Artificial Intelligence in Process Engineering aims to present a diverse sample of Artificial Intelligence (AI) applications in process engineering. The book contains contributions, selected by the editors based on educational value and diversity of AI methods and process engineering application domains. Topics discussed in the text include the use of qualitative reasoning for modeling and simulation of chemical systems; the use of qualitative models in discrete event simulation to analyze malfunctions in processing systems; and the diagnosis of faults in processes that are controlled by Programmable Logic Controllers. There are also debates on the issue of quantitative versus qualitative information. The control of batch processes, a design of a system that synthesizes bioseparation processes, and process design in the domain of chemical (rather than biochemical) systems are likewise covered in the text. This publication will be of value to industrial engineers and process engineers and researchers.

Table of Contents:
Contributors Preface 1. Qualitative Modeling of Chemical Reaction Systems Abstract 1. Introduction 2. The QSIM Algorithm for Qualitative Simulation 3. Building Qualitative Models of Reaction Systems 4. Partial Quantitative Knowledge in Qualitative Models 5. Discussion Acknowledgment Appendix: Curvature Constraint Derivations References 2· Use of Qualitative Models in Discrete Event Simulation to Analyze Malfunctions in Processing Systems Abstract 1. Introduction 2. The Problem 3. Modeling and Simulation Background 4. CONFIG Implementation and Examples 5. Conclusions and Future Work Acknowledgment References 3. An Expert System for Diagnosis of a Sequential, PLC-Controlled Operation 1. Introduction 2. Programmable Logic Controllers 3. The Dead Operating State Diagnostic Scenario 4. Diagnosis 5. Diagnostic Methods 6. Expert System Development 7. General Aspects of the Expert System: "WRAPITUP" 8. Discussion and Summary Acknowledgments References 4. Fault Detection and Diagnosis Using Artificial Neural Networks Abstract 1. Introduction 2. Characteristics of Artificial Neural Networks 3. ZNL Architecture 4. Fault Detection and Diagnosis Examples 5. Conclusion References 5. A Modular Approach to Multiple Faults Diagnosis Abstract 1. Introduction 2. Shallow Versus Deep Knowledge 3. The Model-Based Approach 4. Multiple Faults Diagnosis 5. Other Approaches 6. Divide and Conquer (MFD2) 7. Conclusions References 6. Modeling Real-World Processes: Deep and Shallow Knowledge Integrated with Approximate Reasoning in a Diagnostic Expert System 1. Introduction 2. Overview 3. The Scenario 4. A Real-World Domain: The Power Plant 5. The Plant Model 6. The Diagnostic Expert System 7. Conclusions and Future Work Acknowledgments References 7. XIMKON—An Expert Simulation and Control Program Abstract 1. Introduction 2. Control System Design Process 3. XIMKON 4. Expert Process Modeling 5. Expert Controller Design 6. Conclusion Acknowledgments References 8. Exothermic Batch Chemical Reactor Automation Via Expert System Abstract 1. Introduction 2. Review 3. The Generalized Batch Reactor Control Problem 4. An Expert System Approach 5. Associated Conventional Control Strategies 6. Testing by Simulation and Follow-Up of Control Logic 7. Future Goals and Directions Acknowledgments Nomenclature References 9. Design of Protein Purification Processes by Heuristic Search Abstract 1. Introduction 2. Approaches to Design 3. BioSep Designer 4. Conclusion References 10. An Adaptive Heuristic-Based System for Synthesis of Complex Separation Sequences Abstract 1. Introduction 2. Problem Specifications 3. Knowledge Representation Strategy 4. Reasoning Strategy 5. Adaptation Mechanism 6. Example 7. Discussion and Conclusion Acknowledgment Appendix A: Quantitative Expression for the Term "Vary Widely" Appendix B: Fuzzy Membership Function Representing the Antecedent of Rule 2.9 Appendix C: Pattern Recognition for Stream Division or Separation References Other Suggested Readings Index