為了有效設計及開發不同規格之動力組合之複合動力機車系統，並瞭解其動態特性，準確評估其省油效能及最佳動力分配，必須先建立合理的複合動力機車系統模擬模型，但因機車系統為非線性且複雜，完全以數學模型建立模型，將使模型過於複雜且參數過多，難以完全取得。本論文共分為兩個項目，首先針對具機械式無段變速機構(CVT)之複合動力機車系統，建立數值模擬模型，並且探討最佳化設計之法。本研究之複合動力機車，主要由125c.c.燃料噴射引擎(fuel injection internal combustion engine)、發電機、離合器、無段變速系統、最終傳動機構及動力輔助馬達所組成，本研究將以結合數學模型，物理特性方程式，以及實驗數據所構成之混合模型，建立該複合動力機車系統之模擬模型，並分別在Hardware-in-the-Loop (HIL)實驗平台，以及實車在機車底盤動力計上進行實驗，以驗證所建立模型之正確性。本論文之第二個工作項目為，發展一種整合前饋式類神經網路架構(Feedforward neural networks, FNN)與動態規劃(Dynamic Programming, DP)法之能量管理參數最佳化演算法，以便獲得複合動力機車之全域最佳動力分配值，並探討不同成本函數(Cost function)的定義及權重對於整體油耗及能量損耗之影響，以作為未來開發複合動力機車之參考。

The works of this thesis are divided into two parts. Firstly, a dynamic simulation model for hybrid electric scooters (HES) is developed. Attention is paid to the prediction of key parameters of the scooter such as the engine speed, CVT gear ratio, and fuel consumption on the ECE40 drive cycle in order to evaluate the performance of the HES. The scooter studied in this research consists of an electronically controlled fuel injection internal combustion engine (ICE), a generator, clutch, final drive gears, a power assistant motor, and a mechanical-type continuously variable transmission (CVT). Due to the system being complicated and nonlinear in nature, the simulation model is simultaneously constructed using mathematical models, empirical equations, and experimental data. The effectiveness of the model was verified experimentally using a scooter hardware-in-the-loop (HIL) system, as well as using a HES running on chassis dynamometer. With the developed dynamic simulation model, the second part of this thesis is to develop a parameter optimization algorithm for the energy management system (EMS) of the HES. The proposed algorithm integrates the dynamic programming (DP) method with a feedforward neural network (FNN), which is intended to determine the global optimal values of the power split ratio of the HES on the ECE40 drive cycles. Based on the algorithm, the effects of different cost functions with various weight coefficients on the fuel consumption rate were discussed.

摘 要 I
Abstract II
謝誌 IV
Table of Contents VI
List of Tables VIII
List of Figures IX
Chapter 1 Introductions 1
Chapter 2 Scooter System Modeling 6
2.1 Mathematical model 7
2.2 Dynamic simulation model 11
2.3 Scooter hardware-in-loop system 12
2.4 Verification of the Dynamic simulation model on the HIL 14
Chapter 3 Hybrid Electric Scooter Modeling 17
3.1 Dynamic model of hybrid electric scooter 18
3.2 Battery State-of-charge (SOC) 19
3.3 Simulation and Experiment of a Hybrid Electric Scooter 21
Chapter 4 Parameter Optimization Applying Dynamic Programming Method 27
4.1 Dynamic programming 27
4.1.1 Dynamic programming algorithm 28
4.1.2 Application to hybrid electric scooter 30
4.2 Dynamic simulation model 34
4.3 Velocity predication using feedforward neural networks 38
4.3.1 Overview of feedforward neural network 38
4.3.2 FNNs development procedure 40
4.3.2.1 Widrow-Hoff adaptation algorithm 41
4.3.2.2 Generalized Algorithm 42
4.3.2.3 Three-layer FNN structure 43
4.3.2.4 Application of FNNM 45
4.3.3 Application of FNNM to hybrid electric scooter 47
Chapter 5 Simulations on Parameter Optimization 49
5.1 Results of DP optimization 49
5.2 Effect of cost function 58
Chapter 6 Conclusions 60
Reference 61
About the author 65

