Model-Predictive Control of Polyolefin Processes Y.A. Liu and Niket Sharma Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061, U.S.A. Abstract This chapter presents an in-depth exploration of model-predictive control (MPC) or advanced process control (APC) techniques in the optimization of polyolefin manufacturing processes. Drawing on the foundational motivations outlined in previous discussions, it highlights the pivotal role of APC in enhancing industrial efficiency and innovation. Through a comprehensive introduction to the basic concepts and tools of APC, including definitions of manipulated variables (MV), feedforward/disturbance variables (FF/DV), controlled variables (CV), and the intricacies of multivariable dynamic models, this work delineates the advantages of APC over traditional proportional-integral- derivative (PID) control systems. It further elucidates the mechanisms through which APC achieves its benefits, such as model CV prediction, economic optimization, and dynamic control execution, leveraging Aspen DMCplus and DMC3 control structures for illustration. The chapter provides a detailed walkthrough of developing a dynamic matrix controller model for a copolymerization process utilizing Aspen DMC3 Builder, transitioning to the formulation and control of nonlinear processes. It addresses the challenges inherent in constructing nonlinear dynamic models for polymerization process control, introduces the Wiener model for nonlinear processes, and discusses the state-space, bounded derivative network (SS-BDN) for nonlinear controller modeling. A hands-on workshop for the development of a nonlinear model-predictive control (NMPC) of a polypropylene process is presented, culminating in an overview of recent advancements in MPC with embedded AI technologies. Serving both as a primer for newcomers and a sophisticated reference for experienced practitioners and scholars, this work underscores the transformative potential of APC integrated with AI in the polyolefin production sphere. It advocates for the systematic adoption of these advanced control strategies to realize significant improvements in process efficiency, optimization, and innovation within the chemical processing industries. This is a preprint version of a chapter from our book - Integrated Process Modeling, Advanced Control and Data Analytics for Optimizing Polyolefin Manufacturing. Please cite the original work if referenced [26,35] 8.1 Introduction to Advanced Process Control (APC) This chapter covers the fundamentals and practice of model-predictive control (MPC), or advanced process control (APC), of polyolefin processes. The motivation for this chapter appeared previously in Chapter 1, Section 1.4.2, discussing the industrial and potential applications of advanced process control to optimizing polyolefin manufacturing. We begin by introducing the basic concepts and tools of APC in Section 8.1. Specifically, Section 8.1.1 presents some basic definitions, including manipulated variable (MV), feedforward/disturbance (FF/DV)