-based Capacity Forecasting for Supercapacitor
M. Adib Kamali(아딥),Chigozie U. Udeogu(치고지에),Angela Caliwag(안젤라),Wansu Lim(임완수) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
Data driven-based capacity forecasting has gained remarkable attention in recent years. The many approach of data driven-based capacity forecasting can be divided into two approaches: statistical model and deep learning model. Both model have their own advantages and disadvantages in forecasting capacity. To opening up the route in further systematic research in this area. This paper provide quantitative result with actual supercapacitor dataset to compare both forecasting accuracy and reliability for statistical and deep learning models.
Real-Time State of Charge estimation of Li-Ion Battery Considering The Effect of State of Health
Adib M. Kamali(아딥),Angela Caliwag(안젤라),Donguk Kwon(권동욱),Wansu Lim(임완수) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
Battery health estimation has an important role in safe operation and long-life battery. Battery health estimation can be represented as a battery State of charge (SOC) and state of health (SOH). The existing approach performs SOC without considering SOH effect which assumes that capacity always in the maximum value. This approach is lack of accuracy because capacity of battery degrades with aging not always same. Therefore, we propose SOC estimation considering the effect of SOH. The result shows the SOH adjusted the capacity value to improve SOC estimation accuracy. The experimental result is presented using battery management system C2000 and cloud computing.
Deep learning based SOC estimation for Hybrid Energy Storage System
M. Adib Kamali(아딥),Angela Caliwag(안제라),Wansu Lim(임완수) 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
This paper proposes deep learning-based state of charge (SOC) estimation for battery/supercapacitor hybrid energy storage system. Specifically, the proposed method used measurable battery data and artificial neural network to ensure that the proposed model is suitable for practical application. This paper also focuses on the reduction of the error caused by the disregard of dynamic charge and discharge process. We propose to separate the charge and discharge model to capture the dynamic SOC degradation. Results indicate that estimating SOC with our model reduced the root mean square error with respect to actual datasets.