Shape Memory Alloy Based Dampers for Earthquake Response Mitigation

Shape Memory Alloy Based Dampers for Earthquake Response Mitigation

Available online at www.sciencedirect.com ScienceDirect Available online www.sciencedirect.com Available online at at www.sciencedirect.com Structur...

1MB Sizes 0 Downloads 2 Views

Recommend Documents

No documents
Available online at www.sciencedirect.com

ScienceDirect

Available online www.sciencedirect.com Available online at at www.sciencedirect.com Structural Integrity Procedia 00 (2017) 000–000

ScienceDirect ScienceDirect

Procedia Structural (2017) 705–712 Structural IntegrityIntegrity Procedia500 (2016) 000–000

www.elsevier.com/locate/procedia

www.elsevier.com/locate/procedia

2nd International Conference on Structural Integrity, ICSI 2017, 4-7 September 2017, Funchal, Madeira, Portugal

Shape Memory Alloy BasedPCF Dampers for Earthquake XV Portuguese Conference on Fracture, 2016, 10-12 February 2016, PaçoResponse de Arcos, Portugal Mitigation Thermo-mechanical modeling of a high pressure turbine blade of an a J. Moraisa*, P. Gil deairplane Moraisa, C.gas Santos , A. Campos Costab, P. Candeiasb turbine engine a Scientific

Scientific Instrumentation Centre, National Laboratory for Civil Engineering (LNEC), a b c Av. do Brasil 101, Lisboa, 1700-066 Portugal b Structures Department, LNEC a Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal b IDMEC, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal Abstract c CeFEMA, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal

P. Brandão , V. Infante , A.M. Deus *

This paper describes the development process and initial tests performed with a new energy dissipation damper based on Shape Memory Alloy (SMA) wires. The aim of this study was to develop a new iteration of this type of devices, and eventually develop a methodology to properly design them for any type of application. The underlying concept of our device is the use of a double Abstract counteracting system of pre-strained SMA wire sections as the dissipating component. By using pre-strained wires, this design focuses ontheir maximizing energy dissipation, of the device. operating conditions, During operation, modern aircraftpartially engine relinquishing components the are re-centering subjected tocapabilities increasingly demanding especially the high turbine on (HPT) blades. Such conditions cause these partsmethodology. to undergo different types of time-dependent The experimental partpressure was performed a downscaled prototype based on this design The goal of this study was to degradation, onemechanical of which isconcepts. creep. AThe model using thesubjected finite element method (FEM) was of developed, in order abletotobetter predict validate the basic device was to a considerable number load cycling tests,toinbe order the creep the behaviour of HPT blades. Flight records (FDR) and for to a specific provided by a commercial aviation characterize SMA wire behavior when used data in this arrangement improve aircraft, our understanding of their influence on the company, were used to obtain thermal and mechanical data for three different flight cycles. In order to create the 3D model device’s capabilities. forAuthors. the FEM analysis,byaElsevier HPT blade © needed 2017 The Published B.V. scrap was scanned, and its chemical composition and material properties were © 2017 The Authors. Published bygathered Elsevier B.V. obtained. The data that was was fed into the FEMofmodel and different simulations were run, first with a simplified 3D Peer-review under responsibility of the Scientific Committee ICSI 2017. Peer-review under responsibility the Scientific of ICSI 2017 rectangular block shape, in of order to better Committee establish the model, and then with the real 3D mesh obtained from the blade scrap. The overall expected behaviour in terms of displacement was observed, in particular at the trailing edge of the blade. Therefore such a Keywords: Vibration Damper; Shape Memory Alloy; Earthquake Response. model can be useful in the goal of predicting turbine blade life, given a set of FDR data. © 2016 The Authors. Published by Elsevier B.V.

