State equations from transfer functions

The state-spaee representation in equation (8.33) is ealled the eontrollable eanonieal form and the output equation is 

State-space methods for control system design 239 

---

Steady-state empirical models can be used for instrument calibration, process optimization, and specific instances of process control. Single-input, single-output (SISO) models typically consist of simple polynomials relating an output to an input. Dynamic empirical models can be employed to understand process behavior during upset conditions. They are also used to design control systems and to analyze their performance. Empirical dynamic models typically are low-order differential equations or transfer function models (e.g., first-or second-order model, perhaps with a time delay), with unspecified model parameters to be determined from experimental data. However, in some situations more complicated models are valuable in control system design, as discussed later in this chapter. 

In this rearrangement, xp is the process time constant, and Kd and Kp are the steady state gains.2 The denominators of the transfer functions are identical, they both are from the LHS of the differential equation—the characteristic polynomial that governs the inherent dynamic characteristic of the process. 

G( ) is the transfer function relating 0O and 0X. It can be seen from equation 7.18 that the use of deviation variables is not only physically relevant but also eliminates the necessity of considering initial conditions. Equation 7.19 is typical of transfer functions of first order systems in that the numerator consists of a constant and the denominator a first order polynomial in the Laplace transform parameter s. The numerator represents the steady-state relationship between the input 0O and the output 0 of the system and is termed the system steady-state gain. In this case the steady-state gain is unity as, in the steady state, the input and output are the same both physically and dimensionally (equation 7.16h). Note that the constant term in the denominator of G( ) must be written as unity in order to identify the coefficient of s as the system time constant and the numerator as the system... 

Equation 7.52 is the standard form of a second-order transfer function arising from the second-order differential equation representing the model of the process. Note that two parameters are now necessary to define the system, viz. r (the time constant) and (the damping coefficient). The steady-state gain KMT represents the steady-state relationship between the input to the system AP and the output of the system z (cf. equation 7.50). 

There are distinct similarities between second order systems and two first-order systems in series. However, in the latter case, it is possible physically to separate the two lags involved. This is not so with a true second order system and the mathematical representation of the latter always contains an acceleration term (i.e. a second-order differential of displacement with respect to time). A second-order transfer function can be separated theoretically into two first-order lags having the same time constant by factorising the denominator of the transfer function e.g. from equation 7.52, for a system with unit steady-state gain ... 

These two coupled equations involve the rates of ionization, recombination, and reverse transfer to the excited state. However, reverse transfer does not affect the shape of the distribution when ionization is under kinetic control. In such a case v exp(- t/x/j), p 0, and there is nothing to transfer back from p to v. Kinetic ionization creates a distribution identical in form to that of W/(r), regardless of the rate of reverse transfer. Hence, fo(r) may be affected only if ionization is under diffusional control. The maximal effect is expected to be at to = oo when there are stationary functions of distance ns(r) = limv, o.vv(.v) and ms r) = lim o p(.v), with a large dip in ns(r) near the contact and a hump in ms(r) at the same place (ns + ms = 1). 

To obtain steady-state step responses, one sets s = 0 in the transfer function. The steady-state matrix relating c to x is, from equation 4, DA-1 B + E. This matrix is labeled P and its elements p are listed in Table II. Referring to Figure 1, the equations represented bv the matrix P are... 

Equation (21.5) is the design equation for the dynamic feedforward controller and Figure 21.3b shows the resulting control mechanism. As can be seen from Figure 21.3a and b, the only difference between the steady-state and dynamic feedforward controllers for the tank heater is the transfer function (ts + 1) multiplying the set point. 

Many of the algebraic properties of the single-particle Green s function, in particular Dyson s equation, are transferable to two-particle propagators if they are constructed starting from an orthonormal set of primary states. In this paper we will construct propagators Q(u)) with the help of the extended... 

The space velocity, often used in the technical literature, is the total volumetric feed rate under normal conditions, F o(Nm /hr) per unit catalyst volume (m X that is, PbF o/W. It is related to the inverse of the space time W/F g used in this text (with W in kg cat. and F q in kmol A/hr). It is seen that, for the nominal space velocity of 13,800 (m /m cat. hr) and inlet temperatures between 224 and 274 C, two top temperatures correspond to one inlet temperature. Below 224 C no autothermal operation is possible. This is the blowout temperature. By the same reasoning used in relation with Fig. 11.5.e-2 it can be seen that points on the left branch of the curve correspond to the unstable, those on the right branch to the upper stable steady state. The optimum top temperature (425°C), leading to a maximum conversion for the given amount of catalyst, is marked with a cross. The difference between the optimum operating top temperature and the blowout temperature is only 5°C, so that severe control of perturbations is required. Baddour et al. also studied the dynamic behavior, starting from the transient continuity and energy equations [26]. The dynamic behavior was shown to be linear for perturbations in the inlet temperature smaller than 5°C, around the conditions of maximum production. Use of approximate transfer functions was very successful in the description of the dynamic behavior. 

