THE ITER HEAT REJECTION CHALLENGE.

Transcription

1 THE ITER HEAT REJECTION CHALLENGE Steve Ployhar 1, Warren Curd 1, Ajith Kumar 2, Giovanni Dell Orco 1, Babulal Gopalapillai 1, Dinesh Gupta 2, Hiren Patel 2, Keun-Pack Chang 1, Fan Li 1, Fabio Somboli 1, Liliana Teodoros 1 1 ITER Organization, Route de Vinon sur Verdon, Saint Paul Lez Durance, France 2 ITER-India, Institute for Plasma Research, Gandhinagar, Gujarat , India steve.ployhar@iter.org ITER is an international fusion facility being built in France to demonstrate the scientific and technological feasibility of fusion power. Fusion power at ITER is generated using a Tokamak machine in which burning plasma at temperatures of 150,000,000 C is confined within a vacuum vessel by magnetic fields. The enormous amount of heat generated by the Tokamak and its auxiliary systems is removed by the cooling water systems, consisting of the Tokamak Cooling Water System (TCWS), the Component Cooling Water System (CCWS), the Chilled Water System (CHWS), and the Heat Rejection System (HRS). These systems are designed to remove an initial peak heat load of about 1100MW. ITER is an experimental facility that will operate in a cyclical fashion. High levels of fusion power will be generated during repeated plasma pulses with specified durations. Heat produced by the fusion reaction will not be used to generate electricity, but will be rejected to the environment. The cyclical nature of the ITER machine presents distinct challenges to the design of the HRS which must reject normal facility heat loads plus large, intermittent heat loads from Tokamak pulse operations, while maintaining stable and predictable cooling tower basin water temperatures to meet the needs of cooling water system clients. This paper explores these challenges to the HRS design and describes the selected solutions. I. THE HEAT REJECTION SYSTEM The ITER Heat Rejection System receives the heat produced by the Tokamak and the plant auxiliary systems and rejects it to the atmosphere using mechanical induced-draft wet cooling towers. Such cooling towers were chosen for ITER heat rejection since they are costeffective. Figure 1 is a conceptual schematic of the 2010 baseline design, showing how the Component Cooling Water System collects heat produced by the Tokamak and the many auxiliary systems and transfers it to the HRS 1. II. THE CHALLENGE ITER will experience about 30,000 plasma pulse cycles over an operating life of 20 years 2. Each pulse cycle consists of a burn phase and a dwell phase. During the burn phase, large quantities of heat are generated in the Tokamak and transferred via the TCWS and CCWS to the HRS. During the dwell phase, the amount of heat being produced decreases abruptly to about 10% of the amount produced during the burn phase. In contrast with a fission reactor, the amount of decay heat is very small. Heat produced by some of the plant auxiliary systems, such as the radio frequency heating systems, also decreases substantially during the dwell phase. Figure 2 illustrates the rates of heat generation during the burn and dwell phases of the pulse cycle for the principal modes of ITER operation 3. Large cooling towers are normally designed based on one set of process parameters and operate for extended periods with stable process conditions. For ITER there are effectively two very different sets of process conditions that change abruptly over fairly short time intervals. Table I lists the relevant cooling tower parameters for the burn and dwell phases for inductive operation, the initial plasma operating scenario. While it would be easy to size the cooling towers based on peak conditions this would be an unacceptable use of resources, both in terms of capital cost and space on the site. Additionally, it is desirable to maintain a stable cooling tower basin temperature during the pulse cycle. Clients depend on receiving cooling water at a predictable temperature. For example, the radio frequency transmission lines especially for the electron cyclotron heating and current drive system, must maintain a proper alignment to work efficiently and thermal expansion/contraction caused by varying cooling water temperatures can disturb this alignment. ITER s goal is to maintain a basin temperature that does not vary more than ±1 C during the pulse cycle. The basic challenge then, is how best to meet heat rejection needs while minimizing the cost and footprint of the cooling towers. TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN

2 Fig. 1. ITER Heat Rejection and Component Cooling Water Systems Configuration TABLE I Cooling Tower Parameters (2010 baseline) Parameter Value (burn phase) Value (dwell phase) Cooling Tower (CT) flow rate kg/s kg/s CT inlet water temperature 49.2 C ~28 C CT outlet water temperature 25.6 C 25.6 C Wet Bulb Temperature 22 C 22 C Dry Bulb Temperature 32 C 32 C CT approach 3.6 C 3.6 C CT range 23.6 C ~2.4 C Required Heat Rejection Rate ~1100 MW ~110 MW Required CT Area ~5500 m 2 ~1900 m 2 Fig. 2. Heat Rejection Profiles for ITER Reference Plasma Scenarios 108 TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN. 2012

