- Email: [email protected]
Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask
International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Contents lists available at ScienceDirect
International Journal of Pressure Vessels and Piping journal homepage: www.elsevier.com/locate/ijpvp
Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask Chin-Yu Lin, Tung-Yueh Wu*, Chin-Cheng Huang Mechanical and System Engineering Program, Institute of Nuclear Energy Research (INER), Atomic Energy Council, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 April 2015 Accepted 20 April 2015 Available online xxx
In this paper, a series of dynamic impact analysis for installing a dry storage canister into a vertical concrete cask (VCC) is performed. The dry storage system considered herein is called HCDSS-69, recently developed by INER and being capable of accommodating 69 bundles of BWR spent nuclear fuels. The impact accident is stemming from a conservative consideration of accidental movement when the canister is being hoisted into a VCC. According to NUREG-0554, the accidental movement is conserva tively simulated by 80 mm- and 160 mm-height free-drop motions and then with straight and 2 -oblique impact to a pedestal in VCC. A symmetric fully 3-D finite element model is built and analyzed using the explicit finite element code, LS-DYNA. Geometrical, contact, and material nonlinearities are all taken into account. The analysis result concludes that the permanent deformations of the canister are not severe to affect fuel retrieve after the impact accident and the maximum stress intensity in the canister shell can meet the ASME code appendix F F-1340, preventing the leakage of radioactive materials. The study also found that with properly reducing the wall thickness of the pedestal cylinder, the maximum acceleration and permanent deformation of the canister can be much alleviated, even though the drop height is increased to the double of the required brake distance specified in NUREG-0554. The damages of the pedestal in each analysis are moderate so that the heat transfer condition after the impact accident can be bounded by the off-normal event for half-blockage of air inlets. © 2015 Published by Elsevier Ltd.
Keywords: Dry storage canister Impact analysis Accident movement Explicit finite element
1. Introduction With the aggravation of global warming and climate change, the living with low-carbon emission and energy saving is being advocated worldwide. Having a characteristic in low-carbon emission, the nuclear energy continues to be a good option for power resource in the future. However, the handling of highly radioactive spent nuclear fuels and the safety operation during the complex and huge natural disasters like those happened in Japan 2011 strongly impacts the public confidence on the use of nuclear energy. Fortunately, the Fukushima event also proved that the dry storage system is able to successfully withstand the huge natural disasters. Spent nuclear fuels discharged from reactor core are commonly stored in spent fuel pool (SFP), not only for shielding radiation by the water in SFP but also for removing residual heat through coolant circulation. Such kind of so-called wet-storage is the way
* Corresponding author. E-mail address: [email protected] (T.-Y. Wu).
adopted by most current in-service nuclear power plants (NPP). However, most of NPPs built in 70 s and 80 s with licenses to be renewed do not have enough capacity for spent fuel storage, even though their SFP has been replaced by high-density fuel racks. The independent spent fuel storage installation (ISFSI) is thus adopted as an interim storage method before final disposal policy is determined. In US, many dry storage systems have been successfully developed and applied to practical nuclear engineering practice, such as the HI-STORM and UMS systems respectively developed by Holtec [1] and NAC [2]. Recently, the institute of nuclear energy research (INER) in Taiwan also deployed a project to develop an experimental dry storage system called HCDSS-69. It is designed in accordance with the needs of the NPPs in Taiwan and being capable of accommodating 69 bundles of BWR spent nuclear fuels. The system design and analyses are in accordance with 10 CFR72 [3], ANSI/ANS 57.9 [4], the applicable sections of the ASME Code [5], and the ACI code [6]. The design of HCDSS-69 system also follows guidance of the standard review plans (SRPs), NUREG-1536 [7] and NUREG-1567 [8] published by US NRC. The aim of SRPs is to
http://dx.doi.org/10.1016/j.ijpvp.2015.04.006 0308-0161/© 2015 Published by Elsevier Ltd.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
2
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
standardize the review processes to the license application of dry storage facilities. The SRPs clearly instruct that off-normal and accident conditions must be addressed in the safety report of ISFSI applicants. The off-normal event commonly considers the offnormal ambient temperature, off-normal handling, half-blockage of air inlet, and so on. The occurrence frequency of the events is around once per year. The accident event considers more sever accidents which may occurs only one time for entire service life of ISFSI, such as the extremely natural disasters and handling drop impact accidents. Obviously the scenarios of events should be made up in accordance with the ISFSI operational procedures. The assessment on handling impact accident of dry storage cask should be provided in the safety analysis report.
In order to provide applicants for an NRC license under 10 CFR Part 72 with a method for evaluating storage casks for low velocity impact conditions, the NRC has conducted a series of drop test studies of a solid steel billet and a near-full-scale empty cask in the Sandia National Lab. (SNL) and Lawrence Livermore National Lab. (LLNL), respectively. The experimental results reported in NUREG6608 [9] are compared and verified by those of numerical simulation using the explicit finite element code DYNA3D [10] developed in LLNL. Since then employment of explicit finite element code for waste cask drop impact simulation has become intensive. Many researchers and engineers thus employed the explicit finite element code to conduct their drop impact analyses. Huang and Wu [11] studied a VCC tip-over impact using a simplified canister model
Fig. 1. Schematics of HCDSS-69.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 2. Finite element model of HCDSS-69.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
3
4
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14 Table 1 Material properties of 304L stainless steel [5]. 304L Steel property
Value at 400 F
Mass density, r Modulus of elasticity, E Poisson's ratio, n Yield stress, sy Ultimate stress, Su 0.9Su
8.04E-9 ton/mm3 182.7 GPa 0.31 120.7 MPa 404.7 MPa 364.6 MPa
with pseudo tensor concrete plasticity. Kim et al. [12] conducted finite element analyses and tests for a shock-absorbing effect of a pad in a spent fuel storage cask. Kim et al. [13,14] further used explicit finite element procedures to simulate the drop impact of a transport cask. The simulation of drop impact is usually difficult and may have different results depending on FEA codes. Lee et al. [15] carried out a finite element analysis for a 9 m free drop and the puncture condition under the hypothetical accident condition by using LS-DYNA and ABAQUS/Explicit. Wu et al. [16] investigated the size effect of drop test using LS-DYNA. A numerical simulation is indeed able to compensate for some disadvantages of field test. For example, responses of some key locations where sensor installations are impossible but can be reported by simulations. In addition, the repeating drop angles can be achieved without any difficulties. The current main operation procedures of independent spentfuel dry storage for the NPPs in Taiwan include: (a) Spent-fuel assemblies are loaded into the canister in SFP with water depth over 10 m. (b) A transfer cask with loaded canister is then lifted from the SFP and sits on the border. After lid sealed, water drained, vacuumed and helium filled, the cask will be decontaminated and transferred to the storage site using a dolly. (c) In the phase of transportation, an impact limiter is installed on the top of the transfer cask and a cable tied down system is used to ensure the transfer cask not falling off the dolly deck under SSE level of earthquake. (d) In the storage site, the transfer cask is moved to the top of a VCC casted in-situ. Then the canister is hoisted using a single failure-proof crane to install it into the VCC. Although the single-failure proof crane can prevent the cask from drop accident during loading process, it still requires a braking distance not exceeding 3 in (80 mm), as that specified in NUREG-0554 [17]. It
