Abstract
The 10 MW High Temperature Gas-Cooled Reactor-Test Module (HTR-10) is the first High Temperature Gas-Cooled Reactor in China. With the objective of raising the reactor power from 30% to 100% rated power, the power ascension test was planned and performed in January 2003. The test results verified the practicability and validity of the HTR-10 power regulation methods. In this study, the power ascension process is preliminarily simulated using the THERMIX code. The code satisfactorily reproduces the reactor transient parameters, including the reactor power, the primary helium pressure, and the primary helium outlet temperature. Reactor internals temperatures are also calculated and compared with the test values recorded by a number of thermocouples. THERMIX correctly simulates the temperature variation tendency for different measuring points, with good to fair agreement between the calculated temperatures and the measured ones. Based on the comparison results, the THERMIX simulation capability for the HTR-10 dynamic characteristics during the power ascension process can be demonstrated. With respect to the reactor safety features, it is of utmost importance that the maximum fuel center temperature during the test process is always much lower than the fuel temperature limit of 1620°C.
1. Introduction
The 10 MW High Temperature Gas-Cooled Reactor-Test Module (HTR-10), located at the Institute of Nuclear and New Energy Technology (INET) of Tsinghua University, is the first High Temperature Gas-Cooled Reactor (HTGR) in China. As a pebble bed modular HTGR, the HTR-10 utilizes spherical fuel elements containing ceramic coated particles, whilst it adopts helium as coolant and graphite as moderator. Attaining the first criticality in December 2000, the HTR-10 experienced the power ascension process in January 2003, with the purpose of raising the reactor power from 30% to 100% rated power (RP). After the successful completion of this power ascension test, the HTR-10 achieved the 72 h full power operation with satisfactory technical specifications meeting the design requirements very well [1].
The HTR-10 itself is a complicated system characterized by multi-inputs, multi-outputs, and strong coupling, consequently making the power ascension a somewhat complex process which needs the close coordination of different power regulation methods and encounters the large-range change of operation parameters in both the primary and the secondary circuits. In light of the HTR-10 operation procedure, the strategy of proportionally increasing the secondary feed water mass flow rate, the primary helium mass flow rate, and the reactor power was adopted to complete the power ascension test. Through the test results, the practicability and validity of the HTR-10 power regulation means were fully demonstrated. Besides, transient test data of the reactor core and other components were obtained for the validation of codes and models employed in the design process.
Based on the actual test conditions, the HTR-10 power ascension process is preliminarily simulated using the THERMIX code. The simulation puts the emphasis on the primary circuit and reproduces the reactor transients very well. Important operation parameters, such as the reactor power, the primary pressure, and the internals temperatures, are compared with their measured values. Good to fair agreement between the calculation results and the test ones proves the THERMIX simulation capability for the HTR-10 dynamic characteristics during the power ascension process.
2. Power Ascension Process
The HTR-10 primary system in side-by-side arrangement essentially consists of a reactor, a steam generator, a hot gas duct, and a helium circulator, as shown in Figure 1, where the primary coolant flow direction is also illustrated. Some important design parameters are listed in Table 1 and the detailed design information is referred to elsewhere [2].
Since the HTR-10 steam generator is of once-through type and runs at a medium secondary pressure, water-steam two-phase flow instability may occur in the secondary circuit, especially when the feed water mass flow rate is less than 30% of its rated value [3]. Accordingly, two operation stages are considered in the HTR-10 operation procedure.
(1) Start-Up Operation Stage. At this stage, the reactor power is at 0–30% RP, while the primary helium mass flow rate keeps at 30% of its rated value, and so does the secondary feed water mass flow rate. At this time, the outlet steam cannot meet the requirements of the steam turbine. Alternatively, the steam is bypassed to provide process heat.
