Please use this identifier to cite or link to this item:
Design and Implementation of Battery Monitoring System for Two-Wheeled Vehicles
Back Propagation Neural Network
|Abstract:||本研究以兩輪平衡車(Two-wheeled vehicle)為實驗平台，建構一套監控電動車電源的智慧型系統。因為兩輪平衡車為一倒單擺系統，它有著動態平衡以及負載重量不確定等特性，因此電壓電流的變化大，所以也較難準確且有效地估測載具電池殘電量(State of charge、SOC)、評估尚可工作的時間以及計算可行走的距離等。本文從能量的角度出發，去評估電池在工作狀態下的耗損狀況。首先，利用可程式規劃電子負載與直流電源供應器，對電池組進行特定形式的充放電實驗，找出電池工作時的耗損特性，進一步利用倒傳遞類神經網路(Back Propagation Neural Network)求出電池耗損的特性曲線。在兩輪平衡車方面，同時製作一具有藍牙通訊能力的電壓電流量測模組嵌入載具中。接下來，在電腦上利用所求出的電池耗損特性曲線設計成估測電池電量的演算法。最後，透過實際騎乘兩輪平衡車實驗，實驗結果顯示車上電池電量可有效的估測。|
This thesis using the two-wheeled vehicle as an experimental platform constructs an intelligent system to monitor the power of the two-wheeled vehicle. Two-wheeled vehicle which has the characteristics of dynamic balance and uncertain loading weights is an inverted pendulum system. Because its fluctuations of voltage and current are violent, it is difficult to estimate the state of charge (SOC), and the surplus time and distance of operation correctly and efficiently. Therefore, this thesis mainly studies the power estimation of the two-wheeled vehicle. First of all, based on programmable electronic load and DC power supply, some experiments of charge and discharge are performed in the batteries of the two-wheeled vehicle. Using the experimental data finds the loss features of the batteries. Then, using back propagation neural network approximates the loss features of the batteries. Moreover, a voltage and current measuring model with bluetooth communication is built and integrated into the electric control system of the two-wheeled vehicle. Next, an algorithm to estimate the battery power of the two-wheeled vehicle is developed. Finally, the real experiment result shows that the proposed method is effective.
|Appears in Collections:||學位論文|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.