implies that a drop impact accident may happen within 80 mm. This accident is frequently ignored in many ISFSI safety reports. In order to study the consequence of the accident movement during installation, dynamic impact analyses for installing a dry storage canister into a VCC are performed in this paper. The impact accident is stemming from a conservative consideration of accidental movement when the canister is being hoisted into the VCC. According to NUREG-0554 and the gap between VCC liner and canister, the accidental movement is conservatively simulated by 80 mm-height free-drop motions and with straight and 2 -oblique impacts onto a pedestal in the VCC. The minimum height of free drop is 80 mm, being adopted in accordance with the maximum braking distance of crane specified in NUREG-0554. For defense in depth, the scenario with double drop height, 160 mm is further investigated. A symmetric fully 3-D finite element model is built and analyzed using the explicit finite element code, LS-DYNA [18e20]. The main concerns of the impact accident in this work include: (a) The permanent deformations of the canister cannot be severe so that the retrieve of spent fuel is difficult. This is to meet the basic requirement of CFR72. (b) The magnitude of the stress intensity in the canister shell has to meet the ASME code appendix F, F-1340, preventing the leakage of radioactive materials. (c) The damages of the pedestal must be small enough so that the heat transfer condition after the impact accident can be bounded by the off-normal event for half-blockage of air inlets. (d) The maximum acceleration of the canister cannot be over 35 g, which is the design limit of the internal components in canister. In addition to the main concerns, this work also investigated the effect of reduced wall thickness of the pedestal cylinder to understand the possibility of optimal design for the pedestal. In addition to the Section 1 of this paper, Section 2 introduces the main features and dimensions of the HCDSS-69 dry storage system. Section 3 introduces the finite element model prepared for LS-DYNA input data. Numerical results are reported and discussed in the Section 4. Finally, Section 5 shows the conclusions and remarks. 2. Dry storage cask of HCDSS-69 The HCDSS-69 system is an experimental cask designed to accommodate 69 bundles of BWR spent nuclear fuels. The profile of
Fig. 3. The observation points in the Finite element model.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
the storage cask of HCDSS-69 system is illustrated in Fig. 1(a) and the detailed components in the canister in Fig. 1(b). The height and outer radius of the VCC are 5750 mm and 1850 mm, and those of the canister are 4840 mm and 1678 mm. Most components of the canister of HCDSS-69 are made of AISI 304 series stainless steels. The main components in the canister includes the outer shell, 4 side-support panels, fuel tube assembly, Boral neutron absorbing plates, shield lid and structural lid, as shown in Fig. 1(b). The arrangement of the fuel tubes is assembled in a compact stack form,
5
rather than the support disks used in INER-HPS [11] or NAC UMS [2]. The Boral neutron absorbing plates are embedded between the walls of two fuel tubes. The stacked tubes are finally bundled up by 4 side-support panels connected by fasteners. One design philosophy for the HCDSS-69 is that the fuel assembly must be able to be withdrawn out if underwent hypothetical drop impact accident, in order to satisfy the regulation of 10 CFR72 [3]. The bottoms of fuel tubes and neutron absorbing plates are supported and constrained by a set of spacers installed on the base plate of the canister outer
Fig. 4. Distribution of effective plastic strain.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
6
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
shell for accelerating drainage. The canister is closed by a structural lid and sealed using welding, after loading fuels and covering the shield lid. With full loading the BWR fuels, the total weight of the canister is around 35 metric tons. 3. Explicit finite element model According to NUREG 0554, the accident movement can be conservatively analyzed via 80 mm-height and free-drop induced
impact. An extended analysis on the responses of the pedestal and canister for the doubled drop height (160 mm) is also conducted based on the defense-in-depth policy. The drop impact simulation includes a straight and a 2 -oblique impact onto the steel pedestal. The angle of 2 is evaluated in accordance with the gap between the canister and VCC liner. The rigid body motion during free-drop is replaced by the equivalent initial velocities, V0 ¼ 1520 mm/s and 1772 mm/s for saving computation cost. In order to investigate the characteristic of energy absorption of the pedestal, the wall
Fig. 5. Distribution of maximum shear stress.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14 Table 3 The height change of air inlet.
0.08
Effective plastic strain (%)
0 Degree C1 C2 P1 P2
0.06
Analyzed case
CASE I
Impact angle (Deg.)
0
2
0
2
0
2
Deformed length (mm) Height change ratioa (%)
0.08 0.05
0.41 0.25
0.19 0.12
0.42 0.26
5.93 3.71
27.6 17.3
a
CASE II
CASE III
Height Change Ratio ¼ crush length/original height of the air inlet.
0.04
0.02
0 0
10
20
30
40
30
40
Time (ms)
(a) Straight impact of CASE I 0.4 2 Degree
Effective plastic strain (%)
7
C1 C2 P1 P2
0.3
0.2
0.1
0 0
10
20
Time (ms)
(b) Oblique impact of CASE I Fig. 6. Effective plastic strain responses.
thickness of pedestal is reduced from 25 mm to 10 mm in the case with released height 160 mm. Thus, totally 3 cases with 6 analyses are performed in this work. The symmetric 3-D finite element model of the canister and pedestal for straight and oblique drop impact analysis is illustrated in Fig. 2(a). The fuel tubes, neutron absorbing plates, and side support panels are modeled by shell elements, as shown in Fig. 2(b). In order to obtain more precise stress distribution, the mesh density of the outer shell and side support panels are higher
Table 2 Stress intensity and safety factor. Drop height (mm)
CASE I
Impact angle (Degree)
0
2
CASE II 0
2
CASE III 0
2
Stress intensity (MPa) Safety factor (0.9Su/Sint)
168.1 2.17
195.8 1.86
173.1 2.10
196.0 1.85
140.4 2.59
97.5 3.73
than those of other components. The average element sizes of the outer shell and the support panel are respectively 0.02 0.02 mm2 and 0.04 0.04 mm2. All the element types used in this model are fully integrated elements. The entire finite element model is built up by 118409 elements and 133035 nodes in total. The stiffness of the cask lids is much higher than those of other components, thus they are simplified as a rigid body, and using the surface tied contact function of LS-DYNA to simulate the welding between the lid and the top of outer shell wall. The modeling of contact interactions between each part is carried out using the general contact algorithm in LS-DYNA. In this contact algorithm, the contact search strategy includes global bucket sort and local contact searches. The bucket sort frequency used in this work is set to be 200/cycle. The tangential friction forces are computed by regularized MohreCoulomb's friction law with static and dynamic frictional coefficients 0.6 and 0.5. In contrast to the function of single surface contact in LS-DYNA, the automatic general contact can provide edge contact detection. It is necessary in this study. For the contact formulation details, the reader can consult the references [18] and [21]. In the drop impact process, the 69 bundles of BWR spent fuel assembly also play a crucial role due to its heavy weight. It may induce significant contact pressure on the base plate of canister for straight impact, and on the support panels for the oblique impact. However, the geometry details of the fuel assembly are very complex. Thus only a simplified geometry modeled using 3-D solid elements with equivalent mass density and estimated stiffness is used for analysis, according to the practical engineering experience provided by NAC [22]. The drop impact analyzed in this paper belongs to the category of low-velocity impact, thus the elasticeplastic material model combined with CowpereSymonds strain rate function is adopted. The detail of the material model can be found in the theory manuals of LS-DYNA [18,20] or the FEM textbook [21]. The related material properties for 304L stainless steel at 400 F temperature is listed in Table 1. The system damping ratio for the model is 7%, according to the common codes or standards for nuclear components. 4. Results and discussions The results obtained from the analyses according to the scenarios mentioned previous sections are reported and discussed in this section. The post-processor software, LS-Prepost is employed to review and post-process the data dumped by LS-DYNA. As shown in Fig. 3, five critical locations are selected for observation and judgment, including point F1 which is at the bottom of a fuel assembly closed to the side support panel, points C1 and C2 on the canister shell wall, P1 on the junction of the pedestal cylinder and the pedestal flange, and P2 on the pedestal cylinder. For expediency and simplicity, the 6 analyses are classified into 3 cases. In each case the results of the straight and 2 -oblique impacts onto the steel pedestal are reported and discussed. The stress in the canister outer shell is conservatively checked using ASME code Section III, Appendix F, F-1340 [5].