Ngwompo and his co-authors [24] state that a LTI SISO system is structurally invertible if there is at least one causal path in the causal direct bond graph between the input variable and the output variable ([24, Proposition 1, p. 162]). Furthermore, they show how the state equations of the inverse system can be directly determined from a causal direct bond graph model or from a bicausal bond graph. (In order to support tasks such as bond graph-based system inversion, Gawthrop extended the concept of computational causality by introducing the notion of bicausality [19, 25].) Clearly, the state equations of the inverse model of a SISO system can be converted into a transfer function. 

Transfer functions can also be defined for the other thermodynamic state functions. Since enthalpy changes are often conveniently measurable, the transfer enthalpy, is perhaps the most widely used function. From the second law of thermodynamics, the transfer entropy function is given by Equation 6.9. 

Figure 11.6a shows a Newtonian fluid calculation of the amplitude ratios for the transfer functions of the final area relative to a variety of input disturbances for PET extruded at 295 °C from a 0.5-mm diameter spinneret at a velocity of 13.2 m/min and a draw ratio of 100. The cross-flow air had a velocity of 0.2 m/s and a temperature of 30 °C. Solidification was assumed to occur at 70 °C. Inertia and air drag were included in the steady-state and linearized equations, and perturbations were permitted in the temperature and velocity of the cross-flow air (r and Va,... 

Parametric models are more or less white box or first principle models. They consist of a set of equations that express a set of quantities as explicit functions of several independent variables, known as parameters . Parametric models need exact information about the inner stmcture and have a limited number of parameters. For instance, for the description of the dynamics, the order of the system should be known. Therefore, for these models, process knowledge is required. Examples are state space models and (pulse) transfer functions. Non-parametric models have many parameters and need little information about the inner stmcture. For instance, for the dynamics, only the relevant time horizon shoirld be known. By their stmcture, they are predictive by nature. These models are black box and can be constructed simply from experimental data. Examples are step and pulse response functions. 

In this chapter, we have introduced an important concept, the transfer function. It relates changes in a process output to changes in a process input and can be derived from a linear differential equation model using Laplace transformation methods. The transfer function contains key information about the steady-state and... 

As we have seen before, exact differentials correspond to the total differential of a state function, while inexact differentials are associated with quantities that are not state functions, but are path-dependent. Caratheodory proved a purely mathematical theorem, with no reference to physical systems, that establishes the condition for the existence of an integrating denominator for differential expressions of the form of equation (2.44). Called the Caratheodory theorem, it asserts that an integrating denominator exists for Pfaffian differentials, Sq, when there exist final states specified by ( V, ... x )j that are inaccessible from some initial state (.vj,.... v )in by a path for which Sq = 0. Such paths are called solution curves of the differential expression The connection from the purely mathematical realm to thermodynamic systems is established by recognizing that we can express the differential expressions for heat transfer during a reversible thermodynamic process, 6qrey as Pfaffian differentials of the form given by equation (2.44). Then, solution curves (for which Sqrev = 0) correspond to reversible adiabatic processes in which no heat is absorbed or released. 

A soluble gas is absorbed into a liquid with which it undergoes a second-order irreversible reaction. The process reaches a steady-state with the surface concentration of reacting material remaining constant at (.2ij and the depth of penetration of the reactant being small compared with the depth of liquid which can be regarded as infinite in extent. Derive the basic differential equation for the process and from this derive an expression for the concentration and mass transfer rate (moles per unit area and unit time) as a function of depth below the surface. Assume that mass transfer is by molecular diffusion. 

The low solubility of fullerene (Ceo) in common organic solvents such as THE, MeCN and DCM interferes with its functionalization, which is a key step for its synthetic applications. Solid state photochemistry is a powerful strategy for overcoming this difficulty. Thus a 1 1 mixture of Cgo and 9-methylanthra-cene (Equation 4.10, R = Me) exposed to a high-pressure mercury lamp gives the adduct 72 (R = Me) with 68% conversion [51]. No 9-methylanthracene dimers were detected. Anthracene does not react with Ceo under these conditions this has been correlated to its ionization potential which is lower than that of the 9-methyl derivative. This suggests that the Diels-Alder reaction proceeds via photo-induced electron transfer from 9-methylanthracene to the triplet excited state of Ceo-... 

Electron Nuclear Dynamics (48) departs from a variational form where the state vector is both explicitly and implicitly time-dependent. A coherent state formulation for electron and nuclear motion is given and the relevant parameters are determined as functions of time from the Euler equations that define the stationary point of the functional. Yngve and his group have currently implemented the method for a determinantal electronic wave function and products of wave packets for the nuclei in the limit of zero width, a "classical" limit. Results are coming forth protons on methane (49), diatoms in laser fields (50), protons on water (51), and charge transfer (52) between oxygen and protons. 

According to Eq. (1) the steady-state current across a micro-ITIES is proportional to the bulk concentration of the transferred species. Thus, the micro-ITIES can function as an amperometric ion-selective sensor. Similarly, the peak current in a linear sweep voltam-mogram of ion egress from the micropipette obeys the Randles-Sevcik equation. Both types of measurements can be useful for analysis of small samples [18a]. 

---