Discrete Input-Output Models

Recursive estimation methods are routinely used in many applications where process measurements become available continuously and we wish to re-estimate or better update on-line the various process or controller parameters as the data become available. Let us consider the linear discrete-time model having the general structure  

We shall present three recursive estimation methods for the estimation of the process parameters (ai.ap, b0, b. bq) that should be employed according to the statistical characteristics of the error term sequence e s (the stochastic disturbance). 

---

The general input-output model for discrete data could be written ... 

In the classical concept of predictive control, the trajectory (or set-point) of the process is assumed to be known. Control is implemented in a discrete-time fashion with a fixed sampling rate, i.e. measurements are assumed to be available at a certain frequency and the control inputs are changed accordingly. The inputs are piecewise constant over the sampling intervals. The prediction horizon Hp represents the number of time intervals over which the future process behavior will be predicted using the model and the assumed future inputs, and over which the performance of the process is optimized (Fig. 9.1). Only those inputs located in the control horizon H, are considered as optimization variables, whereas the remaining variables between Hr+1 and Hp are set equal to the input variables in the time interval Hr. The result of the optimization step is a sequence of input vectors. The first input vector is applied immediately to the plant. The control and the prediction horizon are then shifted one interval forward in time and the optimization run is repeated, taking into account new data on the process state and, eventually, newly estimated process parameters. The full process state is usually not measurable, so state estimation techniques must be used. Most model-predictive controllers employed in industry use input-output models of the process rather than a state-based approach. 

The continuous models (e.g., differential equations in the time domain, or input-output models in the Laplace domain) are not convenient to use to analyze the dynamic behavior of loops with computer control discrete-time models are needed. 

D(z) is defined as the transfer function of the discrete system and we notice that it is completely parallel to our familiar transfer function for continuous systems in the Laplace domain. Equation (29.2) constitutes the input-output model for the discrete system and can be used to compute the dynamic response of a discrete system subject to an input change (see Figure 29.2b). 

Input/Output interface, 557-61 Input-output models, 81 discrete-time, 609-26 examples, 81-85, 162, 163, 166 using Laplace transforms, 159-66 Input variables, 12-14 Integral of absolute error, 302 Integral control action, 273, 277-78 advantages and drawbacks, 274-75, 307... 

Bode diagram, 330-31, 334-37 frequency response, 323-24 interacting capacities, 197-200 noninteracting capacities, 194-96 pulse transfer function, 619 Multiple-input multiple-output system, 20 discrete-time model, 586 discrete transfer function, 612 input-output model, 83-85, 163-68 linearization, 121-26 transfer-function matrix, 164, 166 Multiple loop control systems, 394-409 Multiplexer, 560, 564 Multivariable control systems, 461-62 alternative configurations, 467-84 decoupling of loops, 503-8 design questions, 461-62 interaction of loops, 487-94 selection of loops, 494-503 Multivariable process (see Multiple-input multiple-output system)... 

The development of input-output models for discrete-time systems, which constitute the basis for the dynamic analysis and design of control loops... 

In the preceding section the analysis was centered around the response of the discrete components in a direct digital control (DDC) loop with characteristic representative the control algorithm. The use of z-transforms allowed easy and straightforward development of simple input-output models through the discrete transfer functions. 

In this section we consider the continuous elements of a DDC loop (i.e., the process and the hold as shown in Figure 29.1b). Although both elements are continuous, the input to the hold is a discrete-time signal c(nT), and the output from the process is sampled by a sampler. We would like to relate the sampled output values y(nT) with the discrete control commands c(nT), through a simple input-output model in the z -domain of the form... 

Especially in an operating environment where quality has become more important than quantity, there is a strong desire to develop input-output models that can be used in advanced control applications, in order to develop control strategies for quality improvement. These models are usually discrete hnear transfer function (difference equation) type models, which provide a representation of the dynamic behaviour of the process at discrete sampling... 

Now, to be sure, McCulloch-Pitts neurons are unrealistically rendered versions of the real thing. For example, the assumption that neuronal firing occurs synchronously throughout the net at well defined discrete points in time is simply wrong. The tacit assumption that the structure of a neural net (i.e. its connectivity, as defined by the set of synaptic weights) remains constant over time is known be false as well. Moreover, while the input-output relationship for real neurons is nonlinear, real neurons are not the simple threshold devices the McCulloch-Pitts model assumes them to be. In fact, the output of a real neuron depends on its weighted input in a nonlinear but continuous manner. Despite their conceptual drawbacks, however, McCulloch-Pitts neurons are nontrivial devices. McCulloch-Pitts were able to show that for a suitably chosen set of synaptic weights wij, a synchronous net of their model neurons is capable of universal computation. This means that, in principle, McCulloch-Pitts nets possess the same raw computational power as a conventional computer (see section 6.4). 

Continuous Memoryless Channels.—The coding theorem of the last section will be extended here to the following three types of channel models channels with discrete input and continuous output channels with continuous input and continuous output and channels with band limited time functions for input and output. Although these models are still somewhat crude approximations to most physical communication channels, they still provide considerable insight into the effects of the noise and the relative merits of various transmission and detection schemes. 

The integer variables can be used to model, for instance, sequences of events, alternative candidates, existence or nonexistence of units (in their zero-one representation), while discrete variables can model, for instance, different equipment sizes. The continuous variables are used to model the input-output and interaction relationships among individual units/operations and different interconnected systems. 

Here, neural network techniques are used to model these process-model mismatches. The neural network is fed with various input data to predict the process-model mismatch (for each state variable) at the present discrete time. The general input-output map for the neural network training can be seen in Figure 12.2. The data are fed in a moving window scheme. In this scheme, all the data are moved forward at one discrete-time interval until all of them are fed into the network. The whole batch of data is fed into the network repeatedly until the required error criterion is achieved. 