(2) Power Operation Stage. At this stage, the reactor is operated with a power level of 30%–100% RP. Meanwhile, the helium mass flow rate and the feed water mass flow rate are proportional to the reactor power:where , , and are the reactor power, the helium mass flow rate, and the feed water mass flow rate, respectively, while the subscript 0 denotes the rated value. When the HTR-10 is operated in this stage, it can be used for electricity generation because the outlet steam is of quality to drive the steam turbine.
At the HTR-10 power operation stage, main operation parameters as well as the relationship among them are briefly presented in Figure 2.
Based on the HTR-10 dynamic characteristics, three kinds of regulation methods are mainly adopted to change the operation parameters in both the primary and the secondary circuits, as listed below [4].
(1) Introducing Reactivity by Moving the Control Rod. As the first reactivity control system, 10 control rods are designed for the HTR-10. These control rods are symmetrically placed in their corresponding channels in the side reflector and each of them has the same amount of neutron absorber. Therefore, reactivity worth of every control rod can be considered identical. In the HTR-10 commissioning phase, reactivity worth measurement tests were carried out several times for the control rod system [5]. In the helium atmosphere, the integral reactivity worth of a single control rod is about 14 × 10−3 k/k, and the linear segment of the integral curve is in the rod position of 600–1500 mm, as shown in Figure 3. Through the control rod movement, reactivity can be introduced into the core, thus influencing the reactor power directly.
(2) Changing the Helium Mass Flow Rate by Adjusting the Helium Circulator Rotation Speed. The HTR-10 helium circulator, integrated into the upper part of the steam generator pressure vessel, is a vertical single-stage centrifugal compressor driven by an electric motor. Under normal operation, the helium coolant is pumped by the helium circulator and completes its circulation in the primary circuit after taking heat from the reactor core to the steam generator. Connected to the electric motor by a coupling shaft, the helium circulator is powered by a transducer. The rotation speed of the helium circulator is proportional to the output frequency of the transducer, and 1 Hz variation in the output frequency results in 60 rpm change in the rotation speed. Regarding the helium mass flow rate, it is proportional to the helium circulator rotation speed. The latter can be precisely controlled from 10% to 100% of its rated speed, so the helium mass flow rate can be adjusted to satisfy different operation conditions without any doubt.
(3) Changing the Feed Water Mass Flow Rate by Adjusting the Feed Water Pump Rotation Speed. In the HTR-10 secondary circuit, the feed water is driven by its pump to the steam generator, where medium-pressure, superheated steam is produced after the heat exchange between the primary helium and the secondary water. The power supply of the feed water pump is also controlled by a transducer, whose output frequency can be adjusted. Hence, the water mass flow rate can be proportionally regulated via the feed water pump rotation speed that is proportional to the transducer output frequency.
The objective of the power ascension test is to raise the reactor power from 30% RP to 100% RP via the coordination of different regulation methods, while the outlet steam temperature should be kept unchanged as far as possible for the sake of turbine generator stable operation. In addition, other parameters are not allowed to exceed their operation limits during the power ascension process. According to the HTR-10 dynamic characteristics, a specific test procedure was formulated for the power ascension process. And the main points of the procedure are depicted as follows:(1)The water and the helium mass flow rate are increased successively. Due to the internal negative temperature feedback mechanism, the reactor power will automatically rise to a higher level.(2)If the power increase is not sufficient to maintain the outlet steam temperature, then the control rod should be withdrawn for an appropriate interval to compensate the downtrend of the outlet steam temperature.(3)The above-mentioned actions should be repeated until the reactor power achieves 10 MW power level and the helium outlet temperature reaches 700°C.
