| DC 欄位 |
值 |
語言 |
| DC.contributor | 電機工程學系 | zh_TW |
| DC.creator | 許哲瑋 | zh_TW |
| DC.creator | Che-Wei Hsu | en_US |
| dc.date.accessioned | 2019-8-1T07:39:07Z | |
| dc.date.available | 2019-8-1T07:39:07Z | |
| dc.date.issued | 2019 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:444/thesis/view_etd.asp?URN=106521081 | |
| dc.contributor.department | 電機工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 本論文研究目的為研製與發展以人工智慧機器學習為基礎之高性能同步磁阻馬達驅動系統，提出利用遞迴式模糊類神經網路之智慧型每安培最大轉矩追蹤控制法。本論文首先介紹磁場導向控制之傳統每安培最大轉矩控制，但因同步磁阻馬達的磁阻轉矩高度非線性且時變，造成傳統每安培最大轉矩於實際應用中實現困難。有鑒於此，在同步磁阻馬達驅動系統中，結合了適應性計算電流速度控制器和遞迴式勒壤得模糊類神經網路於每安培最大轉矩控制，可不需藉由查表法而線上獲得最佳的電流角命令。適應性計算電流速度控制器用於產生定子電流命令，並利用李亞普諾夫穩定性理論推導出適應律，用於線上適應總集不確定項，證明了適應性計算電流速度控制的漸近穩定。
本文還設計與分析出以電流角控制為基礎之適應性步階迴歸速度控制系統，應用於同步磁阻馬達驅動系統中，描述以磁場導向控制的比例-積分控制系統，但因同步磁阻馬達受到磁飽和與無法模型化的影響， 軸與 軸電流命令目前沒有一種良好的方法可以獲得。因此，設計出以電流角控制為基礎之適應性步階迴歸系統於同步磁阻馬達的速度追隨上。本文提出的適應性步階迴歸速度控制用於產生定子電流命令，並利用李亞普諾夫穩定性理論推導出適應律，用於線上適應總集不確定項，保證適應性步階迴歸速度追隨控制漸進穩定。使用有限元素分析法分析每安培最大轉矩控制，獲得電流最小化的電流角命令，並將結果藉由查表法做應用。為了改善同步磁阻馬達於每安培最大轉矩下的暫態響應，使用遞迴式赫米特模糊類神經網路作為智慧型速度暫態控制系統產生補償電流角命令。
最後本研究以32位元浮點運算數位訊號處理器TMS320F28075分別將適應性計算電流速度控制器結合遞迴式勒壤得模糊類神經網路，及以智慧型電流角控制為基礎的適應性步階迴歸控制，實現於功率4kW的同步磁阻馬達驅動系統，通過實驗結果驗證了所提出之理論的強健性。
| zh_TW |
| dc.description.abstract | In order to construct an artificial intelligence mechine learning based high-performance synchronous reluctance motor (SynRM) drive system, an intelligent maximum torque per ampere (MTPA) tracking control using a recurrent fuzzy neural network (RFNN) is proposed in this study. First, a traditional MTPA (TMTPA) control system based on field-oriented control (FOC) is introduced. Since the reluctance torque of the SynRM is highly nonlinear and time-varying, the MTPA tracking control is very difficult to achieve by using the TMTPA control in practical applications. Then, an adaptive computed current (ACC) speed control using the proposed recurrent legendre fuzzy neural network (RLFNN) for the MTPA tracking control of a SynRM drive system, which does not use a lookup table (LUT) and can effectively obtain the optimal current angle command of MTPA online, is described in detail. The ACC speed control is applied to generate the stator current magnitude command, and an adaptation law is proposed to online adapt the value of a lumped uncertainty in the ACC control. Moreover, the adaptation law is derived using the Lyapunov stability theorem to guarantee the asymptotic stability of the ACC speed control. Furthermore, the proposed RLFNN is employed to produce the incremental command of the current angle.
The design and analysis of a novel current angle-based adaptive backstepping (ABS) speed control system for a SynRM drive system is also presented in this study. First, a proportional-integral (PI) control system with FOC is described. Owing to the unmodeled dynamics and magnetic saturation effects of the SynRM, currently there is no predominant way to design the command of -axis and -axis currents for the SynRM. Therefore, an ABS based on current angle control (ABS-CAC) system is designed for the speed tracking of SynRM. The ABS speed tracking control is proposed to generate the stator current command, and an adaptive law is derived by Lyapunov stability theorem to online adapt the value of a lumped uncertainty and ensure the asymptotic stability of the ABS speed tracking control. Furthermore, a LUT of the results of MTPA analysis by using the finite element analysis (FEA) method is proposed to provide the current angle command for current minimizing solutions. In addition, to improve the transient dynamic response of SynRM under MTPA operating conditions, an intelligent speed transient control (ISTC) system using a recurrent Hermite fuzzy neural network (RHFNN) is developed to generate the compensated current angle command.
Finally, the ACC speed control with RLFNN and ABS-CAC are implemented in a TMS320F28075 32-bit floating-point digital signal processor (DSP) for a 4 kW SynRM drive system. The robustness and effectiveness of the proposed intelligent MTPA tracking control are verified by some experimental results.
| en_US |
| DC.subject | 人工智慧機器學習 | zh_TW |
| DC.subject | 適應性計算電流速度控制 | zh_TW |
| DC.subject | 遞迴式勒壤得模糊類神經網路 | zh_TW |
| DC.subject | 適應性步階迴歸控制 | zh_TW |
| DC.subject | 遞迴式赫米特模糊類神經網路 | zh_TW |
| DC.subject | 電流角控制 | zh_TW |
| DC.subject | 有限元素分析 | zh_TW |
| DC.subject | 每安培最大轉矩 | zh_TW |
| DC.subject | 同步磁阻馬達 | zh_TW |
| DC.subject | artificial intelligence mechine learning | en_US |
| DC.subject | adaptive computed current speed control | en_US |
| DC.subject | recurrent Legendre fuzzy neural network | en_US |
| DC.subject | adaptive backstepping control | en_US |
| DC.subject | recurrent Hermite fuzzy neural network | en_US |
| DC.subject | current angle control | en_US |
| DC.subject | finite element analysis | en_US |
| DC.subject | maximum torque per ampere | en_US |
| DC.subject | synchronous reluctance motor | en_US |
| DC.title | 同步磁阻馬達驅動系統之智慧型每安培最大轉矩追蹤控制 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Intelligent Maximum Torque per Ampere Tracking Control of Synchronous Reluctance Motor Drive System | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |