跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.87) 您好!臺灣時間:2025/02/12 08:28
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:許建民
研究生(外文):Jian-Min Shiu
論文名稱:以混合模式來進行半批式反應器設定點的最適化
論文名稱(外文):SET-POINT OPTIMIZATION OF SEMIBATCH REACTORS VIA HYBRID MODEL
指導教授:張志雄張志雄引用關係
指導教授(外文):Jyh-Shyong Chang
學位類別:碩士
校院名稱:大同工學院
系所名稱:化學工程研究所
學門:工程學門
學類:化學工程學類
論文種類:學術論文
論文出版年:1999
畢業學年度:87
語文別:中文
論文頁數:61
中文關鍵詞:最適操作路徑前饋式類神經網路混合模式
外文關鍵詞:optimal trajectoryFANNhybrid model
相關次數:
  • 被引用被引用:0
  • 點閱點閱:130
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
批式或半批式反應器製程由於操作全程沒有穩態操作狀況,因此必須要有異於連續反應器之控制策略。此批式或半批式製程特性,事先規劃出一適當之操作路徑是必須的。
在本研究中,我們首先用一前饋式類神經網路來建構反應速率與反應物濃度和溫度的關係。在物理模式中我們用此類神經網路來取代其反應速率項而獲得所謂之混合模式。藉由此混合模式並應用動態規劃和GRG非線性規劃有效地找出最適化溫度操作路徑,以期獲得設定之產率。由模擬實驗的驗證,我們證明了以反應速率高低為分區依據所訓練之三組分區網路所建構之混合模式,有優異的內插能力,而能在大操作範圍內決定出正確的最適化溫度操作路徑。

Batch or semibatch processes require a control strategy different from that for continuous processes since they do not operate on steady state. The feature of batch or semibatch processes suggests that determination of a priori optimal trajectory previously is necessary.
In this study, we first used a feedforward artificial neural network (FANN) to interrelate the reaction rate with the concentration and temperature of the reacting medium. In the derived physical model, the reaction rate was replaced with the output of the built FANN, we obtained the so-called hybrid model. Application of the iterative dynamic programming and the generalized reduced gradient nonlinear programming to determine an optimal temperature trajectory was performed to obtain a desired conversion. Simulation studies showed that the hybrid model obtained by imbedding three sub-FANNs in the derived physical model performed the best capability of interpolation over a larger region of operation.

ACKNOWLEDGMENTS iii
ABSTRACT iv
中文摘要 v
TABLE OF CONTENTS vi
LIST OF TABLES viii
LIST OF FIGURES ix
NOTATION xii
1. INTRODUCTION 1
2. MATHEMATICAL MODEL FOR THE SEMIBATCH 6
REACTION
2.1 Detailed Physical Model 6
2.2 Hybrid Model 9
3. ARTIFICIAL NEURAL NETWORKS 10
3.1 Generating a Feedforward Artificial Neural Network 11
3.2 Training a Feedforward Artificial Neural Network 13
4. DETERMINATION OF REACTION RATE VIA ARTIFICIAL 16
NEURAL NETWORKS
4.1 The Learning Procedure 16
4.2 Validation Procedure 26
4.3 Training a FANN based on the partitioned data set 26
5. SET-POINT OPTIMIZATION OF THE SEMIBATCH 40
REACTOR SYSTEM
5.1 Determination of Optimal Operating Path 40
5.2 Solution of the Formulated Optimization Problem 40
5.3 The Calculated Optimal Operating Path based on the 43
Detailed Physical Model
5.4 The Calculated Optimal Operating Path based on the 43
Hybrid Model
5.4.1 Hybrid Model Imbedding the Trained FANN Based 43
on the Whole Data Set
5.4.2 Hybrid Model Imbedding the Trained FANN Based 48
on the Partitioned Data Set
6. CONCLUSIONS 54
REFERENCES 55
APPENDIX: The Iterative Dynamic Programming 57
A.1 Problem Formulation 57
A.2 Construction of Augmented Performance Index 57
A.3 Computation Algorithm of Dynamic Programming 59
個人資料

Bhat, J.; Madhavan, K. P.; Chidambaram, M. "Multivariable Global Input/Output Linearized Internal Model Control of a Semibatch Reactor". Ind. Eng. Chem. Res. 1991, 30, 1541-1547.
Chang, J. S.; Hseih, W. Y. "Optimization and Control of Semibatch Reactors". Ind. Eng. Chem. Res. 1995, 34, 545-556.
De Tremblay, M.; Luus, R. "Optimization of Non-Steady-State Operation of Reactor". Can. J. Chem. Eng. 1989, 67, 494-502.
Galvan, I. M., Zaldivar, J. M. Hernandez, H., Molga, E. "The Use of Neural Networks for Fitting Complex Kinetic Data". Comput. Chem. Engng. 1996, 20, 1451-1465.
Hugo, P. "Start-up and Operation of Exothermic Batch Process". Ger. Chem. Eng. 1981, 4, 161-173.
Korn, G. A., “Interactive Simulation of Backpropagation and Creeping-Random Search Learning in Neural Networks”. Simulation. 1990, Oct, 214-219.
Kravaris, C.; Wright, R. A.; Carrier, J. F. "Nonlinear Controllers for Trajectory Tracking in Batch Processes". Comput. Chem. Engng. 1989, 13, 73-82.
Liptak, B. G. "Controlling and Optimizing Chemical Reactors". Chem. Eng. 1986, 26, 69-81.
Luus, R. "Optimal Control by Dynamic Programming using Systematic Reduction in Grid Size". Int. J. Control. 1990a, 51, 995-1013.
Luus, R.; Rosen, O. "Application of Dynamic programming to Final State Constrained Optimal Control Problems". Ind. Eng. Chem. Res. 1991, 30, 1525-1530.
Morrison, J. D. “A Neural Network Model that Supports Real Time learning of Temporal Relationships in Complex Engineering Domains”. Simulation. 1992, 58, 152-163.
Pollard, J. F.,Broussard, M. R.,Garrison, D. B. and SAN, K. Y. "Process Identification Using Neural Networks" Comput. chem. Engng., 1992, 16, 253-270
Psichogios, D. C., Ungar, L. H. "A Hybrid Neural Network-First Principles Approach to Process Modeling". AIChE J., 1992, 38, 1499-1511
Savkovic'-Stevanovic', J. "A Neural Network Model for Analysis and Optimization of Process". Comput. Chem. Engng. 1993a, 17, S411-S416.
Savkovic'-Stevanovic', J. "Neural Networks for process Analysis and Optimization: Modeling and Applications". Comput. Chem. Engng. 1994, 18, 1149-1155.
Szeifert, F., Chovan, T., Nagy, L. “Process Dynamics and Temperature Control of Fed-batch Reactors”. Comput. Chem. Engng. 1995, 19, S447-S452.
Venkatasubramanian, V., Chan, K. “A Neural Network Methodology for Process Fault Diagnosis. AIChE J. 1989, 35, 1993-2004.
Venkatasubramanian, V., Vaidyanathan, R. and Yamanoto, Y. “Process Fault Detection and Diagnosis Using Neural Networks-I. Steady-State Process”. Comput. Chem. Engng. 1990, 14, 699-712.
Wang, C. J. and Wu, C. H. “Parallel Simulation of Neural Networks”. Simulation. 1991, Apr, 223-232.
Westerterp, K. R.; Van Swaaij; W. P. M.; Beenackers A. A. C. M. Chemical Reactor Design and Operation; Wiley: New York, 1984; pp 278-281.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關論文