Given a set of experimental data, we look for the time profile of A (t) and b(t) parameters in (C.l). To perform this key operation in the procedure, it is necessary to estimate the model on-line at the same time as the input-output data are received [600]. Identification techniques that comply with this context are called recursive identification methods, since the measured input-output data are processed recursively (sequentially) as they become available. Other commonly used terms for such techniques are on-line or real-time identification, or sequential parameter estimation [352]. Using these techniques, it may be possible to investigate time variations in the process in a real-time context. However, tools for recursive estimation are available for discrete-time models. If the input r (t) is piecewise constant over time intervals (this condition is fulfilled in our context), then the conversion of (C.l) to a discrete-time model is possible without any approximation or additional hypothesis. Most common discrete-time models are difference equation descriptions, such as the Auto-.Regression with eXtra inputs (ARX) model. The basic relationship is the linear difference equation ... 

We give only a short description of the three supply chain configurations and their simulation models for details we refer to Persson and Olhager (2002). At the start of our sequential bifurcation, we have three simulation models programmed in the Taylor II simulation software for discrete event simulations see Incontrol (2003). We conduct our sequential bifurcation via Microsoft Excel, using the batch run mode in Taylor II. We store input-output data in Excel worksheets. This set-up facilitates the analysis of the simulation input-output data, but it constrains the setup of the experiment. For instance, we cannot control the pseudorandom numbers in the batch mode of Taylor II. Hence, we cannot apply common pseudorandom numbers nor can we guarantee absence of overlap in the pseudorandom numbers we conjecture that the probability of overlap is negligible in practice. 

In this study we identify an SMB process using the subspace identification method. The well-known input/output data-based prediction model is also used to obtain a prediction equation which is indispensable for the design of a predictive controller. The discrete variables such as the switching time are kept constant to construct the artificial continuous input-output mapping. With the proposed predictive controller we perform simulation studies for the control of the SMB process and demonstrate that the controller performs quite satisfactorily for both the disturbance rejection and the setpoint tracking. 

In order to solve the first principles model, finite difference method or finite element method can be used but the number of states increases exponentially when these methods are used to solve the problem. Lee et u/.[8] used the model reduction technique to reslove the size problem. However, the information on the concentration distribution is scarce and the physical meaning of the reduced state is hard to be interpreted. Therefore, we intend to construct the input/output data mapping. Because the conventional linear identification method cannot be applied to a hybrid SMB process, we construct the artificial continuous input/output mapping by keeping the discrete inputs such as the switching time constant. The averaged concentrations of rich component in raffinate and extract are selected as the output variables while the flow rate ratios in sections 2 and 3 are selected as the input variables. Since these output variables are directly correlated with the product purities, the control of product purities is also accomplished. 

A linear model predictive control law is retained in both cases because of its attracting characteristics such as its multivariable aspects and the possibility of taking into account hard constraints on inputs and inputs variations as well as soft constraints on outputs (constraint violation is authorized during a short period of time). To practise model predictive control, first a linear model of the process must be obtained off-line before applying the optimization strategy to calculate on-line the manipulated inputs. The model of the SMB is described in [8] with its parameters. It is based on the partial differential equation for the mass balance and a mass transfer equation between the liquid and the solid phase, plus an equilibrium law. The PDE equation is discretized as an equivalent system of mixers in series. A typical SMB is divided in four zones, each zone includes two columns and each column is composed of twenty mixers. A nonlinear Langmuir isotherm describes the binary equilibrium for each component between the adsorbent and the liquid phase. 

The discretization of continuous models with dead time is rather straightforward. For example, consider a first-order process with dead time tj between the input m(t) and the output y(t) ... 

Equation (27.9) is the discrete-time dynamic model of a first-order process and shows what the output of the process will be at the next time instant, using current values of the input mn and output y . 

We shall call discrete those dynamic systems that process their input signals only at the sampling instants, thus producing output signals that are only defined at particular time instants. In other words, discrete is a system whose input and output are discrete-time signals. The input-output relationship for such systems is given by a discrete-time model (i.e., by a difference equation). 

This chapter will focus on practicable methods to perform both the model specification and model estimation tasks for systems/models that are static or dynamic and linear or nonlinear. Only the stationary case win be detailed here, although the potential use of nonstationary methods will be also discussed briefly when appropriate. In aU cases, the models will take deterministic form, except for the presence of additive error terms (model residuals). Note that stochastic experimental inputs (and, consequently, outputs) may stiU be used in connection with deterministic models. The cases of multiple inputs and/or outputs (including multidimensional inputs/outputs, e.g., spatio-temporal) as well as lumped or distributed systems, will not be addressed in the interest of brevity. It will also be assumed that the data (single input and single output) are in the form of evenly sampled time-series, and the employed models are in discretetime form (e.g., difference equations instead of differential equations, discrete summations instead of integrals). 

The graphical method is based on the notion that the mathematical model of a discrete-time finite-order (stationary) dynamic system is, in general, a multivariate function /( ) of the appropriate lagged values of the input-output variables... 

A DEVS model of a system takes a sequence of discrete events, i.e. of instantaneous system changes, as inputs and produces an output sequence of events according its initial conditions. Figure 2.22 displays this input/output behaviour of a DEVS model. Events can be characterised by a value and the time point of their occurrence. Accordingly, they are indicated by perpendicular strokes in Fig. 2.22. A sequence of events is called an event trajectory. It is assumed that the number of state changes in any finite time interval is finite. 

---