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
8
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
4.1. CASE I: 80 mm drop height, normal wall thickness of pedestal cylinder Figs. 4 and 5 respectively show the plastic deformation and maximum shear stress distributions under the conditions of this case. Under the straight impact, the plastic deformation concentrates at the junction weld between the canister outer shell and base plate, and so does the maximum shear stress. In addition, the
flange of pedestal has large deflection. On the other hand, under the oblique impact the location of plastic deformation transfers to the corner near contact area. Similar to the experimental result reported by Saegusa et al. [23], a small buckling bulge occurs at the bottom of the canister outer shell, as shown in Figs. 4(b) and 5(b). Fortunately, it is not serious to affect the function of fuel retrieve.
Fig. 7. Distribution of effective plastic strain.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 6(a) and (b) respectively plot the effective plastic strain response at the points P1, P2, C1, and C2 under the straight and oblique impacts. In the case of straight impact, Fig. 6(a) reveals that point C1 and P1 have larger plastic strain than the other 2. The location of point C1 is right at the junction weld between the canister outer shell and base plate. The maximum plastic strain at point C1 is only around 0.062%. In the condition of the 2 -oblique
9
impact, Fig. 4(b) shows the plastic strain at point P1 significantly increases to 0.39%, about 4 times of that under straight impact. It reason is that the contact area in the oblique impact is much smaller than that in the straight impact. The effective plastic strains at points C1 and C2 are also increased to 0.11% and 0.12%. In this case, the location of P2 (the pedestal cylinder) doesn't have any plastic deformation both in the straight and oblique impacts.
Fig. 8. Distribution of maximum shear stress.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
10
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Table 2 lists the maximum shear stress of the canister outer shell under straight and oblique impacts. Compared them with the requirement of ASME code Section III, Appendix F, F-1340 [5], the primary bending plus membrane stress is smaller than 0.9Su. They respectively have safety factors of 2.17 and 1.86. The result of the analyses shows that this canister is safe under the 80 mm-height free drop. It should be noted that in the real situation of crane failure, the brake system in a single-failure proof crane will be activated. Thus the simulation by free drop is conservative enough. Base on the concept of defense in depth, the additional two sets of scenario are further investigated.
4.2. CASE II: 160 mm drop height, normal wall thickness of pedestal cylinder In this case, the drop height is raised to 160 mm, doubling the initial impact energy in the previous case. Figs. 7(a),8(a) and 9(a)
0.12
Effective plastic strain (%)
0 Degree C1 C2 P1 P2
0.08
0.04
0 0
10
20
30
40
respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the straight impacts. Large deflection is found at the top flange of the pedestal due to plastic bending. None of buckling bulge is found in the pedestal cylinder and canister shell. The area of plastic zone does not significantly extend, compared to those in previous case. However, the effective plastic strains increase to 0.118% at point P1 and 0.075% at point C1, as shown in Fig. 9(a). Figs. 7(b), 8(b) and 9(b) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the oblique impacts. Compared to the CASE I, a larger bulge due to local buckling occurred at the bottom of the canister outer shell, as shown in Fig. 8(b). Fortunately, the side support penal still bundles the fuel tubes well and not to jeopardize the function of fuel retrieve. The effective plastic strains increase to 0.4% at point P1, 0.125% at point C1, and 0.21% at point C2, as shown in Fig. 9(b). In other words, the plastic deformation of point C2 becomes larger than that of C1. It reveals that a yielding line forms on the canister outer shell and becomes an impact energy absorbing mechanism. It would be detrimental to the fuel retrieve if a more violent straight impact were encountered. The local buckling of the canister outer shell can be remedied by welding together with the side support panels. Due to the limitation of manuscript page, the technical details does not discussed here. It can also be avoided by adequately adjusting the stiffness and strength of the pedestal cylinder. We will demonstrate this idea in the next case. Compared the stress intensity (2 times of maximum stress) in the canister outer shell with the requirement of ASME code Section III, Appendix F, F-1340 [5], as listed in Table 2, the cases under the straight and oblique impact respectively have safety factors of 2.10 and 1.80. The safety factors in this case slightly decreases but this canister is still safe even under the 160 mmheight free drop. 4.3. CASE III: 160 mm drop height, reduced wall thickness of pedestal cylinder
Time (ms)
(a) Straight impact of CASE II 0.5 2 Degree C1 C2 P1 P2
Effective plastic strain (%)
0.4
0.3
0.2
0.1
0 0
10
20
Time (ms)
(b) Oblique impact of CASE II Fig. 9. Effective plastic strain responses.
30
40
The previous 2 cases show that point P2, which locates on the pedestal cylinder, only slightly deforms in elastic range. In this case, the drop height is kept the same as that of CASE II, but the wall thickness of pedestal cylinder is reduced to 10 mm in order to trigger buckling deformation on the pedestal cylinder. Figs. 10(a), 11(a) and 12(a) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the straight impacts in this case. In Fig. 10(a), one can see that the area of the plastic zone in the pedestal significantly reduces. Furthermore, there is no plastic deformation in the canister outer shell, as the P2 curve shown in Fig. 12(a). In contrast, an elastic buckling deformation forms on the pedestal cylinder, absorbing the impact energy via large area of elastic large deformation (see Fig. 11(a)). Compared to those of the previous 2 cases, the effective plastic strain at point C1 and C2, i.e. the location of canister shell, drastically reduced to the values below 0.08%. Figs. 10(b), 11(b) and 12(b) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the oblique impacts. As expected, the plastic zones gather in the pedestal and the buckling bulge on the canister outer shell vanishes, as those reflected on the response curve C2 in Fig. 12(a). Table 2 shows that the maximum stress intensities of the canister shell in this case also satisfy the
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
11
Fig. 10. Distribution of effective plastic strain.
requirement of ASME code Section III, Appendix F, F-1340 [5]. The cases under the straight and oblique impact respectively have safety factors of 2.59 and 3.73. They are obviously higher and safer than those of previous 2 cases. The result concludes that with a proper design of pedestal, the impact force induced by vertical drop accident can be effectively alleviated, especially for the oblique impact accident. Figs. 13 and 14
respectively plot the velocity and acceleration responses of point F1. The maximum accelerations in each case are below 35 g, satisfying the design limit of the canister internal components. The response curves under the oblique impact accident reveal that CASE I and CASE III almost have the same magnitude of maximum acceleration, revealing again the performance of the pedestal cylinder with properly reduced stiffness and strength. The reduction
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
12
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 11. Distribution of maximum shear stress.
of the pedestal cylinder stiffness and strength can effectively absorb the impact energy, however the heat transfer condition have to be further aware of after the accident. INER had been completed a series of heat transfer analysis for the HCDSS-69 system under the accidents of normal, abnormal high temperature, and high ambient temperature using computational fluid dynamics [24]. The analysis pointed out that the current
design of the HCDSS-69 system is able to meet the temperature design basis of fuel assembly, shielding material, support structure and other components in above conditions. In addition, the fuel components of the maximum temperature are lower than 400 C in air inlet half-blockage accident or high ambient temperature accident. Based on this conclusion, Table 3 summarizes the height change ratio of the VCC air inlet in each case. The values listed show
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
1
0.06 0 Degree
0 Degree
C1 C2 P1
Case I Case II Case III
0.5
P2
0.04
0
Velocity (m/s)
Effective plastic strain (%)
13
0.02
-0.5
-1
-1.5
0
-2 0
10
20
30
40
0
Time (ms)
20
30
40
30
40
Time (ms)
(a) Straight impact.