1. Peer-review Introduction under responsibility of the Scientific Committee of PCF 2016. Passive control techniques have shown to beElement an effective when aimed at structural preservation for seismic Keywords: High Pressure Turbine Blade; Creep; Finite Method;strategy 3D Model; Simulation. events. These systems are designed to eliminate or at least to reduce structural damage on buildings and infrastructures

* Corresponding author. Tel.: +351 218443978; fax: +351 218443041. E-mail address: [email protected] 2452-3216 © 2017 The Authors. Published by Elsevier B.V. Peer-review underauthor. responsibility the Scientific Committee of ICSI 2017. * Corresponding Tel.: +351of218419991. E-mail address: [email protected] 2452-3216 © 2016 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the Scientific Committee of PCF 2016. 2452-3216  2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the Scientific Committee of ICSI 2017 10.1016/j.prostr.2017.07.048

706 2

J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

by limiting the transmitted displacement (Seismic Isolation techniques) or by absorbing the energy of the seismic event (Energy Dissipation techniques). One energy dissipation technique consists on using dampers based on Shape Memory Alloy (SMA) wires. Shape memory alloys have many interesting properties that can be exploited in these applications, namely their Superelasticity, high fatigue resistance, near strain-rate independence (in certain conditions related to temperature control), among others [Dolce and Cardone (2001)]. Superelasticity is the main property explored in this type of devices and represents the material’s capability to change between metallurgical phases (austenitic and martensitic phases) due to stress application cycles. This property allows SMA based dampers to withstand very large strains (when compared to dampers based on other metallic materials) without any residual deformation upon unloading, while dissipating energy during the loading/unloading cycles. Fig. 1 represents a generic stress–strain tensile curve of a Superelastic SMA material, for the case where the SMA is above its austenitic phase transformation temperature (Af in the literature), i.e. the SMA material is in the austenitic phase at room temperature. This metallurgical state provides better properties for this type of dampening application [Dolce et al. (2000)]. Zone A and C correspond to the elastic deformation of the austenitic and martensitic phases, respectively. Zone B is the forward transformation plateau where the material is changing from the austenitic phase to the martensitic phase due to applied stress. Zone D represents the plastic deformation of the martensitic phase, after which if the SMA is unloaded there will be some residual strain (zone E). Zone F and G refer to the unloading path of the SMA while it still is in the elastic deformation region. Here the material recovers in the martensitic phase (F) and then begins the inverse phase transformation (G) when a certain stress level is reached, until it returns to its original austenitic phase with no residual strain.

Fig. 1. Generic stress-strain response of a SMA above temperature Af.

Based on the previous graphic, the following features of a SMA based damper can be discerned:  Energy dissipation: since the forward and inverse transformation take place at different stress levels (zones B and G), a hysteretic cycle occurs. The energy dissipated by the SMA material is equal to the area under the hysteretic cycle [Dolce et al. (2000)]. This reduces the amount of energy transmitted to the structure under protection.  Control of the transmitted force: because the forward transformation plateau (B) has a relative low slope, the device can withstand a large range of deformation under near constant stress. Hence the transmitted load can be controlled based on the device’s characteristics.  Multiple stiffness stages: due to the erratic and unpredictable nature of seismic activity, it can be advantageous to have a device that: i) can stop undesirable movements for low load levels, due to external actuations on the structure unrelated to seismic activity like the wind (initial stiffness from zone A); ii) allow a good range of displacement for moderate earthquakes under near constant load (B); and iii) also resist further displacement of the structure if subjected to high intensity seismic activity (stiffness from zone C). These topics are some of the features that make SMA based dampers a good alternative to more conventional solutions. The final goal of this study is to devise several damper configurations, each tuned to a different set of



J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

707 3

parameters (force levels, displacement amplitudes, shape of the hysteresis cycles) by choosing the arrangement, geometry and number of the SMA wires used in the device. 2. Damper concept description Austenitic SMA wires have been the subject of several works devoted to the study of their mechanical proprieties, particular characteristics and potential usages [Dolce et al. (2000), Zhang and Zhu (2006), Motahari et al. (2007)]. In this work Nitinol austenite SMA wire from MEMRY was used (Ø1.75 mm). Fig. 2 shows its characteristic stressstrain curve from a series of tensile tests performed at LNEC for a previous phase of this study.

Fig. 2. Stress-strain curve of the SMA wire used in this study.

