- Email: [email protected]
Experimental analysis of dummy load prototype for ITER coil power supply system
Fusion Engineering and Design 146 (2019) 1064–1068
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Experimental analysis of dummy load prototype for ITER coil power supply system
T
Li Chuana, , Song Zhiquanb, Luo Bina, Li Jiaweia, Qin Xiuqic, Zhu Guangliangd, Fu Pengb, Zhang Xiuqingb, Li Huab ⁎
a
International Joint Research Laboratory of Magnetic Confinement Fusion and Plasma Physics, State Key Laboratory of Advanced Electromagnetic Engineering and Technology, School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China b Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, 230031, China c Anhui Electric Power Design Institute, Hefei, 230601, China d Hefei Keye Electrophysical Equipment Manufacturing Co., Ltd, Hefei, 230031 China
ARTICLE INFO
ABSTRACT
Keywords: ITER Dummy load Large inductance Experimental analysis Finite element analysis
This paper basically introduces the experimental analysis of the dummy load prototype, which aims to verify the capability of the ITER magnetic power supply systems to operate at their rated power levels without energizing the superconducting coils. The rated inductance of dummy load is 6.73 mH and the pulse test currents are 45 kA, 55 kA and 68 kA. According to the requirements of the large inductance and different pulse test current classes, a dry-type air-core water-cooling prototype with epoxy resin casting technique has been fabricated, which has 24 layers and 72 turns. The experimental analyses, including the consistency test, inductance test, and temperature rise test are described in detail. The consistency of different coil layer is ensured by comparing the voltage wave under the same voltage excitation. The inductance test includes dc inductance test by using the first order circuit method and ac inductance test by LCR instrument and finite element simulation analysis. The temperature rise test results under different current classes are introduced by comparing with the simulation analyses. The above experimental results are coincided with the requirements, which shows the feasibility and availability of the prototype.
1. Introduction ITER (International Thermonuclear Experimental Reactor) is an international nuclear fusion project to realize an experimental fusion reactor based on the concept of a tokamak. The parameters and operational modes of power systems are unique and depend on the various magnetic coils, which play a vital role in the plasma shape and position control. As an inductive component, dummy load is necessary and basically aiming to verify the capability of the various power supply systems to operate at rated power without charging the superconducting coils [1–3]. Thus, the function and performance of components, including converters, switching networks, discharge circuits in the power systems can be tested. Based on the prescribed functions and operational modes, the main requirements of dummy load are shown in Table 1. The rated DC inductance is 6.73 mH with the precision ± 10%. Considering the actual fabrication difficulties, the final precision ( ± 10%) is lower than that ( ± 1%) mentioned in previous work [4,5]. And this alteration has been
⁎
approved by ITER organization. The rated direct current is 7.75 kA. The currents like 45 kA, 55 kA and 68 kA are the typical values of central solenoid (CS) coils, poloidal field (PF) coils and toroidal field (TF) coils, respectively. Thus, the dummy load needs to operate with various modes owing to various testing requirements. The structure performance and thermal stability are crucial parts of dummy load, especially under the most severe pulse mode, i.e. 68 kA with 15 s, then operating with zero-current for 900 s, which is discussed in this paper. To meet the severe requirements in the thermal and dynamic performance under various operational modes, a structure of dry-type aircore water-cooling reactor with epoxy resin casting technique is adopted. A full prototype has 72 turns in total and contains eight parts in series and each part has three nine-turn pipes in parallel to carry more current, with same average diameter of 1.924 m. The full model and actual prototype are shown in Fig. 1. The whole procedures of dummy load mainly include four phases, primary treatment, winding and inspection, vacuum pressure impregnation (VPI), and assembly. VPI can enhance the insulation and mechanical
Corresponding author. E-mail address: [email protected] (C. Li).
https://doi.org/10.1016/j.fusengdes.2019.02.004 Received 8 October 2018; Received in revised form 27 January 2019; Accepted 2 February 2019 Available online 19 February 2019 0920-3796/ © 2019 Elsevier B.V. All rights reserved.