(a) Straight impact of CASE III 1
0.5
2 Degree
2 Degree C1 C2 P1 P2
0.4
Case I Case II Case III
0.5
0
Velocity (m/s)
Effective plastic strain (%)
10
0.3
0.2
-0.5
-1
0.1
-1.5
0
-2 0
10
20
30
40
Time (ms)
(b) Oblique impact of CASE III Fig. 12. Effective plastic strain responses.
that the damage of air inlet is not severe. Even in the most severe case (CASE III), the maximum height change ratio is not greater than 20%. It is obvious that component temperature can be bounded by the off-normal analysis for the air inlet half-blockage condition [24].
5. Concluding remarks In this paper, a series of dynamic impact analysis for installing the HCDSS-69 canister into a VCC is presented. The impact accident focuses on the scenario of accidental movement when the canister is being hoisted into a VCC. The accidental movement is conservatively simulated by free-drop motions and straight and oblique impacts onto a steel pedestal. Some concluding remarks based on the analysis results are drawn as follows:
0
10
20
Time (ms)
(b) Oblique impact Fig. 13. Velocity responses of point F1.
(i) The maximum effective plastic strains on the canister shell are very low under the 80 mm-height free drop, regardless of the straight or oblique impact. The outer shell of the cask only has slight permanent deformation. The analyses demonstrate that the canister is safe under such conditions. It should be noted that in the real situation of crane failure, the brake system in any single-failure proof crane will be activated. Thus the simulation by 80 mm-height free drop impact is conservative enough. (ii) The result of the 3 cases performed in Section 4 shows that the permanent deformations of the canister are not severe, thus the retrieve of spent fuel is viable after the accident and the basic requirement of CFR72 can be met. Meanwhile, the stress intensity in the canister outer shell meets the ASME code appendix F F-1340, preventing the leakage of radioactive materials. In addition, the crush levels of the pedestal are
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
14
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Acknowledgment
40
0 Degree
Acceleration (g)
The research in this paper is under the project of Industrialization Platform Development for Nuclear Technology and sponsored by the funding of National Energy Project phase I, Ministry of Science and Technology in Taiwan (Project No. ANS040401). The authors would like to gratefully express their acknowledgement.
Case I Case II Case III
30
20
References 10
0
-10 0
10
20
30
40
30
40
Time (ms)
(a) Straight impact. 16
2 Degree Case I Case II Case III
Acceleration (g)
12
8
4
0
-4 0
10
20
Time (ms)
(b) Oblique impact Fig. 14. Acceleration responses of point F1.
low enough so that the heat transfer condition after the impact accident can be bounded by the off-normal event of half-blockage air inlets. Finally, the maximum acceleration of the canister is not over 35 g, the design limit of the internal components of the HCDSS-69 canister. In conclusion, the canister of HCDSS-69 system is safe under the accident movement of any single-failure proof crane.
[1] Holtec Inc. Topical safety analysis report for the holtec international storage and transfer operation reinforced concrete module cask system (HI-STORM 100 cask system). NRC Docket No. 72e1014. 1999. [2] NAC International Inc. Final safety analysis report for the UMS universal storage system. Docket No. 72e1015. 2005. [3] 10 CFR 72. Licensing requirements for the independent storage of spent nuclear fuel and high level radioactive waste and reactor-related greater than class C waste. Washington, DC: Code of Federal Regulations, US Government; 2007. [4] ANSI/ANS 57.9-1992. Design criteria for an independent spent fuel storage installation (Dry Type). La Grange Park, IL: American Nuclear Society; 1992. [5] ASME boiler and pressure vessel code, section III. New York, NY: American Society of Mechanical Engineers; 2004. [6] ACI-349-85. Code requirement for nuclear safety related concrete structures and commentary. Farmington Hills, MI: American Concrete Institute; 1985. [7] NUREG-1536. Standard review plan for dry cask storage systems, rev. 1. US Nuclear Regulatory Commission; 2010. [8] NUREG-1567. Standard review plan for spent fuel dry storage facilities. US Nuclear Regulatory Commission; 2000. [9] NUREG-6608. Summary and evaluation of low-velocity impact tests of solid steel billets onto concrete pads. US Nuclear Regulatory Commission; 1998. [10] Whirley RG. DYNA3D, a nonlinear, explicit, three-dimensional finite element code for solid and structural mechanics e user's manual. Lawrence Livermore National Laboratory; 1993. UCRL-MA-107254. [11] Huang CC, Wu TY. A study on dynamic impact of vertical concrete cask tipover using explicit finite element analysis procedures. Ann Nucl Energy 2009;36:213e21. [12] Kim DH, Seo KS, Lee JC, Bang KS, Cho CH, Lee SJ, et al. Finite element analyses and verifying tests for a shock-absorbing effect of a pad in a spent fuel storage cask. Ann Nucl Energy 2007;34:871e82. [13] Kim KS, Kim JS, Choi KS, Shin TM, Yun HD. Dynamic impact characteristics of KN-18 SNF transport cask-part 1: an advanced numerical simulation and validation technique. Ann Nucl Energy 2010;37:546e59. [14] Kim KS, Kim JS, Choi KS, Shin TM, Yun HD. Dynamic impact characteristics of KN-18 SNF transport cask-part 2: sensitivity analysis of modeling and design parameters. Ann Nucl Energy 2010;37:560e71. [15] Lee YS, Ryu CH, Kim HS, Choi YJ. A study on the free drop impact of a cask using commercial FEA codes. Nucl Eng Des 2005;235:2219e26. [16] Wu TY, Lee SY, Kang LC. A study on dynamic responses of a spent-fuel dry storage cask under vertical drop accident. Ann Nucl Energy 2012;42:18e29. [17] NUREG-0554. Single-failure cranes for nuclear power plant. US Nuclear Regulatory Commission; 1998. [18] Hallquist JO. LS-DYNA theory manual. Livermore Software Technology Corporation; 2006. [19] Livermore Software Technology Corporation. LS-DYNA keyword user's manual version 971. 2012. [20] Livermore Software Technology Corporation. LS-DYNA material models. 2012. [21] Belytschko T, Liu WK, Moran B. Nonlinear finite elements nonlinear finite elements for continua and structures. 1st ed. Wiley; 2000. [22] Stephenson RW, Griswold M. Evaluation of the loaded INER transfer cask during a tip-over accident condition. Calculation report, Cal. No. 63005100e2002. NAC International Inc.; 2002. [23] Saegusa KT, Shirai K, Arai T, Tani J, Takeda H, Wataru M, et al. Review and future issues on spent nuclear fuel storage. Nucl Eng Technol 2010;42: 237e48. [24] Huang JJ, Lee SY. High capacity dry storage system-the heat transport analysis of 69 dry storage system and new type vertical concrete cask. INER report No. INER-10358. 2013.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
Contents lists available at ScienceDirect
International Journal of Pressure Vessels and Piping journal homepage: www.elsevier.com/locate/ijpvp
Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask Chin-Yu Lin, Tung-Yueh Wu*, Chin-Cheng Huang Mechanical and System Engineering Program, Institute of Nuclear Energy Research (INER), Atomic Energy Council, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 April 2015 Accepted 20 April 2015 Available online xxx
In this paper, a series of dynamic impact analysis for installing a dry storage canister into a vertical concrete cask (VCC) is performed. The dry storage system considered herein is called HCDSS-69, recently developed by INER and being capable of accommodating 69 bundles of BWR spent nuclear fuels. The impact accident is stemming from a conservative consideration of accidental movement when the canister is being hoisted into a VCC. According to NUREG-0554, the accidental movement is conserva tively simulated by 80 mm- and 160 mm-height free-drop motions and then with straight and 2 -oblique impact to a pedestal in VCC. A symmetric fully 3-D finite element model is built and analyzed using the explicit finite element code, LS-DYNA. Geometrical, contact, and material nonlinearities are all taken into account. The analysis result concludes that the permanent deformations of the canister are not severe to affect fuel retrieve after the impact accident and the maximum stress intensity in the canister shell can meet the ASME code appendix F F-1340, preventing the leakage of radioactive materials. The study also found that with properly reducing the wall thickness of the pedestal cylinder, the maximum acceleration and permanent deformation of the canister can be much alleviated, even though the drop height is increased to the double of the required brake distance specified in NUREG-0554. The damages of the pedestal in each analysis are moderate so that the heat transfer condition after the impact accident can be bounded by the off-normal event for half-blockage of air inlets. © 2015 Published by Elsevier Ltd.