The passive energy dissipation damper developed for this study uses two bundles of these SMA wires in a double counteracting arrangement, with pre-strain applied to both wire bundles. We chose this type of configuration because it focuses on maximizing energy dissipation while maintaining the external load (measurable load exerted by the damper when actuated) on predictable levels [Dolce and Cardone (2001)]. This type of damper configuration has the following features:  Energy dissipation: the double counteracting wire arrangement allows the system to dissipate energy when the damper is under traction and compression, from an external point of view. Intrinsically, only the wires being stretched are dissipating energy, but due to this configuration there is always one bundle dissipating energy.  Predictable external load: due to the Superelastic behavior of the SMA wires, combined with the double counteracting wire configuration, the external measurable load of the damper is always the difference between the load levels of both wire bundles (see Fig. 3). Intrinsically, each wire bundle is describing its corresponding stressstrain curve as the seismic movement occurs (one bundle is stretched, while the other is shortened, thus increasing and decreasing tensile stress, respectively). This creates a hysteretic cycle that is the basis of the dissipation mechanism for this type of SMA based dampers [Dolce and Cardone (2001)].

J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

708 4

EP

SP

SP Transformation Plateaus

Fig. 3. Simulated stress-strain curves of both wire bundles performing the hysteretic cycle (left) and the combined external result (right). This loading path represents the basic dissipation mechanism of this device. SP is the starting strain point for both wire bundles and EP is the final strain point. These graphics are obtained by a very basic spreadsheet simulator created by the author.

 Pre-strain: the damping effect on this type of Superelastic systems can be enhanced by introducing pre-strain in the SMA wires [Dolce and Cardone (2001)]. This initial strain applied to the Superelastic wires guarantees that they start and also work in the martensitic transformation plateau, even for low displacement values, thus improving the overall damping effect of the damper. Both wire bundles are pre-strained to the middle of the phase transformation plateau in order to maximize the usable range of the device. This configuration also exhibits the re-centering effect, albeit at a lower level. The re-centering effect refers to the device’s capability to return to the central position after an external actuation, when the load is released, similar to a regular spring [Dolce et al. (2000)]. But unlike a spring, due to the non-linear Superelastic behavior of the SMA, with this configuration there are multiple possible stable positions (internal force equilibrium), depending on the previous actuation path(see examples on Fig. 4). Another feature of this configuration, related to the Superelastic behavior, is that if the device is forced to the central position, there is a high probability that it will retain some residual stress, unless a specific loading path is used beforehand.

Fig. 4. Simulated stress-strain curves illustrating the re-centering effect of the device. Each illustrated curve represents the loading path of each wire bundle. SP is the starting strain point for both wire bundles. The device is in internal force equilibrium when both wire bundles stop around the horizontal line (same load on both wire bundles), situation visible on all three cases.

In order to begin the validation process of this device and to better understand the SMA material behavior, we began by building a prototype version of the damper, on a smaller scale (see Fig. 5). The experimental tests and results obtained with this prototype are the main focus of this paper. This down-scaled device has the same features of the



J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

709 5

full-scaled damper. The main difference is that it has fewer and shorter wires, thus reducing the allowable load and displacement ranges without compromising the validation goal of the device.

Fig. 5. (left) virtual representation of the fully assembled prototype and (right) a photograph of the manufactured device.

The device prototype is externally similar to a hydraulic cylinder with a tubular body, two cylinder heads and a rod. Both heads have lateral extensions to serve as anchor points to each wire bundle. The other end of each wire bundle anchors on similar extensions on two internal pistons. The piston extensions pass through windows opened on the sides of the tubular body. Each bundle is arranged on perpendicular planes aligned with the longitudinal axis of the device to get a more compact arrangement for a given wire length. The internal pistons are connected to the rod with opposite twisting threads, so that by rotating the rod the initial pre-strain can be simultaneously applied to both wire bundles. One of the cylinder heads and the opposite end of the rod have a swivel head connector. Relative motion between these points stretches one of the wire bundles and shortens the other by the same amount, thus performing the intended damping purpose. 3. Experimental tests The experimental validation process of the damper began with a series of cyclic traction/compression tests. At this stage no tests simulating seismic conditions were performed. The tests were conducted on a servohydraulic testing machine (SCHENCK HYDROPULS PSB – 500 kN) endowed with a displacement control (Fig. 6). A load cell and a displacement transducer were used to measure the external load and the applied displacement, respectively. The analog signals produced by the transducers, after have being amplified and filtered in a signal conditioning module, designed and built at LNEC, were measured and recorded using a NI 9215 data acquisition system.