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Table 1 Main Requirements For Dummy Load. Parameters
Values
Rated DC inductance, mH Rated continuous current, kA Operating ambient temperature, °C Pulse operational modes: current, duty cycle
6.73 ± 10% 7.75 −25∼45 45 kA, 30 s on/900 off; 55 kA, 20 s on/900 s off; 68 kA, 15 s on/900 s off AC, 12/28/75; DC, 12 Demineralized water Cylinder, Aluminium Outdoor
Insulation level, Nominal /testing /impact, kV Cooling medium Shape, conductor material Location
Fig. 3. Schematic diagram of area difference ratio.
circuit. By analyzing and comparing the oscillating voltage wave between the terminals of measured piece (like coil layer) under the same pulse voltage excitation, insulation defect and coil difference can be easily detected. Various methods can be used to detect coil difference, such as by calculating the area difference ratio of standard wave and measured wave, which is defined as
=
Smeasured Sstandard × 100% Sstandard
(1)
Sstandard is the area enclosed by time axis and standard voltage wave of qualified piece. Smeasured is defined in the similar way. Fig. 3 shows the method adopted to measure the sub-layer coils of dummy load. The ± 2% compared to standard wave is the eligibility criteria by calculating the area difference ratio in the 1.5 periods of voltage wave starting with the second peak. According to the experimental records, the maximum difference is -1.4%. And all the sub-layer coils have passed the test.
Fig. 1. Model and actual prototype of dummy load ((a) main body, (b)distributive pipe, (c) collecting pipe, (d) vibration damper plate, (e)support components, (f)insulator).
performance [6]. Different with the previous work in reference [4,5], which focuses on the structural, thermal, electromagnetic environment and seismic analysis of dummy load prototype, this manuscript mainly introduces the experimental methods and results in detail, focusing on the consistency test, inductance test, and temperature rise test of dummy load prototype.
3. Inductance test 3.1. DC inductance As an inductive component, dummy load plays a crucial role in limiting the rate of current variation. Nevertheless, the dc inductance test with high precision, especially for air coil, is hard to realize due to the high sensitivity of inductance to the surrounding ferromagnetic material, and the calibration of additional inductance caused by test circuit. The lowest test frequency of existing series of LCR tester is generally 20 Hz, which still can’t regard as dc inductance test. The brief schematic diagram of dc inductance test of dummy load is shown in Fig. 4. The SCCPS has the ability of supplying a constant current up to 300 A to the measured piece, which is dummy load here. The power supply charges the dummy load to a certain current until flattop, then close the contactor to short and turn of the SCCPS. The current of dummy load freewheels through the closed contactor. Since the test works at a low power state, which is less than the short-circuit capacity of power supply, so the impact of short test can be ignored.
2. Consistency test Factors such as winding materials, magnetic materials, skeletons, processing techniques, etc., can cause insulation performance degradation between coil layers, and turns of coil products (such as transformers, motors, etc.). Therefore, it is extremely important to test the electrical properties without damaging the device under test. Pulse insulation tester Tonghui TH2882AS-5 adopted in this paper, not only can judge the insulation quality of the coil in a short time, but also can detect the individual differences of multiple coils, which is essential for quality of sub-layer coil with the same size [7]. Fig. 2 shows the schematic diagram of insulation tester, which mainly consists of a dc high voltage generator with adjustable voltage (maximum voltage 5 kV), a switch to trigger the test, and interior charge capacitance (C1). The stray capacitance of the circuit equalizes to C2. When connected to an inductive load, it becomes a second order
Fig. 4. Schematic diagram of dc inductance test. (SCCPS-superconducting constant current power supply, DL-dummy load).
Fig. 2. Schematic diagram of impulse insulation tester. 1065
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Table 2 Typical Inductance Value From Various Methods. Freq. (Hz)
Design value/mH
Measurement 1 (mH)
Measurement 2 (mH)
Measurement 3 (mH)
0 20 1000
6.92 6.47 5.47
– 6.30 5.34
– 6.12 5.45
7.55 6.20 5.21
* Dummy load in measurement 1 and 2 are assembled without resin-cast, so DC inductance cannot be measured by the method mentioned in Part 3.1. Thus, it is filled with dash. The design value at zero Hz is 6.92 mH owing to the consideration of some margin.