Keywords: Dry storage canister Impact analysis Accident movement Explicit finite element
1. Introduction With the aggravation of global warming and climate change, the living with low-carbon emission and energy saving is being advocated worldwide. Having a characteristic in low-carbon emission, the nuclear energy continues to be a good option for power resource in the future. However, the handling of highly radioactive spent nuclear fuels and the safety operation during the complex and huge natural disasters like those happened in Japan 2011 strongly impacts the public confidence on the use of nuclear energy. Fortunately, the Fukushima event also proved that the dry storage system is able to successfully withstand the huge natural disasters. Spent nuclear fuels discharged from reactor core are commonly stored in spent fuel pool (SFP), not only for shielding radiation by the water in SFP but also for removing residual heat through coolant circulation. Such kind of so-called wet-storage is the way
* Corresponding author. E-mail address: [email protected] (T.-Y. Wu).
adopted by most current in-service nuclear power plants (NPP). However, most of NPPs built in 70 s and 80 s with licenses to be renewed do not have enough capacity for spent fuel storage, even though their SFP has been replaced by high-density fuel racks. The independent spent fuel storage installation (ISFSI) is thus adopted as an interim storage method before final disposal policy is determined. In US, many dry storage systems have been successfully developed and applied to practical nuclear engineering practice, such as the HI-STORM and UMS systems respectively developed by Holtec [1] and NAC [2]. Recently, the institute of nuclear energy research (INER) in Taiwan also deployed a project to develop an experimental dry storage system called HCDSS-69. It is designed in accordance with the needs of the NPPs in Taiwan and being capable of accommodating 69 bundles of BWR spent nuclear fuels. The system design and analyses are in accordance with 10 CFR72 [3], ANSI/ANS 57.9 [4], the applicable sections of the ASME Code [5], and the ACI code [6]. The design of HCDSS-69 system also follows guidance of the standard review plans (SRPs), NUREG-1536 [7] and NUREG-1567 [8] published by US NRC. The aim of SRPs is to
http://dx.doi.org/10.1016/j.ijpvp.2015.04.006 0308-0161/© 2015 Published by Elsevier Ltd.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
2
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
standardize the review processes to the license application of dry storage facilities. The SRPs clearly instruct that off-normal and accident conditions must be addressed in the safety report of ISFSI applicants. The off-normal event commonly considers the offnormal ambient temperature, off-normal handling, half-blockage of air inlet, and so on. The occurrence frequency of the events is around once per year. The accident event considers more sever accidents which may occurs only one time for entire service life of ISFSI, such as the extremely natural disasters and handling drop impact accidents. Obviously the scenarios of events should be made up in accordance with the ISFSI operational procedures. The assessment on handling impact accident of dry storage cask should be provided in the safety analysis report.
In order to provide applicants for an NRC license under 10 CFR Part 72 with a method for evaluating storage casks for low velocity impact conditions, the NRC has conducted a series of drop test studies of a solid steel billet and a near-full-scale empty cask in the Sandia National Lab. (SNL) and Lawrence Livermore National Lab. (LLNL), respectively. The experimental results reported in NUREG6608 [9] are compared and verified by those of numerical simulation using the explicit finite element code DYNA3D [10] developed in LLNL. Since then employment of explicit finite element code for waste cask drop impact simulation has become intensive. Many researchers and engineers thus employed the explicit finite element code to conduct their drop impact analyses. Huang and Wu [11] studied a VCC tip-over impact using a simplified canister model
Fig. 1. Schematics of HCDSS-69.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 2. Finite element model of HCDSS-69.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
3
4
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14 Table 1 Material properties of 304L stainless steel [5]. 304L Steel property
Value at 400 F
Mass density, r Modulus of elasticity, E Poisson's ratio, n Yield stress, sy Ultimate stress, Su 0.9Su
8.04E-9 ton/mm3 182.7 GPa 0.31 120.7 MPa 404.7 MPa 364.6 MPa
with pseudo tensor concrete plasticity. Kim et al. [12] conducted finite element analyses and tests for a shock-absorbing effect of a pad in a spent fuel storage cask. Kim et al. [13,14] further used explicit finite element procedures to simulate the drop impact of a transport cask. The simulation of drop impact is usually difficult and may have different results depending on FEA codes. Lee et al. [15] carried out a finite element analysis for a 9 m free drop and the puncture condition under the hypothetical accident condition by using LS-DYNA and ABAQUS/Explicit. Wu et al. [16] investigated the size effect of drop test using LS-DYNA. A numerical simulation is indeed able to compensate for some disadvantages of field test. For example, responses of some key locations where sensor installations are impossible but can be reported by simulations. In addition, the repeating drop angles can be achieved without any difficulties. The current main operation procedures of independent spentfuel dry storage for the NPPs in Taiwan include: (a) Spent-fuel assemblies are loaded into the canister in SFP with water depth over 10 m. (b) A transfer cask with loaded canister is then lifted from the SFP and sits on the border. After lid sealed, water drained, vacuumed and helium filled, the cask will be decontaminated and transferred to the storage site using a dolly. (c) In the phase of transportation, an impact limiter is installed on the top of the transfer cask and a cable tied down system is used to ensure the transfer cask not falling off the dolly deck under SSE level of earthquake. (d) In the storage site, the transfer cask is moved to the top of a VCC casted in-situ. Then the canister is hoisted using a single failure-proof crane to install it into the VCC. Although the single-failure proof crane can prevent the cask from drop accident during loading process, it still requires a braking distance not exceeding 3 in (80 mm), as that specified in NUREG-0554 [17]. It
implies that a drop impact accident may happen within 80 mm. This accident is frequently ignored in many ISFSI safety reports. In order to study the consequence of the accident movement during installation, dynamic impact analyses for installing a dry storage canister into a VCC are performed in this paper. The impact accident is stemming from a conservative consideration of accidental movement when the canister is being hoisted into the VCC. According to NUREG-0554 and the gap between VCC liner and canister, the accidental movement is conservatively simulated by 80 mm-height free-drop motions and with straight and 2 -oblique impacts onto a pedestal in the VCC. The minimum height of free drop is 80 mm, being adopted in accordance with the maximum braking distance of crane specified in NUREG-0554. For defense in depth, the scenario with double drop height, 160 mm is further investigated. A symmetric fully 3-D finite element model is built and analyzed using the explicit finite element code, LS-DYNA [18e20]. The main concerns of the impact accident in this work include: (a) The permanent deformations of the canister cannot be severe so that the retrieve of spent fuel is difficult. This is to meet the basic requirement of CFR72. (b) The magnitude of the stress intensity in the canister shell has to meet the ASME code appendix F, F-1340, preventing the leakage of radioactive materials. (c) The damages of the pedestal must be small enough so that the heat transfer condition after the impact accident can be bounded by the off-normal event for half-blockage of air inlets. (d) The maximum acceleration of the canister cannot be over 35 g, which is the design limit of the internal components in canister. In addition to the main concerns, this work also investigated the effect of reduced wall thickness of the pedestal cylinder to understand the possibility of optimal design for the pedestal. In addition to the Section 1 of this paper, Section 2 introduces the main features and dimensions of the HCDSS-69 dry storage system. Section 3 introduces the finite element model prepared for LS-DYNA input data. Numerical results are reported and discussed in the Section 4. Finally, Section 5 shows the conclusions and remarks. 2. Dry storage cask of HCDSS-69 The HCDSS-69 system is an experimental cask designed to accommodate 69 bundles of BWR spent nuclear fuels. The profile of
