Hierarchical controller design

The two-time-scale behavior of the material balance of integrated processes with large recycle suggests the use of a hierarchical control structure with two tiers of control action  

To this end, let us complete the description of Equation (3.10) with a vector of output variables y  

The above time-scale decomposition provides a transparent framework for the selection of manipulated inputs that can be used for control in the two time scales. Specifically, it establishes that output variables y1 need to be controlled in the fast time scale, using the large flow rates u1, while the control of the variables ys is to be considered in the slow time scale, using the variables us. Moreover, the reduced-order approximate models for the fast (Equation (3.11)) and slow (the state-space realization of Equation (3.16)) dynamics can serve as a basis for the synthesis of well-conditioned nonlinear controllers in each time scale. 

---

The interactions between units do, however, become significant over long periods of time processes with recycle exhibit a slow, core dynamic component that must be addressed in any effective process-wide control strategy. This chapter presented an approach for systematically exploiting this two-time-scale behavior in a well-coordinated hierarchical controller design. The proposed framework relies on the use of simple distributed controllers to address unit-level control objectives in the fast time scale and a multivariable supervisory controller to accomplish process-wide control objectives over an extended time horizon. 

We proposed a method for deriving nonlinear low-dimensional models for the dynamics in each time scale. Subsequently, we proposed a hierarchical controller design framework that takes advantage of the time-scale multiplicity, and relies on a multi-tiered structure of coordinated decentralized and supervisory controllers in order to address distributed and process-level control objectives. 

The previous chapters have concentrated on analyzing the material-balance dynamics of several classes of integrated process systems. We demonstrated that the dynamic behavior of the systems considered exhibits several time scales and described a method for the derivation of reduced-order models describing the dynamics in each time scale. Also, a hierarchical controller design framework was introduced, with distributed control of the fast dynamics and supervisory control of the dynamics at the systems level. 

Figure 3.4 Hierarchical control relies on separate, but coordinated, fast and slow controllers, designed on the basis of the respective reduced-order models, to compute the values of the separate inputs that influence the fast and slow dynamics of the process. Tighter coordination between the distributed and supervisory control layers is achieved by using a cascaded configuration.

Section 2.2, in that two layers of control action involving separate controllers are proposed, whereas composite control relies on a single (possibly multivariable) controller with two components, a fast one and a slow one. Thus, the hierarchical control structure accounts for the separation of the flow rates of the process streams into two groups of inputs that act upon the dynamics in the different time scales. On the other hand, composite controller design (Figure 2.9) presupposes that the available manipulated inputs impact both the fast and the slow dynamics and relies on one set of inputs to regulate both components of the system dynamics. 

Remark 3.4. In the context of the present chapter (and of the remainder of the book), the term hierarchical control structure reflects the use of two (or multiple) coordinated tiers of control action, and should not be confused with hierarchical plant-wide controller design strategies (see, e.g. Ponton and Laing 1993, Luyben et al. 1997, Zheng et al. 1999, Antelo et al. 2007, Scattolini 2009, and references therein), which use the term hierarchy to denote a set of guidelines, to be followed in sequence, for designing the control system for a chemical plant. 

The resulting hierarchical control structure is represented schematically in Figure 5.2. Note that, while controller design proceeds in a bottom-up manner, starting from the fastest time scale, during the operation of the process there will exist a tight top-down interconnection via control cascades between the supervisory and regulatory layers. 

Integrating plantwide control in Hierarchical Conceptual Design... 

This chapter presents a simple synthesis-oriented approach methodology for integrating plantwide control in hierarchical conceptual design (Bildea, 2001). Two ideas are central ... 

Accidents in STAMP are the result of a complex process that results in the system behavior violating the safety constraints. The safety constraints are enforced by the control loops between the various levels of the hierarchical control structure that are in place during design, development, manufacturing, and operations. 

Creating a richer model of causation. (Leveson and Dulac 2005) propose the STAMP accident model and the STPA hazard assessment approach. STAMP is based on systems-theoretic concepts of hierarchical control, internal models of the environment and a classification of control errors. STPA takes that classification as the basis for iterative integrated control system safety assessment. At each design iteration the design is assessed and constraints are derived (equivalent to derived safety requirements) and imposed on further design iterations. 

Table H.l provides the key steps in a systematic procedure recommended here for design of plantwide control structures. It is based on the combined top-down/bottom-up approach of Larsson and Skogestad (2000) and Skogestad (2002) and the hierarchical organization that generally matches Fig. H.l. The proposed systematic plantwide control design approach consists of the four major steps shown in Table H.l.

---