Due to the skin effect and proximity effect, the ac inductance declines with the frequency, and the inductance at 1 kHz only accounts for 79% of that at 0 Hz (i.e. dc inductance) according to the simulation results. Comparisons among three measure results show that the inductance off the ground is more close to the design value than that near the ground, which demonstrates the influence mentioned above. The reasons for the fact that the relative deviation 3 is a little higher may be the enhancement of capacitance effect, since the resin for casting increases the permittivity of materials between the layers. The measured dc inductance is 7.55 mH according to the first order circuit method, which is 9.1% higher than the corresponding design value, owing to the possibility that the measured circuit resistance is higher than the actual value and the measured inductance is larger than that of dummy load. The ac inductance relative deviation is within 5% below 1 kHz. Since the measured ac inductance at 20 Hz is 6.2 mH with a relative deviation -4.17%, the actual dc inductance can be reasonably inferred to be between 6.2 mH and 7.55 mH (actual value should be less than this value), which is acceptable for the actual debugging test.
Fig. 5. Declining current wave at 63 A with time constant 0.86 s.
The circuit is equivalent to a first order circuit (LR). The time constant (τ = L/R) depends on the inductance and resistance of circuit. The former approximates the inductance of dummy load. The inductance of test wires can be calibrated and is relatively small to that of dummy load. The latter consists of the resistance of dummy load, wires and closed contactor. Time constant can be deduced from the recorded wave of declining current as shown in Fig. 5. The dc inductance is 7.55 mH, according to the time constant 0.86 s and circuit resistance 8.78 mΩ as an average value of multiple tests. 3.2. AC inductance Dummy load consists of nine layer in radials direction and eight layer in axial direction. To enhance the thermal stability under pulse current like 68 kA and improve the current-carrying capability, three aluminium conductors with the size 40 mm × 50 mm are connected in parallel, which intensifies skin effect and proximity effect on the ac inductance, especially for this large equivalent section area. One design value based on finite element analysis and three measurement results with relative deviation compared to design value are described in Fig. 6 and Table 2, which also shows the simulation and measured results of ac inductance under several typical frequencies from 20 Hz to 1 kHz. The measurement 1 shows the test results of dummy load before resin casting and located on the ground, which means that the test may be affected by the ferromagnetic material, such as rebar to enhance the structural strength in the concrete. In the measurement 2, the dummy load is hung up to eliminate the influence of surrounding materials. The measurement 3 indicates the actual results of final assembled dummy load.
4. Temperature rise test 4.1. Steady model (8.02 kA) The relevant experimental conditions are as follows: water flow is 24.6 m3/h (i.e. water velocity is 0.58 m/s in each pipe), dc current RMS is 8.02 kA (a little higher than the rated value 7.75 kA), the initial temperature of input water and ambient are 17.2 °C and 19.0 °C (approx. value), respectively. The steady temperature test results are shown in Fig. 7. The steady state with temperature rise changing within 1.0 °C has lasted more than 4 h. It is worth mentioning that the water temperature at input in the simulation is set to be a constant value (17.2 °C), whereas that of test is changeable due to the difficulty in maintaining the water temperature
Fig. 6. AC inductance of simulation and measurement.
Fig. 7. Steady temperature test under RMS 8.02 kA. 1066
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Fig. 8. Temperature distribution of coil at steady-stage 8.02 kA.
at input unchanged. In Fig. 8, the calculated average temperature at outlet section is 24.6 °C, which means that the calculated average temperature rise between water input and output is 7.4 °C under the same conditions, such as water flow, dc current and water temperature at input. There is a little difference temperature rise (0.1 °C) between simulation result (7.4 °C) and measured value (7.3 °C shown in Fig. 7) under long term test, which not only demonstrates the accuracy of design and fabrication, but also justify the high thermal reliability of dummy load.
Fig. 10. Pulse temperature curves under periodical test.
measurement points. Since the data in the simulation come from the outlet of tube, whereas test points are located at bottom of collecting pipe. It is hard for the temperature sensor to respond to the rapid changes of water temperature, especially at the downstream after all distributive pipes meet. The difference of water temperature rise between test and simulation at the end of each phase is the result of small current during the controlling stage in the test. The temperature variation curves of 20 points aiming to probe the temperature state of dummy load, such as body surface, busbar, aluminium conductor at insider and outsider, ambient and so on, are shown in Fig. 11. The curvilinear trend is in accordance with that shown in Fig. 10. The maximum temperature is less than 56.0 °C, located at the end of conductor where the cooling water flows out. Obviously, the temperature of conductor at outsides are higher than that at insider, since the water flow from insider to outsider.