Fig. 3. The observation points in the Finite element model.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
the storage cask of HCDSS-69 system is illustrated in Fig. 1(a) and the detailed components in the canister in Fig. 1(b). The height and outer radius of the VCC are 5750 mm and 1850 mm, and those of the canister are 4840 mm and 1678 mm. Most components of the canister of HCDSS-69 are made of AISI 304 series stainless steels. The main components in the canister includes the outer shell, 4 side-support panels, fuel tube assembly, Boral neutron absorbing plates, shield lid and structural lid, as shown in Fig. 1(b). The arrangement of the fuel tubes is assembled in a compact stack form,
5
rather than the support disks used in INER-HPS [11] or NAC UMS [2]. The Boral neutron absorbing plates are embedded between the walls of two fuel tubes. The stacked tubes are finally bundled up by 4 side-support panels connected by fasteners. One design philosophy for the HCDSS-69 is that the fuel assembly must be able to be withdrawn out if underwent hypothetical drop impact accident, in order to satisfy the regulation of 10 CFR72 [3]. The bottoms of fuel tubes and neutron absorbing plates are supported and constrained by a set of spacers installed on the base plate of the canister outer
Fig. 4. Distribution of effective plastic strain.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
6
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
shell for accelerating drainage. The canister is closed by a structural lid and sealed using welding, after loading fuels and covering the shield lid. With full loading the BWR fuels, the total weight of the canister is around 35 metric tons. 3. Explicit finite element model According to NUREG 0554, the accident movement can be conservatively analyzed via 80 mm-height and free-drop induced
impact. An extended analysis on the responses of the pedestal and canister for the doubled drop height (160 mm) is also conducted based on the defense-in-depth policy. The drop impact simulation includes a straight and a 2 -oblique impact onto the steel pedestal. The angle of 2 is evaluated in accordance with the gap between the canister and VCC liner. The rigid body motion during free-drop is replaced by the equivalent initial velocities, V0 ¼ 1520 mm/s and 1772 mm/s for saving computation cost. In order to investigate the characteristic of energy absorption of the pedestal, the wall
Fig. 5. Distribution of maximum shear stress.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14 Table 3 The height change of air inlet.
0.08
Effective plastic strain (%)
0 Degree C1 C2 P1 P2
0.06
Analyzed case
CASE I
Impact angle (Deg.)
0
2
0
2
0
2
Deformed length (mm) Height change ratioa (%)
0.08 0.05
0.41 0.25
0.19 0.12
0.42 0.26
5.93 3.71
27.6 17.3
a
CASE II
CASE III
Height Change Ratio ¼ crush length/original height of the air inlet.
0.04
0.02
0 0
10
20
30
40
30
40
Time (ms)
(a) Straight impact of CASE I 0.4 2 Degree
Effective plastic strain (%)
7
C1 C2 P1 P2
0.3
0.2
0.1
0 0
10
20
Time (ms)
(b) Oblique impact of CASE I Fig. 6. Effective plastic strain responses.
thickness of pedestal is reduced from 25 mm to 10 mm in the case with released height 160 mm. Thus, totally 3 cases with 6 analyses are performed in this work. The symmetric 3-D finite element model of the canister and pedestal for straight and oblique drop impact analysis is illustrated in Fig. 2(a). The fuel tubes, neutron absorbing plates, and side support panels are modeled by shell elements, as shown in Fig. 2(b). In order to obtain more precise stress distribution, the mesh density of the outer shell and side support panels are higher
Table 2 Stress intensity and safety factor. Drop height (mm)
CASE I
Impact angle (Degree)
0
2
CASE II 0
2
CASE III 0
2
Stress intensity (MPa) Safety factor (0.9Su/Sint)
168.1 2.17
195.8 1.86
173.1 2.10
196.0 1.85
140.4 2.59
97.5 3.73
than those of other components. The average element sizes of the outer shell and the support panel are respectively 0.02 0.02 mm2 and 0.04 0.04 mm2. All the element types used in this model are fully integrated elements. The entire finite element model is built up by 118409 elements and 133035 nodes in total. The stiffness of the cask lids is much higher than those of other components, thus they are simplified as a rigid body, and using the surface tied contact function of LS-DYNA to simulate the welding between the lid and the top of outer shell wall. The modeling of contact interactions between each part is carried out using the general contact algorithm in LS-DYNA. In this contact algorithm, the contact search strategy includes global bucket sort and local contact searches. The bucket sort frequency used in this work is set to be 200/cycle. The tangential friction forces are computed by regularized MohreCoulomb's friction law with static and dynamic frictional coefficients 0.6 and 0.5. In contrast to the function of single surface contact in LS-DYNA, the automatic general contact can provide edge contact detection. It is necessary in this study. For the contact formulation details, the reader can consult the references [18] and [21]. In the drop impact process, the 69 bundles of BWR spent fuel assembly also play a crucial role due to its heavy weight. It may induce significant contact pressure on the base plate of canister for straight impact, and on the support panels for the oblique impact. However, the geometry details of the fuel assembly are very complex. Thus only a simplified geometry modeled using 3-D solid elements with equivalent mass density and estimated stiffness is used for analysis, according to the practical engineering experience provided by NAC [22]. The drop impact analyzed in this paper belongs to the category of low-velocity impact, thus the elasticeplastic material model combined with CowpereSymonds strain rate function is adopted. The detail of the material model can be found in the theory manuals of LS-DYNA [18,20] or the FEM textbook [21]. The related material properties for 304L stainless steel at 400 F temperature is listed in Table 1. The system damping ratio for the model is 7%, according to the common codes or standards for nuclear components. 4. Results and discussions The results obtained from the analyses according to the scenarios mentioned previous sections are reported and discussed in this section. The post-processor software, LS-Prepost is employed to review and post-process the data dumped by LS-DYNA. As shown in Fig. 3, five critical locations are selected for observation and judgment, including point F1 which is at the bottom of a fuel assembly closed to the side support panel, points C1 and C2 on the canister shell wall, P1 on the junction of the pedestal cylinder and the pedestal flange, and P2 on the pedestal cylinder. For expediency and simplicity, the 6 analyses are classified into 3 cases. In each case the results of the straight and 2 -oblique impacts onto the steel pedestal are reported and discussed. The stress in the canister outer shell is conservatively checked using ASME code Section III, Appendix F, F-1340 [5].
