Introduction

12.197. During recent years, the nuclear industry, together with NRC, has been implementing plans to manage severe accidents. By accident management is meant the innovative use of various resources, systems, and actions to prevent or mitigate a severe accident. Although such accidents are “beyond design basis” and would involve some degree of core melting, an understanding of the various possible accident mechanisms can permit the development of strategies to control the release of radionuclides to the environment. For example, the Three Mile Island accident could have been prevented by proper operator action, but once it progressed, it could have been better managed if needed strategies had been developed previously.

12.198. A severe accident management program consists of several dif­ferent elements.

1. Information and analysis

2. Supporting instrumentation

3. Accident management strategy development

4. Equipment modification

5. Personnel training

These are interrelated and may be expressed somewhat differently. Our primary purpose here is to identify the features needed for a severe accident program.

8.25. As the power of computers increased over the years, their use to accomplish so-called “thinking” processes which simulate certain human activities became the subject of a subdiscipline known as artificial intelli­gence (AI). Applications have varied from simple game playing (chess) to speech recognition. Normally required are a bank of stored knowledge and a search procedure in accordance with specified rules that lead to a decision. The boundaries of these efforts are not well defined. For example, many optimization methods follow a similar approach, although optimi­zation is not normally considered an AI procedure. Our initial attention will be focused on “expert systems,” a so-called single-purpose AI pro­cedure that is now widely used commercially [5].

8.26. An expert system is a computer program that incorporates the knowledge and reasoning of human experts as a decision tool. A typical expert system includes a knowledge base, and inference engine, and a system — human interface. The knowledge base includes available relevant infor­mation from data bases as well as various rules-of-thumb facts developed from experience by so-called “heuristic” processes. The inference engine manipulates the knowledge base information by programmed rules of logic to reach the desired conclusions. Translation of input into computer lan­guage and then from the computed results to output is accomplished by the interface feature.

8.27. The major advantage of this type of program is the ability to apply the wisdom of experts systematically in a repeatable manner to the solution of a problem without the experts being present. Prescribed logic paths for decision making are incorporated in the inference engine. “Reasoning” techniques may be forward “chaining” starting from a set of known facts or backward chaining in which the system works backward from goals and tries to establish needed supporting evidence. Common features are “if — then,” “and,” and “or” decision gates. Inprecise or “fuzzy” information may be incorporated into the program through the use of “certainty fac­tors,” which represent judgments of the validity of a fact on a numerical scale of 0 to 1, where 1 means complete certainty.

8.28. In nuclear reactor control rooms, expert systems can provide the operator with guidance regarding measures to be taken in the event of unanticipated incidents, thus reducing operator error. However, the ap­plicability of an expert system is limited by the scope of information in the knowledge base. Therefore, development of expert systems to cover a very wide range of emergency situations is taking place. Other uses include the monitoring of the performance of various plant components and recom­mending operating changes; various training situations, including the analy­sis of trainee performance; and many management-related applications.

8.29. Neural network procedures utilize a type of computer architecture in which the processing elements are interconnected so that a great many calculations can be carried out in parallel, imitating the way that neuron cells in the brain process information. The theory and descriptions of some neural network applications are available in a number of references [6]. In general, neural networks consist of processing units, arranged in layers, with connections between units. These connections have “weights” which have a memory capability and may be “trained” to adjust to changing conditions. Information is processed in a distributed manner rather than by a conventional computer serial approach. Hence a general characteristic is the ability of a network quickly to recognize the various conditions or states of a complex system once it has been suitably “trained.” Thus, applications in pattern recognition, language translation, and speech under­standing are typical.

8.30. Applications to the operation of nuclear power plants are being developed. For example, efficient pattern recognition can be a useful di­agnostic tool in interpreting control room data. Another area is the mod­eling of nonlinear systems with application to process dynamics [7].

9.94. In practical reactor systems the coolant is not stationary, and the boiling which takes place, called flow boiling, is hydrodynamically quite different from pool boiling. Flow boiling commonly occurs under forced- convection conditions, as in boiling-water reactors (BWRs) and to some extent in pressurized-water reactors (PWRs). It can also be experienced, however, when there is natural circulation in a loop configuration, such as may be present during the transient conditions that arise when a coolant circulation pump fails.

9.95. Suppose that water, below the saturation temperature, is forced through a channel between or around the solid fuel elements of a reactor; heat is then transferred from the solid surface (or wall) to the water. As long as the fuel-wall temperature, which increases along its length (§9.143 eq seq.), remains below the steam saturation temperature, single-phase heat transfer only will occur. In PWRs, the pressure on the cooling system is increased in order to raise the saturation temperature and thus prevent bulk boiling, but some local boiling is tolerated; PWRs and BWRs may therefore be regarded as having certain features in common.