4.2. Pulse model (70.7 kA/15 s) The temperature curves under three periodical test (phase A to phase C) with DC 70.7 kA (RMS) have been recorded, comparing with pulse simulation under same current excitation and cooling conditions. The detailed information of pulse test is shown in Fig. 9. The demineralized water flow is 25.0 m3/h, which means the water flow velocity in a single aluminium tube is 0.59 m/s. The water temperature at inlet is a constant 20.6 °C in the simulation analysis, whereas a little step rise of water temperature at input exists at the end of each phase, due to joule heat produced by each pulse current. According to the requirements from ITER, dummy load needs to be tested under three various operational conditions with 45 kA/30 s, 55 kA/20 s and 68 kA/15 s. The experimental results with thrice 68 kA/ 15 s on and 900 s off are shown in Fig. 10, which is more serious than the other two operational current in the joule heat aspect. The curves with hollow symbols reflect the changing process of temperature of water and conductor from simulation results [8–10]. Similarly, the experimental results are described by the curves with solid symbols. In addition, the experimental results are recorded at the sampling frequency of one per second to ensure the data integrity based on the test platform at ASIPP [11]. The consistency in the trend and rate of change of water temperature rise of water from actual test and simulation analysis are demonstrated. The difference in the peak may be due to the various
5. Conclusion This paper basically describes the experimental analysis of the dummy load prototype, focusing on the consistency test, inductance test, and temperature rise test, especially the 68 kA pulse test. The experimental results are coincided with the requirements and it shows the feasibility and availability of the prototype.
Fig. 9. Experimental current under three periodical pulse test.
Fig. 11. Temperature curve at typical position on the surface of dummy load.
1067
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Acknowledgements [5]
The authors would like to express gratitude to Ministry of Science and Technology of China for the foundation and staff of ASIPP for helpful discussions and suggestions. The view and opinion expressed herein does not necessarily reflect those of the ITER organization.
[6] [7] [8]
References
[9]
[1] P. Fu, G. Gao, et al., Preliminary design of the poloidal field AC/DC converter system for the ITER coil power supply, Fusion Sci. Technol. 64 (4) (2013) 741–747. [2] P.L. Mondino, et al., ITER R&D: auxiliary systems: coil power supply components, Fusion Eng. Des. 55 (2001) 325–330. [3] P. Fu, G. Gao, et al., Review and analysis of the ac/dc converter of ITER coil power supply, Proc. Applied Power Electronics Conference (2010) 1810–1816. [4] C. Li, M. Zhang, K.X. Yu, et al., R&D on dummy load prototype for ITER coil power
[10] [11]
1068
supply system[C]//Fusion engineering (SOFE), 2015 IEEE 26th Symposium on. (2015) 1–5. C. Li, M. Zhang, K.X. Yu, et al., Dummy load prototype design for ITER coil power supply system, Ieee Trans. Plasma Sci. 44 (9) (2016) 1711–1715. C. Li, Z. Song, et al., Dynamic stability analysis of high-power DC reactor for ITER poloidal field converter, J. Fusion Energy 34 (2) (2015) 202–206. Operation Manual of TH2882AS Series of Impulse Wingding Tester V1.0.0 http:// www.tonghui.com.cn/uploads/soft/150506/4-150506160F1.pdf. F. Marignetti, V.D. Colli, Y. Coia, Design of axial flux PM synchronous machines through 3-D coupled electromagnetic thermal and fluid-dynamical finite-element analysis, IEEE Trans. Ind. Electron. 55 (10) (2008) 3591–3601. A. Boglietti, A. Cavagnino, D. Staton, M. Shanel, M. Mueller, C. Mejuto, Evolution and modern approaches for thermal analysis of electrical machines, IEEE Trans. Ind. Electron. 56 (3) (2009) 871–882. Chuan Li, Zhiquan Song, et al., Thermal design of high-power DC reactor applied to ITER poloidal field converter, J. Fusion Energy 33 (5) (2014) 588–593. Xiuqing Zhang, Peng Fu, et al., Transient fault current test for ITER DC reactor on DC test platform in ASIPP, J. Fusion Energy 33 (5) (2014) 607–611.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Experimental analysis of dummy load prototype for ITER coil power supply system
T
Li Chuana, , Song Zhiquanb, Luo Bina, Li Jiaweia, Qin Xiuqic, Zhu Guangliangd, Fu Pengb, Zhang Xiuqingb, Li Huab ⁎