Before the power ascension test, the HTR-10 was under normal operation with a power level of 30% RP, a primary helium pressure of 2.8 MPa, a helium inlet temperature of 212°C, and a helium outlet temperature of 675°C. In the power ascension process, the reactor power, the helium mass flow rate, and the water mass flow rate went up proportionally, and the outlet steam temperature always stayed at about 420°C. During the test, the reactor power is monitored by the nuclear measurement system [6]. This system is composed of three kinds of ex-core neutron flux instrumentation facilities covering the source range, the intermediate range, and the power range. Other key parameters are recorded by the thermal measurement system [7]. Due to the internal structure limitation, there is no proper flow passage accommodating the common flowmeters, for example, the orifice flowmeter. Thus, the helium mass flow rate is calculated by a specific formula as a function of the helium pressure, the cold helium temperature, and some related parameters of the helium circulator, such as the motor power, the pressure head, and the rotation speed. The cold and the hot helium temperatures are measured by class 1E thermocouples installed at the outlet and the inlet sections of the steam generator, respectively. The measuring points of hot helium temperature are in the downstream location of the hot helium plenum, so the coolant here has been adequately homogenized with high mixing degree [8]. Figure 4 shows the distribution of the temperature measuring points arranged in the reactor internals [9]. TR1~TR6 represent two symmetrical columns of thermocouples in the top internals. SR1~SR6 and SR7~SR12 are two rows of thermocouples in the side internals at two different heights. In the bottom reflector, BR1 and BR3, and BR2 and BR4 are installed in pairwise symmetrical positions. Around the fuel discharging tube there are also two symmetrical columns of measuring points labeled as FD1~FD6. Similarly, CB1~CB8 are located symmetrically in the bottom carbon brick.
3. Analysis Methods
The system analysis code THERMIX is applied for the simulation of the HTR-10 power ascension test. This code can analyze the thermal-hydraulic performance of pebble bed HTGRs under normal operation and accident conditions. As a modular software package, THERMIX mainly comprises analysis modules for neutron kinetics, solid heat conduction in reactor, gas convection in reactor, and fluid flow in primary circuit. Brief descriptions of these modules are presented as follows [10, 11].
3.1. Neutron Kinetics Module
In this module, nuclear characteristics are evaluated by a conventional point kinetics model with six groups of delayed neutrons. Fission power is calculated by a balance of feedback reactivity and external reactivity. The former results from the variation of fuel, moderator, and reflector temperatures as well as the change of xenon concentration, while the latter is caused by the movement of control rods. In addition, decay heat is estimated by kinetic equations of fission products.
3.2. Solid Heat Conduction Module
This module consists of a two-dimensional transient temperature model for solid materials and a one-dimensional transient temperature model for spherical fuel elements. It solves the time-dependent general heat conduction equation with temperature-dependent material properties. Main components of the HTR-10, such as the pebble bed core, the graphite reflectors, the carbon bricks, and the reactor pressure vessel, can be represented using different material compositions.
3.3. Gas Convection Module
In this module, a quasi-stationary gas flow model is used to simulate the complex flow conditions in a pebble bed HTGR. Coolant flow through the fuel elements is regarded as the flow in the homogeneous media. Coupling with a given time-dependent temperature profile of the solid structures, this module solves steady-state continuity, momentum, and energy equations of gas in the reactor. The HTR-10 gas convection model, which is a two-dimensional axisymmetrical one in (r, z) geometry, contains 19 different flow regions divided into 18 radial and 38 axial mesh points, as shown in Figure 5. The calculating model covers the main flow passages in the HTR-10, for example, the reactor core, the cold helium channels, the cold and the hot helium plena, and the control rod channels.
3.4. Primary Circuit Module
This module calculates pressure, temperature, and mass flow rate of coolant in the primary circuit. Using a quasi-stationary model composed of steady-state continuity, momentum, and energy equations of fluid, this module can model different components in the HTR-10 primary circuit, involving the hot gas duct, the steam generator, the helium circulator, and so forth.
Besides the past validation work, the THERMIX code is now further checked in INET against the test data from the HTR-10 with the aim of verifying its simulation capability for different scenarios of pebble bed HTGRs [12].
