Most modern chemical plants are complex networks of multiple interconnected nonlinear process units, often with multiple recycle and by-pass streams and energy integration. Interactions between process units often lead to plant-wide operability problems (i.e., difficulties in process control). Plant-wide operability analysis is often difficult due to the complexity and nonlinearity of the processes. This article provides a new framework of dynamic operability analysis for plant-wide processes, based on the dissipativity of each process unit and the topology of the process network. Based on the concept of dissipative systems, this approach can deal with nonlinear processes directly. Developed from a network perspective, the proposed framework is also inherently scalable and thus can be applied to large process networks. V V C 2009 American Institute of Chemical Engineers AIChE J, 55: [963][964][965][966][967][968][969][970][971][972][973][974][975][976][977][978][979][980][981][982] 2009 Keywords: plant-wide control, nonlinear systems, process networks, process operability
IntroductionProcess control has been playing an increasingly important role in the process industries as more process integration and tighter operating conditions are putting greater demands on control system performance. The traditional approach to process design and control has been to perform the design of the process and the control system sequentially.1 As little consideration is given to dynamic operability in the process design stage, the outcome of this approach sometimes is a plant whose dynamic characteristics lead to serious operating difficulties associated with significant economic penalties.2 It is well known that a process design fundamentally determines the inherent operability properties of the process since it imposes limitations on control performance regardless of what control method is implemented. Therefore, process operability analysis should be performed in the process design stage for better integration of process design and control. 1 Modern chemical plants have two important features: (1) they are networks of significant number of process units with recycle and heat integration configurations; (2) process units are generally nonlinear. Interactions between the process units are often the main causes of operability problems. As such, it is important to develop appropriate tools to quantitatively assess the effects of nonlinear dynamics of process units and the interactions between them on plant-wide operability. 3 In the past two decades, a number of operability analysis tools have been reported. Many of these were developed for linear processes, based on, e.g., process zero dynamics, minimum singular values and loop interactions. [4][5][6] For nonlinear processes, a direct approach to operability analysis is to solve a dynamic performance optimisation problem numerically. Although this approach allows seamless integration of process and control design, 6-9 the complexity of the resulting problem grows quickly with t...