3 I.A. The Principles of Cooling Tower Operation The heat produced by ITER is ultimately transferred to the atmosphere by the evaporation of water, taking advantage of water s large latent heat of vaporization to cool the remaining water. Evaporation of water is facilitated by increasing the surface area of the water (droplet size or film area), increasing the volume of air exposed to the water (cooling tower area and fan speed), and increasing the time in which the water is in contact with the air (cooling tower height). Figure 3 illustrates a conceptual cooling tower of the type proposed to be used for ITER, and associated nomenclature 4. The rate of heat removal from the water passing through the cooling tower is: Q cl( T ) 1 T2 (1) The water flow rate is not constant because of the evaporation of water and drift losses. However this mass is very small and is neglected here. Conservation of energy requires that the heat removed from the water as it passes through the cooling tower be equal to the heat absorbed by the air. 5 Thus, KaV L T1 dt h h ' T 2 where: K = air mass transfer coefficient (kgs air/hour-m 2 transfer area) a = specific transfer surface (m 2 of transfer area/m 3 of fill) V = total fill volume (m3) L = water flow rate (kg/hour) T 1 = water inlet temperature (ºC) T 2 = water outlet temperature (ºC) h = enthalpy of saturated air at water temp. (J/kg) h = enthalpy of air (J/kg) The right-hand side of the Merkel equation is expressed in terms of the properties of the water and air, so for a given set of design or operating conditions it is (3) Q cl T T ) G( h ) or L G ( h1 ( h2 h1 ) (2) c( T T ) 1 2 where, for equations (1) and (2): Q = heat removal rate (Joules/hour) L = water flow rate (kg/hour) T1 = water inlet temperature (ºC) T2 = water outlet temperature (ºC) c = heat capacity of water (J/kg- ºC) G = air flow rate (kg/hour) h1 = enthalpy of inlet air (J/kg) h2= enthalpy of outlet air (J/kg) For a given set of cooling tower operating conditions the right-hand side of equation (2) will be constant. Thus, the water/air ratio, L/G, must be constant. The air flow rate can be increased by increasing the area (footprint) of the cooling tower, or to some extent by increasing fan speed or blade angle. ITER does not currently plan to have variable air flow through the cooling tower. Cooling tower heat transfer can be represented mathematically by the Merkel equation. 6 Fig. 3. Cooling Tower Nomenclature TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN

4 constant. Reference 6 (Figure 12-12) provides a useful illustration of relationships between the air and water conditions for a counterflow cooling tower of the type planned for use at ITER, and. also shows the correlation between enthalpy of the air and the cooling tower range and approach temperatures. The left-hand side of the Merkel equation is expressed in terms of the parameters related to tower size and water flow rate and is referred to as the tower characteristic. During plasma operations ITER has two very different design conditions, those for the burn and dwell phases of the pulse cycle. Thus it must satisfy two very different tower characteristics. From the preceding equations and other references, three general principles regarding cooling tower size can be deduced. Other things being equal: Principle #1: Cooling tower size increases as the required rate of heat transfer increases Equation 1 represents the rate of heat removal from the water passing through the cooling tower. For a fixed range (T 1 -T 2 is constant), water flow must increase to increase the total rate of heat transfer. G is assumed to be constant (no variable air flow), so cooling tower size must increase. Since cooling towers are designed with fill characteristics and a height that allows the air to reach saturation at the outlet, increasing water flow means that the cooling tower area (footprint) must be increased. Principle #2: Cooling tower size increases as approach decreases Keeping cooling tower range constant but decreasing the approach causes the L/G ratio to decrease. This can be seen in Reference 6, Figure Since water flow is unchanged, G must increase. Additional air flow and cooling tower volume are required to cause the outlet water temperature to more closely approach the entering air wet-bulb temperature. The impact of changes in cooling tower approach on cooling tower area is quite significant. Figure 4 shows the relationship between approach and cooling tower area for both the burn and dwell phases of the plasma cycle, using Table I parameters. The representative cooling tower area data used to prepare this figure was obtained from an online cooling tower calculator offered by GEA Energy Technology GmbH. 7 It is apparent that even a small increase in approach dramatically reduces required cooling tower area. Principle #3: Cooling tower size increases as range decreases Referring again to equation (1) it is apparent that reducing the range will require a commensurate increase in water flow in order to maintain the same heat removal rate (maintaining the left side of the equation constant). Again, increasing water flow means that the cooling tower area must be increased. II.B. Objectives for Reducing Cooling Tower Size The cooling towers could be designed only for the burn phase of the pulse cycle, accepting the fact that they would be grossly oversized for dwell conditions and for other operating modes, such as short-term and long-term maintenance. But considering that the plasma burn periods represent less than 5% of ITER s operating life, it is well worthwhile to consider means of reducing required cooling tower size. To reach this goal, the following objectives have been identified, focusing on the three general principles described above. The bulleted items are possible means of achieving the objectives. 1. Reduce peak heat transfer rate by averaging the heat load over the pulse cycle Hold up the hot water, dilute it with colder water, and circulate it through the cooling tower during the full duration of the pulse cycle, at a temperature closer to the average temperature. Split the flow returning from CCWS-1 (carrying heat from the Tokamak) and slow part of it down so that the temperature of the flow reaching the HRS is self-averaged over the pulse cycle. Fig. 4. Cooling Tower Area vs. Approach 110 TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN. 2012

