Reactor Diagnosis

Reactor Diagnosis Equation (10-26) can provide a useful check on operating data. In the laboratory, in a pilot plant, or in a full-scale plant, the volumetric flow rate usually is fairly easy to set and/or measure. However, the volume of reactor that is actually fi lled by fluid is not always so easy to determine. Mechanical drawings that can be used to determine a value of V when the reactor was installed may (or may not) be available. Moreover, things can change over time. Consider the following example. 

A small, continuous polymoization reactor has been in service for about 5 years, during which time it has been started up and shut down frequently, sometimes according to established procedure and sometimes not. The original drawings show the volume of the reactor to be 500 gallons. 

The performance of the reactor appears to have deteriorated over time. There is some concern that solid polymer has built up in the reactor, reducing the volume in which the polymerization reaction takes place. Therefore, a tracer test was ran, as follows. 

Water was passed continuously through the reactor at a flow rate of 1000 gallons per hour. (Water is not a solvent for any polymer that might have accumulated.) When the flow of water was at steady state, a sharp pulse of tracer was injected right at the point where the water entered the reactor. The total amount of injected tracer was 100,000 units. The concentration of tracer was measured at the point where the water stream left the reactor. The results are shown in the following table. 

Time after injection (min) Tracer concentration in effluent (units/gal) Time after injection (min) Tracer concentration in effluent (units/gal) 

---

To illustrate the power of the object-oriented style of representation, consider the reactor diagnosis example used eadier in the discussion of rules. Assume that there are several reactors, R-101, R-102, etc, each served by a common cooling system. Relating coolant malfunctions to the temperature in each reactor would need multiple rules, or rules with multiple disjunctions. Instead, if rules are used in combination with the object representation described above, a single general rule can be written to cover all the specific instances, as follows. 

H. Schuler and C.U. Schmidt. Calorimetric state estimators for chemical reactor diagnosis and control reviews of methods and application. Chemical Engineering Science, 47(4) 899-915, 1992. 

The neutrons in a research reactor can be used for many types of scientific studies, including basic physics, radiological effects, fundamental biology, analysis of trace elements, material damage, and treatment of disease. Neutrons can also be dedicated to the production of nuclear weapons materials such as plutonium-239 from uranium-238 and tritium, H, from lithium-6. Alternatively, neutrons can be used to produce radioisotopes for medical diagnosis and treatment, for gamma irradiation sources, or for heat energy sources in space. 

Water as coolant in a nuclear reactor is rendered radioactive by neutron irradiation of corrosion products of materials used in reactor constmction. Key nucHdes and the half-Hves in addition to cobalt-60 are nickel-63 [13981 -37-8] (100 yr), niobium-94 [14681-63-1] (2.4 x 10 yr), and nickel-59 [14336-70-0] (7.6 x lO" yr). Occasionally small leaks in fuel rods allow fission products to enter the cooling water. Cleanup of the water results in LLW. Another source of waste is the residue from appHcations of radionucHdes in medical diagnosis, treatment, research, and industry. Many of these radionucHdes are produced in nuclear reactors, especially in Canada. 

C. Comprehensive Example 3 Diagnosis of Operating Problems in a Batch Polymer Reactor... 

Consider again a batch polymerization process where the process is characterized by the sequential execution of a number of steps that take place in the two reactors. These are steps such as initial reactor charge, titration, reaction initiation, polymerization, and transfer. Because much of the critical product quality information is available only at the end of a batch cycle, the data interpretation system has been designed for diagnosis at the end of a cycle. At the end of a particular run, the data are analyzed and the identification of any problems is translated into corrective actions that are implemented for the next cycle. The interpretations of interest include root causes having to do with process problems (e.g., contamination or transfer problems), equipment malfunctions (e.g., valve problems or instrument failures), and step execution problems (e.g., titration too fast or too much catalyst added). The output dimension of the process is large with more than 300 possible root causes. Additional detail on the diagnostic system can be found in Sravana (1994). 

The production of the various metallic radionuclides with utility to either diagnosis or treatment are described, separated by route of production (either cyclotron/accelerator or reactor). 

At present, the chief value of RTD studies is for the diagnosis of the performance of existing equipment, for instance maldistribution of catalyst in a packed reactor, or the presence of bypassing or stagnant zones in stirred tanks. No correlations have been achieved for cr2(tr) or njrriang in terms of operating conditions, and only limited correlations for De. 

J.P. Steyer, D. Roland, J.C. Bouvier, and R. Moletta. Hybrid fuzzy neural network for diagnosis - Application to the anaerobic treatment of wine distillery wastewater in a fluidized bed reactor. Wat. Sci. TechnoL, 36(6-7) 209-217,... 

There are many diagnosis and/or advisory systems under development, applied to geology, nuclear reactors, software debugging and use, manufacturing and related financial services. 

Biological Reactors. In this section I discuss some applications that are at least indirectly related to chemical science and engineering. The first example, illustrated in Figure 1, is derived from a simulation and diagnosis of a biological reactor that we put together for a demonstration. 

On the basis of these arguments, the chapter and the book concludes with a few suggestions for developing future research work in this field, for applying the methods presented in this book to real reactors, and for improving the proposed control and diagnosis strategies. 

The aim and the hope of the authors is to provide, through this book, a unitary perspective of the main problems and challenges related to modeling, control, and diagnosis of chemical batch reactors. A special emphasis is put on the interaction between the development of effective and reliable mathematical models of the plant and on the subsequent design of the control and diagnosis systems. Hence, the recommendation for the reader is to read this monograph as a whole. 

The model reduction procedure must be adapted to the use of the simplified models and to the availability of experimental data needed to evaluate the unknown parameters, as discussed in Chap. 3. In general, more complex models are used for the design of the reactor and for the simulation of the entire process, whereas more simplified models are best fit for feedback control. In the following chapters it is shown that fairly accurate results are obtained when a strongly simplified kinetic model is used for control and fault diagnosis purposes. 

After performing the kinetic analysis of the reacting system, the researchers possess suitable kinetic models of different complexity to be used to design and control the entire process. The more complex model should be used to design the reactor this subject is outside the purpose of this book and is only briefly considered in Sect. 7.4. On the contrary, in Chaps. 5 and 6 the kinetic model is used to design adaptive model-based control and fault diagnosis schemes for a class of reactions taking place in batch reactors. 

The literature focused on model-based FD presents a few applications of observers to chemical plants. In [10] an unknown input observer is adopted for a CSTR, while in [7] and [21] an Extended Kalman Filter is used in [9] and [28] Extended Kalman Filters are used for a distillation column and a CSTR, respectively in [45] a generalized Luenberger observer is presented in [24] a geometric approach for a class of nonlinear systems is presented and applied to a polymerization process in [38] a robust observer is used for sensor faults detection and isolation in chemical batch reactors, while in [37] the robust approach is compared with an adaptive observer for actuator fault diagnosis. 

---