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
8
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
4.1. CASE I: 80 mm drop height, normal wall thickness of pedestal cylinder Figs. 4 and 5 respectively show the plastic deformation and maximum shear stress distributions under the conditions of this case. Under the straight impact, the plastic deformation concentrates at the junction weld between the canister outer shell and base plate, and so does the maximum shear stress. In addition, the
flange of pedestal has large deflection. On the other hand, under the oblique impact the location of plastic deformation transfers to the corner near contact area. Similar to the experimental result reported by Saegusa et al. [23], a small buckling bulge occurs at the bottom of the canister outer shell, as shown in Figs. 4(b) and 5(b). Fortunately, it is not serious to affect the function of fuel retrieve.
Fig. 7. Distribution of effective plastic strain.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 6(a) and (b) respectively plot the effective plastic strain response at the points P1, P2, C1, and C2 under the straight and oblique impacts. In the case of straight impact, Fig. 6(a) reveals that point C1 and P1 have larger plastic strain than the other 2. The location of point C1 is right at the junction weld between the canister outer shell and base plate. The maximum plastic strain at point C1 is only around 0.062%. In the condition of the 2 -oblique
9
impact, Fig. 4(b) shows the plastic strain at point P1 significantly increases to 0.39%, about 4 times of that under straight impact. It reason is that the contact area in the oblique impact is much smaller than that in the straight impact. The effective plastic strains at points C1 and C2 are also increased to 0.11% and 0.12%. In this case, the location of P2 (the pedestal cylinder) doesn't have any plastic deformation both in the straight and oblique impacts.
Fig. 8. Distribution of maximum shear stress.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
10
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Table 2 lists the maximum shear stress of the canister outer shell under straight and oblique impacts. Compared them with the requirement of ASME code Section III, Appendix F, F-1340 [5], the primary bending plus membrane stress is smaller than 0.9Su. They respectively have safety factors of 2.17 and 1.86. The result of the analyses shows that this canister is safe under the 80 mm-height free drop. It should be noted that in the real situation of crane failure, the brake system in a single-failure proof crane will be activated. Thus the simulation by free drop is conservative enough. Base on the concept of defense in depth, the additional two sets of scenario are further investigated.
4.2. CASE II: 160 mm drop height, normal wall thickness of pedestal cylinder In this case, the drop height is raised to 160 mm, doubling the initial impact energy in the previous case. Figs. 7(a),8(a) and 9(a)
0.12
Effective plastic strain (%)
0 Degree C1 C2 P1 P2
0.08
0.04
0 0
10
20
30
40
respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the straight impacts. Large deflection is found at the top flange of the pedestal due to plastic bending. None of buckling bulge is found in the pedestal cylinder and canister shell. The area of plastic zone does not significantly extend, compared to those in previous case. However, the effective plastic strains increase to 0.118% at point P1 and 0.075% at point C1, as shown in Fig. 9(a). Figs. 7(b), 8(b) and 9(b) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the oblique impacts. Compared to the CASE I, a larger bulge due to local buckling occurred at the bottom of the canister outer shell, as shown in Fig. 8(b). Fortunately, the side support penal still bundles the fuel tubes well and not to jeopardize the function of fuel retrieve. The effective plastic strains increase to 0.4% at point P1, 0.125% at point C1, and 0.21% at point C2, as shown in Fig. 9(b). In other words, the plastic deformation of point C2 becomes larger than that of C1. It reveals that a yielding line forms on the canister outer shell and becomes an impact energy absorbing mechanism. It would be detrimental to the fuel retrieve if a more violent straight impact were encountered. The local buckling of the canister outer shell can be remedied by welding together with the side support panels. Due to the limitation of manuscript page, the technical details does not discussed here. It can also be avoided by adequately adjusting the stiffness and strength of the pedestal cylinder. We will demonstrate this idea in the next case. Compared the stress intensity (2 times of maximum stress) in the canister outer shell with the requirement of ASME code Section III, Appendix F, F-1340 [5], as listed in Table 2, the cases under the straight and oblique impact respectively have safety factors of 2.10 and 1.80. The safety factors in this case slightly decreases but this canister is still safe even under the 160 mmheight free drop. 4.3. CASE III: 160 mm drop height, reduced wall thickness of pedestal cylinder
Time (ms)
(a) Straight impact of CASE II 0.5 2 Degree C1 C2 P1 P2
Effective plastic strain (%)
0.4
0.3
0.2
0.1
0 0
10
20
Time (ms)
(b) Oblique impact of CASE II Fig. 9. Effective plastic strain responses.
30
40
The previous 2 cases show that point P2, which locates on the pedestal cylinder, only slightly deforms in elastic range. In this case, the drop height is kept the same as that of CASE II, but the wall thickness of pedestal cylinder is reduced to 10 mm in order to trigger buckling deformation on the pedestal cylinder. Figs. 10(a), 11(a) and 12(a) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the straight impacts in this case. In Fig. 10(a), one can see that the area of the plastic zone in the pedestal significantly reduces. Furthermore, there is no plastic deformation in the canister outer shell, as the P2 curve shown in Fig. 12(a). In contrast, an elastic buckling deformation forms on the pedestal cylinder, absorbing the impact energy via large area of elastic large deformation (see Fig. 11(a)). Compared to those of the previous 2 cases, the effective plastic strain at point C1 and C2, i.e. the location of canister shell, drastically reduced to the values below 0.08%. Figs. 10(b), 11(b) and 12(b) respectively show the distribution of effective plastic strain, distribution of maximum shear stress, and response curves of effective plastic strain at the points observed under the oblique impacts. As expected, the plastic zones gather in the pedestal and the buckling bulge on the canister outer shell vanishes, as those reflected on the response curve C2 in Fig. 12(a). Table 2 shows that the maximum stress intensities of the canister shell in this case also satisfy the
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
11
Fig. 10. Distribution of effective plastic strain.
requirement of ASME code Section III, Appendix F, F-1340 [5]. The cases under the straight and oblique impact respectively have safety factors of 2.59 and 3.73. They are obviously higher and safer than those of previous 2 cases. The result concludes that with a proper design of pedestal, the impact force induced by vertical drop accident can be effectively alleviated, especially for the oblique impact accident. Figs. 13 and 14
respectively plot the velocity and acceleration responses of point F1. The maximum accelerations in each case are below 35 g, satisfying the design limit of the canister internal components. The response curves under the oblique impact accident reveal that CASE I and CASE III almost have the same magnitude of maximum acceleration, revealing again the performance of the pedestal cylinder with properly reduced stiffness and strength. The reduction
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
12
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Fig. 11. Distribution of maximum shear stress.
of the pedestal cylinder stiffness and strength can effectively absorb the impact energy, however the heat transfer condition have to be further aware of after the accident. INER had been completed a series of heat transfer analysis for the HCDSS-69 system under the accidents of normal, abnormal high temperature, and high ambient temperature using computational fluid dynamics [24]. The analysis pointed out that the current
design of the HCDSS-69 system is able to meet the temperature design basis of fuel assembly, shielding material, support structure and other components in above conditions. In addition, the fuel components of the maximum temperature are lower than 400 C in air inlet half-blockage accident or high ambient temperature accident. Based on this conclusion, Table 3 summarizes the height change ratio of the VCC air inlet in each case. The values listed show
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
1
0.06 0 Degree
0 Degree
C1 C2 P1
Case I Case II Case III
0.5
P2
0.04
0
Velocity (m/s)
Effective plastic strain (%)
13
0.02
-0.5
-1
-1.5
0
-2 0
10
20
30
40
0
Time (ms)
20
30
40
30
40
Time (ms)
(a) Straight impact.