9.96. The flow patterns in boiling two-phase flow are complex. Primarily for conceptual purposes, we show the classical Fig. 9.14 representation for flow boiling in a vertical hollow tube [10]. At first, the rising flowing water is merely heated by convective transfer with no boiling occurring. As the temperature of the flowing water increases, a point is reached where the temperature of the water in the slowly moving laminar layer next to the wall is above the saturation temperature although the temperature at the same level in the more rapidly moving core is still below saturation. Initial vaporization therefore occurs along the wall under subcooled (local boiling) conditions. Bubbles grow and are carried along in the superheated layer close to the wall, but they condense on being mixed with the subcooled liquid core.

9.97. After the central core reaches saturation conditions, the vapor bubbles being fed from the superheated layer near the surface no longer collapse but are carried along in the stream. As boiling progresses, the

TEMP. QUALITY

Fig. 9.14. Regimes of two-phase flow.

flow mechanism becomes quite complex and depends on how the vapor and liquid phases flowing in the same direction distribute themselves. As the vapor volume increases, the flow must accelerate. Bubble flow is char­acterized by a regular distribution of vapor bubbles in the continuous liquid
phase. With increased vaporization, and a resulting higher void fraction or quality (§9.99), a transition to annular flow occurs. This is characterized by a continuous vapor phase in the central core in which some liquid drops may be dispersed.

10.22. The terms nuclear fuel management and in-core fuel management are often used interchangeably. However, it is useful to refer to those decisions that relate to multioperating cycle planning as nuclear fuel man­agement. This might also include strategic decisions associated with pre­reactor fuel operations. For example, planned cycle length is dependent on utility requirements, while feed enrichment and feed batch size are cycle dependent. Thus, there is a necessary coupling between the larger multi­cycle, and often multigenerating unit picture and the smaller single-cycle design requirements. For our purposes, the term in-core fuel management refers to the design decisions for a single burnup cycle, which are concerned primarily with specifying the loading pattern and control strategies within the constraints given by plant design limits and technical specifications.

11.4. The most important contaminants from fossil-fueled power plant operation are carbon dioxide, sulfur dioxide, nitrogen oxides, and partic­ulate matter. Combustion processes contribute about 10 percent of the carbon dioxide emitted to the atmosphere, the remainder being primarily the result of natural decay. The carbon dioxide concentration in the at­mosphere is the result of the balance between generation and removal processes such as photosynthesis. However, as will be discussed in the next section, increasing carbon dioxide concentration levels could affect the earth’s climate as a result of the greenhouse effect.

11.5. Power plants contribute the major proportion of sulfur dioxide emitted to the atmosphere, but industrial sources can be very significant in some areas. Local concentrations vary widely and tend to be highest in urban industrial regions. High concentrations on the order of 10 parts per million (ppm) can cause breathing problems.

11.6. Only about 4 percent of the total emission of nitrogen oxides is caused by power plants. However, when carried along by wind currents, sulfur compounds and nitrogen oxides combine with the water vapor in the atmosphere to form sulfuric and nitric acids as aerosols which precip­itate back to earth with rain or snow long distances from the point of origin. This phenomenon, known as acid rain, has become a serious environmental problem. Therefore, flue-gas desulfurization systems are now required for new fossil-fueled power plants to minimize sulfur emission. These systems are expensive, generate a great deal of solid waste requiring disposal, and require significant maintenance.

12.17. The term redundancy refers to the use of two or more similar systems in parallel, so that the failure of one will not affect the plant operation. Redundant components and systems are commonly employed in nuclear power plants. They are of special importance in systems, such as instrumentation, shutdown controls, and emergency cooling, upon which safety depends.

12.18. The components of a redundant system could all be rendered inoperative by a common-mode failure, that is, when one failure leads to another or a number of failures result from a single cause. For example, duplicate components of the same design could conceivably fail simulta­neously when subjected to the same stress. Many potential common-mode failures can be foreseen and appropriate steps taken to circumvent them. In some cases, however, they are unpredictable and are revealed only after they have occurred. One way to minimize common-mode failure is by diversity, that is, by the use of two or more independent and dif­ferent methods for achieving the same result, e. g., reactor shutdown in an emergency.

12.19. The electric power supply system for a nuclear plant provides an illustration of redundancy and diversity. Instruments are operated by direct current which is available from two independent storage batteries. The alternating current required to operate pumps, valves, and air blowers is normally supplied by the plant’s generator connected to two separate bus­bar sets. In addition, two independent offsite power sources are available for use when the plant is shut down or the generator is not operating. If all the onsite and offsite electrical power should fail (“station blackout”), alternating current would be supplied by onsite diesel generators. The plant might then be without power for half a minute; this fact is taken into consideration in determining the conditions for safe operation.