a
International Joint Research Laboratory of Magnetic Confinement Fusion and Plasma Physics, State Key Laboratory of Advanced Electromagnetic Engineering and Technology, School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China b Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, 230031, China c Anhui Electric Power Design Institute, Hefei, 230601, China d Hefei Keye Electrophysical Equipment Manufacturing Co., Ltd, Hefei, 230031 China
ARTICLE INFO
ABSTRACT
Keywords: ITER Dummy load Large inductance Experimental analysis Finite element analysis
This paper basically introduces the experimental analysis of the dummy load prototype, which aims to verify the capability of the ITER magnetic power supply systems to operate at their rated power levels without energizing the superconducting coils. The rated inductance of dummy load is 6.73 mH and the pulse test currents are 45 kA, 55 kA and 68 kA. According to the requirements of the large inductance and different pulse test current classes, a dry-type air-core water-cooling prototype with epoxy resin casting technique has been fabricated, which has 24 layers and 72 turns. The experimental analyses, including the consistency test, inductance test, and temperature rise test are described in detail. The consistency of different coil layer is ensured by comparing the voltage wave under the same voltage excitation. The inductance test includes dc inductance test by using the first order circuit method and ac inductance test by LCR instrument and finite element simulation analysis. The temperature rise test results under different current classes are introduced by comparing with the simulation analyses. The above experimental results are coincided with the requirements, which shows the feasibility and availability of the prototype.
1. Introduction ITER (International Thermonuclear Experimental Reactor) is an international nuclear fusion project to realize an experimental fusion reactor based on the concept of a tokamak. The parameters and operational modes of power systems are unique and depend on the various magnetic coils, which play a vital role in the plasma shape and position control. As an inductive component, dummy load is necessary and basically aiming to verify the capability of the various power supply systems to operate at rated power without charging the superconducting coils [1–3]. Thus, the function and performance of components, including converters, switching networks, discharge circuits in the power systems can be tested. Based on the prescribed functions and operational modes, the main requirements of dummy load are shown in Table 1. The rated DC inductance is 6.73 mH with the precision ± 10%. Considering the actual fabrication difficulties, the final precision ( ± 10%) is lower than that ( ± 1%) mentioned in previous work [4,5]. And this alteration has been
⁎
approved by ITER organization. The rated direct current is 7.75 kA. The currents like 45 kA, 55 kA and 68 kA are the typical values of central solenoid (CS) coils, poloidal field (PF) coils and toroidal field (TF) coils, respectively. Thus, the dummy load needs to operate with various modes owing to various testing requirements. The structure performance and thermal stability are crucial parts of dummy load, especially under the most severe pulse mode, i.e. 68 kA with 15 s, then operating with zero-current for 900 s, which is discussed in this paper. To meet the severe requirements in the thermal and dynamic performance under various operational modes, a structure of dry-type aircore water-cooling reactor with epoxy resin casting technique is adopted. A full prototype has 72 turns in total and contains eight parts in series and each part has three nine-turn pipes in parallel to carry more current, with same average diameter of 1.924 m. The full model and actual prototype are shown in Fig. 1. The whole procedures of dummy load mainly include four phases, primary treatment, winding and inspection, vacuum pressure impregnation (VPI), and assembly. VPI can enhance the insulation and mechanical
Corresponding author. E-mail address: [email protected] (C. Li).
https://doi.org/10.1016/j.fusengdes.2019.02.004 Received 8 October 2018; Received in revised form 27 January 2019; Accepted 2 February 2019 Available online 19 February 2019 0920-3796/ © 2019 Elsevier B.V. All rights reserved.