(a) Straight impact of CASE III 1
0.5
2 Degree
2 Degree C1 C2 P1 P2
0.4
Case I Case II Case III
0.5
0
Velocity (m/s)
Effective plastic strain (%)
10
0.3
0.2
-0.5
-1
0.1
-1.5
0
-2 0
10
20
30
40
Time (ms)
(b) Oblique impact of CASE III Fig. 12. Effective plastic strain responses.
that the damage of air inlet is not severe. Even in the most severe case (CASE III), the maximum height change ratio is not greater than 20%. It is obvious that component temperature can be bounded by the off-normal analysis for the air inlet half-blockage condition [24].
5. Concluding remarks In this paper, a series of dynamic impact analysis for installing the HCDSS-69 canister into a VCC is presented. The impact accident focuses on the scenario of accidental movement when the canister is being hoisted into a VCC. The accidental movement is conservatively simulated by free-drop motions and straight and oblique impacts onto a steel pedestal. Some concluding remarks based on the analysis results are drawn as follows:
0
10
20
Time (ms)
(b) Oblique impact Fig. 13. Velocity responses of point F1.
(i) The maximum effective plastic strains on the canister shell are very low under the 80 mm-height free drop, regardless of the straight or oblique impact. The outer shell of the cask only has slight permanent deformation. The analyses demonstrate that the canister is safe under such conditions. It should be noted that in the real situation of crane failure, the brake system in any single-failure proof crane will be activated. Thus the simulation by 80 mm-height free drop impact is conservative enough. (ii) The result of the 3 cases performed in Section 4 shows that the permanent deformations of the canister are not severe, thus the retrieve of spent fuel is viable after the accident and the basic requirement of CFR72 can be met. Meanwhile, the stress intensity in the canister outer shell meets the ASME code appendix F F-1340, preventing the leakage of radioactive materials. In addition, the crush levels of the pedestal are
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006
14
C.-Y. Lin et al. / International Journal of Pressure Vessels and Piping xxx (2015) 1e14
Acknowledgment
40
0 Degree
Acceleration (g)
The research in this paper is under the project of Industrialization Platform Development for Nuclear Technology and sponsored by the funding of National Energy Project phase I, Ministry of Science and Technology in Taiwan (Project No. ANS040401). The authors would like to gratefully express their acknowledgement.
Case I Case II Case III
30
20
References 10
0
-10 0
10
20
30
40
30
40
Time (ms)
(a) Straight impact. 16
2 Degree Case I Case II Case III
Acceleration (g)
12
8
4
0
-4 0
10
20
Time (ms)
(b) Oblique impact Fig. 14. Acceleration responses of point F1.
low enough so that the heat transfer condition after the impact accident can be bounded by the off-normal event of half-blockage air inlets. Finally, the maximum acceleration of the canister is not over 35 g, the design limit of the internal components of the HCDSS-69 canister. In conclusion, the canister of HCDSS-69 system is safe under the accident movement of any single-failure proof crane.
[1] Holtec Inc. Topical safety analysis report for the holtec international storage and transfer operation reinforced concrete module cask system (HI-STORM 100 cask system). NRC Docket No. 72e1014. 1999. [2] NAC International Inc. Final safety analysis report for the UMS universal storage system. Docket No. 72e1015. 2005. [3] 10 CFR 72. Licensing requirements for the independent storage of spent nuclear fuel and high level radioactive waste and reactor-related greater than class C waste. Washington, DC: Code of Federal Regulations, US Government; 2007. [4] ANSI/ANS 57.9-1992. Design criteria for an independent spent fuel storage installation (Dry Type). La Grange Park, IL: American Nuclear Society; 1992. [5] ASME boiler and pressure vessel code, section III. New York, NY: American Society of Mechanical Engineers; 2004. [6] ACI-349-85. Code requirement for nuclear safety related concrete structures and commentary. Farmington Hills, MI: American Concrete Institute; 1985. [7] NUREG-1536. Standard review plan for dry cask storage systems, rev. 1. US Nuclear Regulatory Commission; 2010. [8] NUREG-1567. Standard review plan for spent fuel dry storage facilities. US Nuclear Regulatory Commission; 2000. [9] NUREG-6608. Summary and evaluation of low-velocity impact tests of solid steel billets onto concrete pads. US Nuclear Regulatory Commission; 1998. [10] Whirley RG. DYNA3D, a nonlinear, explicit, three-dimensional finite element code for solid and structural mechanics e user's manual. Lawrence Livermore National Laboratory; 1993. UCRL-MA-107254. [11] Huang CC, Wu TY. A study on dynamic impact of vertical concrete cask tipover using explicit finite element analysis procedures. Ann Nucl Energy 2009;36:213e21. [12] Kim DH, Seo KS, Lee JC, Bang KS, Cho CH, Lee SJ, et al. Finite element analyses and verifying tests for a shock-absorbing effect of a pad in a spent fuel storage cask. Ann Nucl Energy 2007;34:871e82. [13] Kim KS, Kim JS, Choi KS, Shin TM, Yun HD. Dynamic impact characteristics of KN-18 SNF transport cask-part 1: an advanced numerical simulation and validation technique. Ann Nucl Energy 2010;37:546e59. [14] Kim KS, Kim JS, Choi KS, Shin TM, Yun HD. Dynamic impact characteristics of KN-18 SNF transport cask-part 2: sensitivity analysis of modeling and design parameters. Ann Nucl Energy 2010;37:560e71. [15] Lee YS, Ryu CH, Kim HS, Choi YJ. A study on the free drop impact of a cask using commercial FEA codes. Nucl Eng Des 2005;235:2219e26. [16] Wu TY, Lee SY, Kang LC. A study on dynamic responses of a spent-fuel dry storage cask under vertical drop accident. Ann Nucl Energy 2012;42:18e29. [17] NUREG-0554. Single-failure cranes for nuclear power plant. US Nuclear Regulatory Commission; 1998. [18] Hallquist JO. LS-DYNA theory manual. Livermore Software Technology Corporation; 2006. [19] Livermore Software Technology Corporation. LS-DYNA keyword user's manual version 971. 2012. [20] Livermore Software Technology Corporation. LS-DYNA material models. 2012. [21] Belytschko T, Liu WK, Moran B. Nonlinear finite elements nonlinear finite elements for continua and structures. 1st ed. Wiley; 2000. [22] Stephenson RW, Griswold M. Evaluation of the loaded INER transfer cask during a tip-over accident condition. Calculation report, Cal. No. 63005100e2002. NAC International Inc.; 2002. [23] Saegusa KT, Shirai K, Arai T, Tani J, Takeda H, Wataru M, et al. Review and future issues on spent nuclear fuel storage. Nucl Eng Technol 2010;42: 237e48. [24] Huang JJ, Lee SY. High capacity dry storage system-the heat transport analysis of 69 dry storage system and new type vertical concrete cask. INER report No. INER-10358. 2013.
Please cite this article in press as: Lin C-Y, et al., Nonlinear dynamic impact analysis for installing a dry storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping (2015), http://dx.doi.org/10.1016/j.ijpvp.2015.04.006