12.118. Antimony remains in elemental form upon being released from failed fuel elements and tends not to react with steam or hydrogen in the cooling system. Since its boiling point is 1380°C, it will vaporize under accident conditions and deposit on system surfaces. However, it can alloy with other metals, which would tend to reduce its vapor pressure.

Alkaline earths

12.119. Strontium and barium are released from failed fuel rods as oxides and could react in the cooling system with steam to form hydroxides
which are moderately soluble in water. The presence of carbonates in emergency cooling water would result in precipitation of the alkaline earths from the hydroxides, thus preventing their release from the containment to the environment.

Other fission product groups

12.120. Noble metals, rare earths, and refractory oxides are generally chemically inactive. However, upon possible release into the containment during a severe accident, they could be transported as fine particles such as aerosols.

9.20. In most reactor designs, the power output is limited by the max­imum permissible temperature of the fuel elements, i. e., by the value of Pmax. In some cases, it is possible to increase the power output by decreasing Ртах/Рау. In other words, the power output of the reactor can be increased by a more uniform distribution of the power density. This is sometimes referred to as a “flattening” of the power distribution or of the neutron flux.

9.21. We have seen that the presence of the reflector provides some flux flattening (§3.136). In addition, reduction in the concentration of fissile atoms in fuel that would otherwise have a high neutron flux could provide a flattening effect. Such fuel management schemes, described in Chapter 10, may also provide improved utilization of the available neutrons and decrease the cost of the electricity produced by the reactor plant.

9.125. In the design of the first PWRs, local (nucleate) boiling was considered to be undesirable, and conditions were chosen to prevent its occurrence. In later designs, however, local boiling during steady-state operation became a design requirement. Local boiling will affect the pres­sure drop and, in an open lattice, will result in flow redistribution with the possibility of burnout. An underestimating of the factors affecting the pressure drop is therefore essential.

9.126. Where the parallel flow channels have the same inlet and outlet headers, the pressure drop must be equal in all channels. If there is a tendency in any channel for the resistance per unit flow rate to be higher, because of local boiling or other factors, the flow rate in that channel will automatically decrease until the pressure drop is the same as that in the other channels. Such flow distribution may result in instabilities because the decreased flow will mean a lower thermal transport and, consequently, higher surface and fluid temperatures. This could lead to ultimate failure

if the higher temperatures resulted in bulk boiling, accompanied by in­creased flow resistance, and so on (cf. §9.98).

9.127. It has been found [17] that the pressure drop for a nonboiling, open, parallel-tube system may be predicted by means of the standard Fanning friction-factor correlation based on an equivalent diameter. Under nonboiling conditions, there are only minor variations caused by differences in bulk-water temperatures. When local boiling occurs, however, there is an increase in momentum exchange and in heat transfer which may be due to the increase in hydrodynamic turbulence caused by repeated growth and collapse of steam bubbles. The bubbles may also be considered, in a broad sense, to contribute an increased surface roughness. Whatever the mechanism, there is an increase in pressure drop, and this may be repre­sented by a friction factor applicable to the local-boiling conditions.

9.128. The friction factor/for local boiling at 13.8 MPa (2000 psi) may

be expressed empirically in terms of the isothermal (nonboiling) factor /iso as a function of the bulk temperature tm alone; thus,

where the bulk temperature is between 293 and 335°C (saturation tem­perature of water) and above the local-boiling temperature. The pressure drop is equal to that for nonboiling until the water temperature reaches 335°C. Alternative procedures for local-boiling, pressure-drop calculations have been based on two-phase flow predictions.

10.58. Boiling-water reactors are operated so that the fuel economy is enhanced by shifting the neutron spectrum to control reactivity. During the first half, or so, of the operating cycle, the coolant flow rate is reduced so that the core coolant void fraction is relatively high. Thus, the core is somewhat undermoderated. The resulting increased resonance region neu­tron flux results in the increased conversion of uranium-238 to fissile plu­tonium. Since this undermoderation acts in the direction of reducing the reactivity at the beginning of the operating cycle, when compensation for the reactivity introduced by fresh fuel is needed, the neutrons used for conversion would otherwise be lost by absorption in control poisons.

10.59. As reactor operation proceeds, fuel is consumed and fission prod­uct poisons accumulate. The coolant flow is then increased, thus reducing the core coolant void fraction and softening the neutron spectrum. This acts in the direction of increasing the reactivity, as necessary to maintain operation. The fissile atoms formed by the previous conversion process now contribute to energy production.