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Table 1 Main Requirements For Dummy Load. Parameters
Values
Rated DC inductance, mH Rated continuous current, kA Operating ambient temperature, °C Pulse operational modes: current, duty cycle
6.73 ± 10% 7.75 −25∼45 45 kA, 30 s on/900 off; 55 kA, 20 s on/900 s off; 68 kA, 15 s on/900 s off AC, 12/28/75; DC, 12 Demineralized water Cylinder, Aluminium Outdoor
Insulation level, Nominal /testing /impact, kV Cooling medium Shape, conductor material Location
Fig. 3. Schematic diagram of area difference ratio.
circuit. By analyzing and comparing the oscillating voltage wave between the terminals of measured piece (like coil layer) under the same pulse voltage excitation, insulation defect and coil difference can be easily detected. Various methods can be used to detect coil difference, such as by calculating the area difference ratio of standard wave and measured wave, which is defined as
=
Smeasured Sstandard × 100% Sstandard
(1)
Sstandard is the area enclosed by time axis and standard voltage wave of qualified piece. Smeasured is defined in the similar way. Fig. 3 shows the method adopted to measure the sub-layer coils of dummy load. The ± 2% compared to standard wave is the eligibility criteria by calculating the area difference ratio in the 1.5 periods of voltage wave starting with the second peak. According to the experimental records, the maximum difference is -1.4%. And all the sub-layer coils have passed the test.
Fig. 1. Model and actual prototype of dummy load ((a) main body, (b)distributive pipe, (c) collecting pipe, (d) vibration damper plate, (e)support components, (f)insulator).
performance [6]. Different with the previous work in reference [4,5], which focuses on the structural, thermal, electromagnetic environment and seismic analysis of dummy load prototype, this manuscript mainly introduces the experimental methods and results in detail, focusing on the consistency test, inductance test, and temperature rise test of dummy load prototype.
3. Inductance test 3.1. DC inductance As an inductive component, dummy load plays a crucial role in limiting the rate of current variation. Nevertheless, the dc inductance test with high precision, especially for air coil, is hard to realize due to the high sensitivity of inductance to the surrounding ferromagnetic material, and the calibration of additional inductance caused by test circuit. The lowest test frequency of existing series of LCR tester is generally 20 Hz, which still can’t regard as dc inductance test. The brief schematic diagram of dc inductance test of dummy load is shown in Fig. 4. The SCCPS has the ability of supplying a constant current up to 300 A to the measured piece, which is dummy load here. The power supply charges the dummy load to a certain current until flattop, then close the contactor to short and turn of the SCCPS. The current of dummy load freewheels through the closed contactor. Since the test works at a low power state, which is less than the short-circuit capacity of power supply, so the impact of short test can be ignored.
2. Consistency test Factors such as winding materials, magnetic materials, skeletons, processing techniques, etc., can cause insulation performance degradation between coil layers, and turns of coil products (such as transformers, motors, etc.). Therefore, it is extremely important to test the electrical properties without damaging the device under test. Pulse insulation tester Tonghui TH2882AS-5 adopted in this paper, not only can judge the insulation quality of the coil in a short time, but also can detect the individual differences of multiple coils, which is essential for quality of sub-layer coil with the same size [7]. Fig. 2 shows the schematic diagram of insulation tester, which mainly consists of a dc high voltage generator with adjustable voltage (maximum voltage 5 kV), a switch to trigger the test, and interior charge capacitance (C1). The stray capacitance of the circuit equalizes to C2. When connected to an inductive load, it becomes a second order
Fig. 4. Schematic diagram of dc inductance test. (SCCPS-superconducting constant current power supply, DL-dummy load).
Fig. 2. Schematic diagram of impulse insulation tester. 1065
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Table 2 Typical Inductance Value From Various Methods. Freq. (Hz)
Design value/mH
Measurement 1 (mH)
Measurement 2 (mH)
Measurement 3 (mH)
0 20 1000
6.92 6.47 5.47
– 6.30 5.34
– 6.12 5.45
7.55 6.20 5.21
* Dummy load in measurement 1 and 2 are assembled without resin-cast, so DC inductance cannot be measured by the method mentioned in Part 3.1. Thus, it is filled with dash. The design value at zero Hz is 6.92 mH owing to the consideration of some margin.