4. Simulation Results
Based on the initial operation parameters listed in Section 2 and the calculating models established in Section 3, the power ascension process of the HTR-10 is simulated using the THERMIX code. The preliminary simulation puts the emphasis on the primary circuit so that it decouples the steam generator model. As a result, the following inputs obtained from the test are adopted by the computation: (1) the primary helium mass flow rate; (2) the primary helium inlet temperature which is affected directly by the secondary operation parameters, for example, the water mass flow rate; (3) the control rod position history which is converted into positive reactivity introduced into the reactor core according to the reactivity worth curve given in Figure 3. The helium mass flow rate and the helium inlet temperature recorded during the test are given in Figure 6.
It was confirmed that the HTR-10 had achieved its equilibrium state before the power ascension test. After the initiation of the test, the reactor power was gradually raised up to 100% RP in accordance with the test procedure. When the reactor came to 50% RP and 75% RP, it stayed at those power levels for some time. The whole simulation lasts 12 h, including 3 h for the steady-state operation under 30% RP and 2.6 h for the full power operation.
As illustrated in Figure 7, the THERMIX code accurately reproduces the reactor power transient during the test process. Following the power ascension from 30% RP to 100% RP, the primary helium pressure concomitantly rises from 2.8 MPa to 2.9 MPa, as depicted in Figure 8, from which it can be seen that the calculated pressure corresponds very well with the test one. Moreover, the test result shows that the primary helium outlet temperature reaches 700°C when the reactor is operated under 100% RP, as shown in Figure 9. A difference of about 10°C exists between the calculation and the test. At present, such deviation is preliminarily attributed to the mass flow rate used as an input condition by the simulation, because the test value was derived from the indirect calculation rather than the direct measurement.
Temperature fields of the reactor under 30% RP and 100% RP are shown in Figure 10, with the broken lines indicating the representative configuration of the pebble bed core. For the two different power levels, some similar phenomena can be qualitatively observed from their corresponding temperature fields: (1) both the radial and the axial temperature gradients are greatly intensive in the core; (2) heat transport in the top and the bottom ceramic internals is mainly determined by the longitudinal heat transfer, while the one in the side ceramic internals is primarily dependent on the lateral heat transfer; (3) with regard to the axial temperature profiles in the pebble bed, the core temperature increases along the axial direction at first and then descends after achieving a maximum value; (4) at the core inlet, the radial temperature distribution is nearly flat, but in the lower part of the core the central zone is much hotter than the peripheral zone.
However, temperature values of the HTR-10 main components are raised up to different extent along with the power ascension. For example, temperatures of the core region, the side internals, and the bottom internals are in the ranges of 200–900, 200–600, and 200–800°C, respectively, at 30% RP. For the 100% RP case, the three temperature ranges are 300–1000, 200–700, and 200–900°C. Due to the combination of the axial power distribution and the helium flow direction, the hottest spots at the two power levels both appear at the points of cm and cm, which means the intersection of the core centerline and the bottom surface of the cylindrical part of the pebble bed. And the maximum core temperatures are 875 and 953°C, respectively.
As mentioned in Section 2, there are a number of thermocouples installed in the reactor internals. In this study, temperatures of the measuring points in the following components are calculated and compared with the measured values: the top internals, the side internals, the bottom reflector, and the fuel discharging tube. Since the two-dimensional axisymmetrical calculating models are used by the THERMIX code and the azimuthal temperature distribution cannot be taken into account at present, one calculated temperature will be compared with two measured values for the top internals, the bottom reflector, and fuel discharging tube where every two symmetrical thermocouples are in different circumferential positions but have the same radius and height.
Figure 11 shows the temperature transition of the top internals. During the test process, TR1 and TR4 both experience a temperature rise. THERMIX basically predicts such variation tendency, although it overestimates the temperature values of these two measuring points. For TR2 and TR5, the calculated temperature which almost keeps unchanged is higher than their test values from which a temperature increase can be observed. Likewise, measured temperatures of TR3 and TR6 show a higher rise than the corresponding code result. Based on the comparison, the maximum calculation deviation in the top internals is found to be 36°C that occurs at the measuring position of TR1.
