Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/8203
DC FieldValueLanguage
dc.contributor.advisor蔡清池zh_TW
dc.contributor.advisorC.C. Tsaien_US
dc.contributor.author林湧淳zh_TW
dc.contributor.authorLin, Yung-Chunen_US
dc.date2003zh_TW
dc.date.accessioned2014-06-06T06:41:12Z-
dc.date.available2014-06-06T06:41:12Z-
dc.identifier.urihttp://hdl.handle.net/11455/8203-
dc.description.abstract本論文旨的在發展迴授類神經網路(Recurrent Neural networks)為基礎的適應預測控制方法及其工業程序之應用技術。為了準確的估測工業程序系統的系統參數與輸出預測值,本文提出類神經線性化驗證器(Neural-based Linearized Identifier)及長時間類神經預測器(Long-Range Predictor)。為達到良好的控制成效,本文提出二種適應預測控制策略:類神經線性化模型參考預估控制 (Neural-based Linearized Model Predictive Control)及適應多步類神經預估控制 (Adaptive Multi-step Neural Predictive Control)。經由電腦模擬和變頻油溫冷卻實驗結果證明本論文所提之的控制策略,皆能有效地在工業程序上得到優異之控制性能。zh_TW
dc.description.abstractAbstract This thesis develops methodologies and techniques for predictive control of industrial processes using recurrent neural networks. The neural-based linearized identifier and long-range predictor are presented in order to estimate the parameters of industrial processes and to obtain output predictions. To achieve satisfactory control performance, two adaptive predictive control strategies, neural-based linearized predictive control and adaptive multi-step neural predictive control are proposed. Numerous simulation results are provided to show the effectiveness and feasibility of proposed control methods. Experimental results for a variable-frequency oil-cooling process have been conducted to show that the presented control approaches would perform well for the industrial processes with nonlinear, time-delay and nonminimum-phase properties.en_US
dc.description.tableofcontentsContents Chinese Abstract i English Abstract ii Contents iii List of Figures vi Chapter 1 Introduction 1 1.1 Introduction 1 1.2 Survey of Related Research 3 1.3 Contributions of the Thesis 5 1.4 Organization of the Thesis 6 Chapter 2 Neural-based Predictive Control using Dynamic Linearization 7 2.1 Introduction 7 2.2 Simple Recurrent Neural Networks Modeling and Approximation 8 2.3 Generalized Predictive Control 15 2.4 Neural-based Dynamic Linearized Predictive Control Algorithm 20 2.5 Simulation Results 21 2.6 Concluding Remarks 23 Chapter 3 Adaptive Neural Nonlinear Predictive Control 24 3.1 Introduction 24 3.2 System Identification 26 3.3 Multiple-Step Neural Predictive Control 30 3.3.1 Nonlinear Long-Range SRNN-based Prediction 30 3.3.2 Multiple-Step Neural Networks Controller 32 3.4 Adaptive Recurrent Neural Predictive Control Algorithm 35 3.5 Analysis of Stability 36 3.5.1 Stability of the SRNN-based Identifier 36 3.5.2 Stability of the SRNN-based controller 38 3.5.3 convergence of the weightings of the SRNN system identifier 42 3.5.4 convergence of the SRNN Controller's Weightings 44 3.6 Simulation Results and Discussion 46 3.7 Concluding Remarks 48 Chapter 4 Apply Experimental Setup, Results and Discussion 49 4.1 Introduction 49 4.2 Brief Description of the Variable-frequency oil-cooling Control System 50 4.3. Experimental Results and Discussion 54 4.3.1. Neural-based Predictive Control using Dynamic Linearization 54 4.3.2. Adaptive Neural Nonlinear Predictive Control 61 4.4 Concluding Remarks 66 Chapter 5 Summaries and Recommendations 68 5.1 Summaries 68 5.2 Recommendations 70 Bibliography 71 List of Figures Figure 1.1 Configuration of the oil-cooling machine 4 Figure 2.1 Block diagram of the neural-base predictive control 8 Figure 2.2 Simple recurrent neural networks dynamic linearized structure 9 Figure 2.3 The tracking response for rigid non-minimum phase plant. 22 Figure 2.4 The set-point tracking errors. 22 Figure 3.1 Block diagram of adaptive neural nonlinear predictive control. 25 Figure 3.2 SRNN-based system identifier structure 26 Figure 3.3 Structure of Long-Range Predictor 30 Figure 3.4 The tracking response of SRNN-based predictive output and rigid non-minimum phase plant. 47 Figure 3.5 The set-point tracking errors. 47 Figure 3.6 The error of the SRNN controllers for the rigid non-minimum phase plant. 48 Figure 4.1 Oil-cooling processes. 51 Figure 4.2 A recent picture of the PT100 senor. 51 Figure 4.3 Photo of FR-E500 actuator 52 Figure 4.4 The hardware is PCI-1202L AD/DA card 52 Figure 4.5 The schematics of the isolated circuit. 53 Figure 4.6 Block diagram of the oil-cooling control system 53 Figure 4.7 Temperature step tracking response with no load 55 Figure 4.8 Tracking errors with no load 56 Figure 4.9 Behavior of the temperature controller parameters with a fixed heat load of no load 56 Figure 4.10 Behavior of the time history delay times with a fixed heat load of no load 57 Figure 4.11 Temperature step tracking response with the load of 500W 57 Figure 4.12 Tracking errors with the a fixed heat load of 500W 58 Figure 4.13 Behavior of the temperature controller parameters with a fixed heat load of the load of 500W 58 Figure 4.14 Behavior of the delay time for the system with a fixed heat load of the load of 500W 59 Figure 4.15 Temperature step tracking response with the load changed from 0 Watts to 500 Watts after the 200th period 59 Figure 4.16 Tracking errors with the load changed from 0 Watts to 500 Watts after the 200th period 60 Figure 4.17 Behavior of the temperature controller parameters with the load changed from 0 Watts to 500 Watts after the 200th period 60 Figure 4.18 Behavior of the delay times with the load changed from 0 Watts to 500 Watts after the 200th period 61 Figure 4.19 Temperature step tracking response with no load 63 Figure 4.20 Tracking errors with no load 63 Figure 4.21 Tracking errors between processes and designing temperature with no load 64 Figure 4.22 Temperature step tracking response with the load of 500W 64 Figure 4.23 Tracking errors with the load of 500W 65 Figure 4.24 Tracking errors between processes and designing temperature with 500W load 65en_US
dc.language.isoen_USzh_TW
dc.publisher電機工程學系zh_TW
dc.subjectpredictive controlen_US
dc.subject預估控制zh_TW
dc.subjectrecurrent neural networken_US
dc.subjectNeural-based Linearized Identifieren_US
dc.subject迴授式類神經網路zh_TW
dc.subject類神經辨識器zh_TW
dc.title適應類神經預估控制與其工業程序應用zh_TW
dc.titleAdaptive Neural Predictive Control with Applications toa Class of Industrial Processesen_US
dc.typeThesis and Dissertationzh_TW
item.fulltextno fulltext-
item.languageiso639-1en_US-
item.openairetypeThesis and Dissertation-
item.cerifentitytypePublications-
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.grantfulltextnone-
Appears in Collections:電機工程學系所
Show simple item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.