5 2. Increase cooling tower approach temperature Reduce the design wet-bulb temperature during which pulse operations are performed, allowing a reduction in plant availability during the summer. Increase basin temperature (and cooling water supply temperature to clients) 3. Avoid operation with very high and very low cooling tower ranges Hold up the hot water, dilute it with colder water, and circulate it through the cooling tower during the full duration of the pulse cycle, at a temperature closer to the average temperature. Increase the range during the dwell by minimizing cold water dilution III. THE PROPOSED SOLUTION Following evaluation of many alternatives, the following design solution has been proposed 8. It includes features that address all three objectives described above. The principal features are: Feature #1: Addition of a hot basin and recirculation pumps. The area of the cooling tower basin is enlarged and partitioned into a hot basin and a cold basin. New vertical pumps take suction from the hot basin and discharge to the cooling towers at a constant rate. The cooled water collects in the cold basin. Feature #2: Introduction of variable speed drives for the pumps which circulate water from the cold basin to the hot basin via the CCWS-1 heat exchangers (removing heat generated in the Tokamak). Feature #3: Increased cooling tower approach, achieved by increasing peak summertime basin temperature. The combination of Features 1 and 2 is effective because it reduces the peak heat transfer rate and associated water flow rate (general principle #1) by accumulating some of the hot water in the hot basin during the burn phase of the pulse cycle. By reducing HRS flow through the CCWS-1 heat exchangers to about 40% during the dwell phase, when heat load is minimal and the water is cold, the temperature of the hot basin is allowed to vary between about 40º and 50º during the pulse cycle. This mitigates the extreme variation in cooling tower range and avoids operation at very low range (general principle #3). Feature #3 is very effective at decreasing cooling tower size. Even a small increase in approach results in a significant decrease in cooling tower area (general principle #2). The revised configuration of CCWS and HRS incorporating these proposed design features is illustrated in Figure 5. Table II lists the relevant cooling tower parameters. Fig. 5. Proposed Configuration for CCWS and HRS TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN

6 TABLE II Cooling Tower Parameters (proposed) Parameter Value (average) Cooling Tower (CT) flow rate ~6517 kg/s CT inlet water temperature 44.5 C CT outlet water temperature 27 C Wet Bulb Temperature 22 C Dry Bulb Temperature 32 C CT approach temperature 5 C CT range 17.5 C CT Heat Rejection Capacity ~480 MW Required CT Area ~2500 m 2 6. R. PERRY, Perry's Chemical Engineers' Handbook, 6 th edition, pp gem/en/calculators/ct_calculator.html 8. H. PATEL, Feasibility Report on ITER Heat Rejection System, version 1.0 (draft April 20, 2011) IV. CONCLUSION Actual cooling tower size configuration will be determined by the cooling tower designer. But the results of these proposed design changes are expected to result in a total cooling tower area about 50% less than might otherwise be required. ACKNOWLEDGMENTS This report is based on work undertaken within the framework of the ITER Project and supported by the ITER Organization and/or its Members, i.e. China, European Union, India, Japan, Korea, Russia and the United States of America. Dissemination of information contained in this paper is governed by the applicable terms of the ITER Agreement. Disclaimer: The views and opinions expressed herein do not necessarily reflect those of the ITER Organization, its members or any agency thereof. REFERENCES 1. W. CURD et al, CCWS-1 Process Flow Diagram version 2.1 (2011), CCWS-2 Process Flow Diagram version 2.1 (2011), HRS Process Flow Diagram version 1.0 (2010) 2. S. CHIOCCHIO et al, ITER Project Requirements, version 4.6 (2010) 3. S. PLOYHAR, CCWS Heat Loads for Tokamak Operation version 2.1 (2011) 4. Bureau of Energy Efficiency Book, Chapter 7, Cooling Towers, Ministry of Power, Gov t of India 5. S. LEEPER, Wet Cooling Towers: Rule-of- Thumb U.S. Department of Energy July TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY VOL. 61 JAN. 2012