Figure 12 presents the temperature transient in the upper part of the side internals. In this part, the calculated temperature of SR1 agrees very well with the test one. As regards the other five thermocouples, obviously the calculated temperature curves are all above the test ones. However, THERMIX still reproduces their general behaviors with a largest deviation of 46°C which appears at thermocouple SR5.
Comparatively speaking, more accurate simulation results are obtained at the measuring points of the lower part, as shown in Figure 13. Except for SR11, all the thermocouples get satisfactory prediction temperatures which are in good accordance with the measured ones. At the position of SR11, THERMIX generates the correct temperature change tendency and the largest discrepancy is 48°C.
Figure 14 gives the analysis and the test temperature transients of the bottom reflector. For the first pair of thermocouples, the calculated temperature curve is located between the test curve of BR1 and the one of BR3, so it is considered that the code result is pretty reasonable. In the bottom reflector, thermocouples are located below the core outlet channels through which hot helium with different high temperatures flows to the hot helium plenum for uniform mixing. Compared with BR1 and BR3, the location of BR2 and BR4 is closer to the core centerline and that makes them affected by the hotter helium from the core outlet. Thus, an underestimation can be seen from the comparison between the analysis and the test temperatures of BR2 and BR4 and the maximum deviation is 141°C. However, the calculated curve still reflects the actual temperature change tendency and becomes closer and closer to the measured curves during the test process.
The temperature variation of the fuel discharging tube is presented in Figure 15. Whether for FD1 and FD4 or for FD3 and FD6, sufficient agreement between the calculated temperature and the test ones is obtained. Underestimating the measured temperatures of FD2 and FD5, THERMIX simulates the general trend and gets a better result when the reactor is at 100% RP. At the beginning of the test, the maximum calculation deviation of FD2 and FD5 reaches 97°C.
With respect to the safety features of the HTR-10, the maximum fuel center temperature rises from 900 to 1020°C, and it does not exceed the limit value 1620°C all through the test time, as shown in Figure 16.
5. Conclusions
The power ascension test was conducted on the HTR-10 with the purpose of raising the reactor power from 30% to 100% rated power. The test results prove the practicability and validity of the three power regulation means of the HTR-10 and lead to the following main findings:(1)The strategy of proportionally raising the secondary feed water mass flow rate, the primary helium mass flow rate, and the reactor power facilitates the goal of the power ascension test.(2)Main operation parameters, such as the reactor power, the helium inlet/outlet temperatures, and the helium pressure, change smoothly during the whole test process. In the meantime, the outlet steam temperature keeps stable. As a consequence, none of the operation limits is exceeded.
The THERMIX code is used to simulate the HTR-10 power ascension process. The simulation reproduces the reactor transients very well, for example, the reactor power, the helium pressure, and the helium outlet temperature. Additionally, the reactor internals temperatures are calculated and compared with actual values recorded by the thermocouples. Generally speaking, THERMIX can simulate the general temperature change trend of different measuring points and the calculated temperatures show good to fair agreement with the test ones. Some calculation differences are larger (141°C for BR2 and BR4, 97°C for FD2 and FD5) in some small hot zones beyond the range of the model symmetry at the bottom of the reactor internals. Considering the measured temperatures that exceed 750°C, the relative deviations are not so great. On the basis of the above-mentioned comparisons, the THERMIX simulation capability for the HTR-10 dynamic characteristics during the power ascension process can be demonstrated.
With respect to the reactor safety features, it is of utmost importance that the maximum fuel center temperature during the test process is always much lower than 1620°C, which is the limit value of the HTR-10 fuel.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgment
This work has been supported by the Chinese National S&T Major Project (Grant no. ZX069).