類神經網路應用於兩輪平衡載具之電池能量預測研究
No Thumbnail Available
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究的目的在於開發出電動載具所使用的電池能量監測與電量預測系統。首先，監測載具電池的充放電狀態與實際騎乘的車況資訊，建立雲端載具電量資料庫，其中車況資訊包括溫度、電流、電壓、傾斜度、顛頗度、車速等。接著，根據資料庫的電量資訊，利用倒傳遞類神經網路來學習載具電池耗損曲線，並且建立電量估測模型。此外，利用歷史電量資料來即時預測未來電池能量數值變化，以提供使用者了解電動載具電池未來電量使用狀況，以及電動載具可再移動距離資訊。
最後，為了驗證電動載具電量估測與預測方法的可行性與精確度，本研究以兩輪平衡電動車為例，在國立臺灣師範大學校園進行實際騎乘實驗，實驗結果顯示該電量估測與預測方法可行，精確度符合需求。
The purpose of this study is to develop a battery power monitoring and forecasting system for electric vehicle. First of all, the state of charge of the vehicle and real road experiment data can be saved to a cloud database. The data of the database include temperature, battery current, battery voltage, road conditions, speed, etc. Based on these data, the back propagation artificial neural networks (BPANN) are used to learn the loss feature of the battery and to build the estimate model of the battery. Besides, the historical data are utilized to forecast the future power result. Therefore, the users can acquire the power information and running distance immediately. Finally, some experiments are performed to verify the accuracy and feasibility of the power monitoring and forecasting system by using the two-wheeled vehicle at National Taiwan Normal University. The experiment results show that the power monitoring and forecasting system is feasible, and its accuracy is acceptable
The purpose of this study is to develop a battery power monitoring and forecasting system for electric vehicle. First of all, the state of charge of the vehicle and real road experiment data can be saved to a cloud database. The data of the database include temperature, battery current, battery voltage, road conditions, speed, etc. Based on these data, the back propagation artificial neural networks (BPANN) are used to learn the loss feature of the battery and to build the estimate model of the battery. Besides, the historical data are utilized to forecast the future power result. Therefore, the users can acquire the power information and running distance immediately. Finally, some experiments are performed to verify the accuracy and feasibility of the power monitoring and forecasting system by using the two-wheeled vehicle at National Taiwan Normal University. The experiment results show that the power monitoring and forecasting system is feasible, and its accuracy is acceptable
Description
Keywords
倒傳遞類神經網路, 殘電量, 電動載具, Back propagation artificial neural networks, state of charge, electric vehicle