Due to the skin effect and proximity effect, the ac inductance declines with the frequency, and the inductance at 1 kHz only accounts for 79% of that at 0 Hz (i.e. dc inductance) according to the simulation results. Comparisons among three measure results show that the inductance off the ground is more close to the design value than that near the ground, which demonstrates the influence mentioned above. The reasons for the fact that the relative deviation 3 is a little higher may be the enhancement of capacitance effect, since the resin for casting increases the permittivity of materials between the layers. The measured dc inductance is 7.55 mH according to the first order circuit method, which is 9.1% higher than the corresponding design value, owing to the possibility that the measured circuit resistance is higher than the actual value and the measured inductance is larger than that of dummy load. The ac inductance relative deviation is within 5% below 1 kHz. Since the measured ac inductance at 20 Hz is 6.2 mH with a relative deviation -4.17%, the actual dc inductance can be reasonably inferred to be between 6.2 mH and 7.55 mH (actual value should be less than this value), which is acceptable for the actual debugging test.
Fig. 5. Declining current wave at 63 A with time constant 0.86 s.
The circuit is equivalent to a first order circuit (LR). The time constant (τ = L/R) depends on the inductance and resistance of circuit. The former approximates the inductance of dummy load. The inductance of test wires can be calibrated and is relatively small to that of dummy load. The latter consists of the resistance of dummy load, wires and closed contactor. Time constant can be deduced from the recorded wave of declining current as shown in Fig. 5. The dc inductance is 7.55 mH, according to the time constant 0.86 s and circuit resistance 8.78 mΩ as an average value of multiple tests. 3.2. AC inductance Dummy load consists of nine layer in radials direction and eight layer in axial direction. To enhance the thermal stability under pulse current like 68 kA and improve the current-carrying capability, three aluminium conductors with the size 40 mm × 50 mm are connected in parallel, which intensifies skin effect and proximity effect on the ac inductance, especially for this large equivalent section area. One design value based on finite element analysis and three measurement results with relative deviation compared to design value are described in Fig. 6 and Table 2, which also shows the simulation and measured results of ac inductance under several typical frequencies from 20 Hz to 1 kHz. The measurement 1 shows the test results of dummy load before resin casting and located on the ground, which means that the test may be affected by the ferromagnetic material, such as rebar to enhance the structural strength in the concrete. In the measurement 2, the dummy load is hung up to eliminate the influence of surrounding materials. The measurement 3 indicates the actual results of final assembled dummy load.
4. Temperature rise test 4.1. Steady model (8.02 kA) The relevant experimental conditions are as follows: water flow is 24.6 m3/h (i.e. water velocity is 0.58 m/s in each pipe), dc current RMS is 8.02 kA (a little higher than the rated value 7.75 kA), the initial temperature of input water and ambient are 17.2 °C and 19.0 °C (approx. value), respectively. The steady temperature test results are shown in Fig. 7. The steady state with temperature rise changing within 1.0 °C has lasted more than 4 h. It is worth mentioning that the water temperature at input in the simulation is set to be a constant value (17.2 °C), whereas that of test is changeable due to the difficulty in maintaining the water temperature
Fig. 6. AC inductance of simulation and measurement.
Fig. 7. Steady temperature test under RMS 8.02 kA. 1066
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Fig. 8. Temperature distribution of coil at steady-stage 8.02 kA.
at input unchanged. In Fig. 8, the calculated average temperature at outlet section is 24.6 °C, which means that the calculated average temperature rise between water input and output is 7.4 °C under the same conditions, such as water flow, dc current and water temperature at input. There is a little difference temperature rise (0.1 °C) between simulation result (7.4 °C) and measured value (7.3 °C shown in Fig. 7) under long term test, which not only demonstrates the accuracy of design and fabrication, but also justify the high thermal reliability of dummy load.
Fig. 10. Pulse temperature curves under periodical test.
measurement points. Since the data in the simulation come from the outlet of tube, whereas test points are located at bottom of collecting pipe. It is hard for the temperature sensor to respond to the rapid changes of water temperature, especially at the downstream after all distributive pipes meet. The difference of water temperature rise between test and simulation at the end of each phase is the result of small current during the controlling stage in the test. The temperature variation curves of 20 points aiming to probe the temperature state of dummy load, such as body surface, busbar, aluminium conductor at insider and outsider, ambient and so on, are shown in Fig. 11. The curvilinear trend is in accordance with that shown in Fig. 10. The maximum temperature is less than 56.0 °C, located at the end of conductor where the cooling water flows out. Obviously, the temperature of conductor at outsides are higher than that at insider, since the water flow from insider to outsider.