Bertsekas D. P., 2007, “Dynamic Programming and Optimal Control: 3rd edition,”Athena Scientific, Vols. 1 and 2.
Bertsekas D. P., 1995, Programming and optimal control, Belmont MA, Dynamic, Athena Scientific.
Carbone G., Mangialardi L., Bonsen B., Tursi C., and Veenhuizen P.A., 2007, “CVT dynamics: Theory and experiments,” Mechanism and Machine Theory, Vol. 42, pp.409–428,.
Delprat S., Guerra T. M., Rimaux J., 2002, Optimal control of a parallel powertrain: From global optimization to real time control strategy, IEEE.
Francoa J., Franchekb M. A., Grigoriadisb K., “Real-time brake torque estimation for internal combustion engines,” Mechanical Systems and Signal Processing, Vol. 22, pp.338–361, 2008.
Fukuo K., Fujimura A., Saito M., Tsunoda K., Takiguchi S., 2002, “Development of the ultra-low-fuel-consumption hybrid car-INSIGHT,”JSAE Review, Vol.22, pp. 95-103.
Gillespie T. D., 1992, Fundamentals of Vehicle Dynamics, Society of Automotive Engineers, ISBN:1-56091-199-9.
Jalil N., Kheir N. A., Salman M., 1997,“A Rule-Based Energy Management Strategy for a Series Hybrid Vehicle,”Proceedings of the American Control Conference, Albuquerque, New Mexico.
Karwowski, W., Zurada, J., Marras, W.S., Gaddie, P., 1994, A prototype of the artificial neural network-based system for classification of industrial jobs with respect to risk of low back disorders. In: Aghazadeh, F. (Ed.), Advances in Industrial Ergonomics and Safety VI Taylor & Francis, Bristol, PA, pp 19-22.
Keller D.L., Wilson R.E., 1972, “Design and development of a high horsepower torque-sensing variable-speed drive,” SAE, Vol.61, No.720709.
Kheir N. A., Salman M. A., Schouten N. J., 2004, “Emissions and fuel economy trade-off for hybrid vehicles using fuzzy logic”Mathematics and computers in simulation, Vol. 66, pp.155-172, USA.
Killough, M.K., Crumpton, L.L., Calvert, A., Bowden, R., 1995, “An investigation of using neural networks to identify the presence of carpal tunnel syndrome,”the 4th Industrial Engineering Research Conference Proceedings, pp 695-667.
Kimura A., Abe T., Sasaki S., 1999, “Drive force control of a parallel-series hybrid system,”JSAE Review, Vol.20, pp.337-341.
Kuschewski J. F., 1993, ‘Application of Feedforward Neural Networks to Dynamical System Identification and Control’, IEEE Transactions on Control Systems Technology, Vol. 1, pp.37-49.
Lin B., 2000, “Conceptual design and modeling of a fuel cell scooter for urban Asia,” Journal of Power Sources, Vol. 86, pp.202–213.
Lin C. C., Peng H., Grizzle J. W., Kang J. M., 2003, “Power Management Strategy for a Parallel Hybrid Electric Truck,”IEEE Transactions on Control Syetems Technology, Vol.11, No.6, pp.839-849.
Maten J. and Anderson B., 2006, Continuously Variable Transmission (CVT), Society of Automotive Engineers.
Oliver L.R., Henderson D.D., 1972, “Torque-sensing variable-speed V-belt drive,” SAE Trans, Vol.81, No.720708.
Oliver L.R., Hornung K.G., Swenson J.E., Shapiro H.N., 1973, “Design equations for a speed and torque controlled variable-ratio V-belt transmission,” SAE Trans, Vol.82, No.730003.
Paganelli G., Delprat S., Guerra T. M., Rimaux J., Santin J. J., 2002, Equivalent Consumption Minimization Strategy For Parallel Hybrid Powertrains, IEEE.
Pérez L. V. and Pilotta E. A., 2007, “Optimal power split in a hybrid electric vehicle using direct transcription of an optimal control problem,”Mathematics and Computers in Simulation, 2007.
Pérez L. V., Bossio G. R., Moitre D., García G. O., 2006, “Optimization of power management in an hybrid electric vehicle using dynamic programming,”Science Direct-Mathematics and Computers in Simulation, Vol.73, pp.244-254.
Pérez L. V., Bossio G. R., Moitre D., García G. O., 2006, “Mathematics and Computers in Simulation,”Vol. 73, pp 244-254.
Schouten N. J., Salman M. A., Kheir N. A., 2003,“Energy management strategies for parallel hybrid vehicles using fuzzy logic” Control Engineering Practice, control engineering, USA, pp.171-177.
Sira-Ramírez H. and Eliezer C. M., 1995, “A sliding mode strategy for adaptive learning in Adalines”IEEE Transactions on circuits and systems-I:Fundamental theory and applications, Vol.42, No.12.
Sira-Ramírez H. and Żak S. H., 1991, “The adaptation of perceptrons with applications to inverse dynamics identification of unknown dynamic systems,” IEEE Trans. Syst., Nan, Cybern, vol. 21, pp. 634-643.
Sira-Ramírez H., 1995, “A sliding mode strategy for adaptive learning in adalines,” IEEE Transactions on Circuits and Systems, Vol. 42, No.12, pp.1001-1012.
Slotine J. E. and Li W. P., 1991, “Applied nonlinear control,”Prentice Hall, pp.311-388.
Tseng C. Y., Chen L. W., Lin Y. T., Li J. Y., 2008, “A Hybrid Dynamic Simulation Model for Urban Scooters with a Mechanical-Type CVT,”International Conference on Automation and Logistics, China.
Widrow B. and A.Lehr M., 1984,“30 years of adaptive neural networks : Perceptron, madaline, and backpropagation,” Proc. Nat. Acad Sci., Vol. 81, pp.3088-3092.
Worley W.S., 1955, “Designing adjustable-speed V-belt drives for farm implements,” SAE Trans, Vol. 63, No.550256, pp.321–33.
Żak S. H. and Sira-Ramírez H., 1990, “On the adaptation algorithms for feedforward neural networks,” in Theoretical Aspects of Industrial Design, D.A. Flield and V. Komkov, New York, pp.116-133.
Zurada, J. Karwowski, W., Marras, W.S., 1997, “A neural network-based system for classification of industrial jobs with respect to risk of low back disorders due to workplace design,”Appl. Ergon, Vol.28, No.1, pp.49-58.
丁一銘，2007, 複合動力機車線上最佳化系統之實驗研究，碩士論文，國立屏東科技大學，車輛工程系，屏東。
吳建勳，2003，正向混合動力車輛動態模擬與分析，碩士論文，大葉大學，車輛工程學系，彰化。
林政江，施保重，2002，混合動力車的理論與實際，全華科技圖書股份有限公司，台北。
孫宗瀛、楊英魁，Fuzzy 控制：理論、實作與應用，全華科技圖書出版公司，台北。