4.2. Pulse model (70.7 kA/15 s) The temperature curves under three periodical test (phase A to phase C) with DC 70.7 kA (RMS) have been recorded, comparing with pulse simulation under same current excitation and cooling conditions. The detailed information of pulse test is shown in Fig. 9. The demineralized water flow is 25.0 m3/h, which means the water flow velocity in a single aluminium tube is 0.59 m/s. The water temperature at inlet is a constant 20.6 °C in the simulation analysis, whereas a little step rise of water temperature at input exists at the end of each phase, due to joule heat produced by each pulse current. According to the requirements from ITER, dummy load needs to be tested under three various operational conditions with 45 kA/30 s, 55 kA/20 s and 68 kA/15 s. The experimental results with thrice 68 kA/ 15 s on and 900 s off are shown in Fig. 10, which is more serious than the other two operational current in the joule heat aspect. The curves with hollow symbols reflect the changing process of temperature of water and conductor from simulation results [8–10]. Similarly, the experimental results are described by the curves with solid symbols. In addition, the experimental results are recorded at the sampling frequency of one per second to ensure the data integrity based on the test platform at ASIPP [11]. The consistency in the trend and rate of change of water temperature rise of water from actual test and simulation analysis are demonstrated. The difference in the peak may be due to the various
5. Conclusion This paper basically describes the experimental analysis of the dummy load prototype, focusing on the consistency test, inductance test, and temperature rise test, especially the 68 kA pulse test. The experimental results are coincided with the requirements and it shows the feasibility and availability of the prototype.
Fig. 9. Experimental current under three periodical pulse test.
Fig. 11. Temperature curve at typical position on the surface of dummy load.
1067
Fusion Engineering and Design 146 (2019) 1064–1068
C. Li, et al.
Acknowledgements [5]
The authors would like to express gratitude to Ministry of Science and Technology of China for the foundation and staff of ASIPP for helpful discussions and suggestions. The view and opinion expressed herein does not necessarily reflect those of the ITER organization.
[6] [7] [8]
References
[9]
[1] P. Fu, G. Gao, et al., Preliminary design of the poloidal field AC/DC converter system for the ITER coil power supply, Fusion Sci. Technol. 64 (4) (2013) 741–747. [2] P.L. Mondino, et al., ITER R&D: auxiliary systems: coil power supply components, Fusion Eng. Des. 55 (2001) 325–330. [3] P. Fu, G. Gao, et al., Review and analysis of the ac/dc converter of ITER coil power supply, Proc. Applied Power Electronics Conference (2010) 1810–1816. [4] C. Li, M. Zhang, K.X. Yu, et al., R&D on dummy load prototype for ITER coil power
[10] [11]
1068
supply system[C]//Fusion engineering (SOFE), 2015 IEEE 26th Symposium on. (2015) 1–5. C. Li, M. Zhang, K.X. Yu, et al., Dummy load prototype design for ITER coil power supply system, Ieee Trans. Plasma Sci. 44 (9) (2016) 1711–1715. C. Li, Z. Song, et al., Dynamic stability analysis of high-power DC reactor for ITER poloidal field converter, J. Fusion Energy 34 (2) (2015) 202–206. Operation Manual of TH2882AS Series of Impulse Wingding Tester V1.0.0 http:// www.tonghui.com.cn/uploads/soft/150506/4-150506160F1.pdf. F. Marignetti, V.D. Colli, Y. Coia, Design of axial flux PM synchronous machines through 3-D coupled electromagnetic thermal and fluid-dynamical finite-element analysis, IEEE Trans. Ind. Electron. 55 (10) (2008) 3591–3601. A. Boglietti, A. Cavagnino, D. Staton, M. Shanel, M. Mueller, C. Mejuto, Evolution and modern approaches for thermal analysis of electrical machines, IEEE Trans. Ind. Electron. 56 (3) (2009) 871–882. Chuan Li, Zhiquan Song, et al., Thermal design of high-power DC reactor applied to ITER poloidal field converter, J. Fusion Energy 33 (5) (2014) 588–593. Xiuqing Zhang, Peng Fu, et al., Transient fault current test for ITER DC reactor on DC test platform in ASIPP, J. Fusion Energy 33 (5) (2014) 607–611.