Fig. 6. (left) damper prototype ready to be tested, (right) prototype mounted on the testing machine.

J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

710 6

Table 1 shows the set of parameters used for each series of tests. Each series was performed with three repetitions and an average of 10 cycles per repetition, at room temperature (average 22.5 ºC). The table also shows the average measured strain-rate for each test series. Table 1: set of parameters used in each test series. Test Frequency (Hz)

Test Amplitude (peak to peak) (mm)

Average Strain-Rate (%/s)

0,05

6

0,16

0,05

8

0,21

0,05

10

0,27

0,05

12

0,33

0,1

6

0,32

0,1

8

0,43

0,1

10

0,54

0,1

12

0,64

0,5

6

1,59

0,5

8

2,13

0,5

10

2,69

0,5

12

3,22

1

6

3,21

1

8

4,29

1

10

5,37

1

12

6,42

4. Results The main goal of these tests was to validate the proposed damper configuration, in terms of its mechanical performance as well as its proficiency as a damping device. Mechanically the device behaved according to expectations, provided that it is adequately lubricated and properly aligned with the press clamps. To determine its performance as a damping device, the obtained results were analyzed and compared to ultimately calculate the equivalent viscous damping ratio (ξ) of the device, according to formulae (1) [Dolce et al. (2000)], where ED is the energy dissipated in each loading cycle and ES is the maximum strain energy.

 

1 ED 4 ES

(1)

Fig. 7 illustrates the mechanical behaviour of the device as a function of the applied displacement amplitude and cyclic frequency by comparing different test series. The first graphic shows that the resulting load is highly dependent on the applied displacement amplitude, while the second graphic illustrates that the device has lower sensitivity to frequency changes.



J. Morais al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ et Structural Integrity Procedia 00 (2017) 000–000

7117

Fig. 7. Stacked stress-strain curves of several test series. The left graphic represents all test series performed with a cyclic frequency of 0.5 Hz and the right graphic represents all the test series realized with displacement amplitude of 10 mm.

Finally, Fig. 8 illustrates the device’s sensitivity to displacement amplitude (here shown as strain amplitude) and strain-rate relative to it’s damping ratio. Here we can see that both parameters have a decreasing effect on the equivalent viscous damping ratio of the device, in view of the specific conditions of the tests (room temperature, applied amplitude and cyclic frequency ranges, specific SMA tested, among others):

Fig. 8. Influence of strain amplitude (left) and strain-rate (right) on the device’s damping ratio.

 Strain amplitude sensitivity: as the strain amplitude increases, the ED also increases as expected. But since ES grows proportionally more than ED, the ratio ED/ES decreases and so does the damping ratio. This is true for the specific SMA wire in use. Other alloys with a lower martensitic transformation slope can attenuate this effect and even invert this tendency [as we believe was the case in Dolce et al. (2000)].  Strain-rate sensitivity: as expected, higher strain-rates result in lower damping ratios. This is due to the temperature changes during the test cycles. These temperature variations occur because the forward martensitic transformation is an exothermic process and the inverse transformation is an endothermic process [Dolce et al. (2000)]. This temperature differential influences the stress level associated with the forward and inverse martensitic transformation, changing the overall shape of the hysteretic cycle. ○ Thermal readings during the tests were carried out with a thermographic recorder. The maximum temperature reached was around 35ºC and the minimum was close to 20 ºC (see Fig. 9).

712 8

J. Morais et al. / Procedia Structural Integrity 5 (2017) 705–712 Morais J et al./ Structural Integrity Procedia 00 (2017) 000–000

Fig. 9. Thermographic image from one of the tensile tests.

5. Conclusions On this paper, we presented a new iteration of a SMA wire based damper for earthquake response mitigation. This concept was thoroughly explained and the manufactured prototype was submitted to a preliminary validation process. This device exhibited four useful features, suitable for this type of application:  Energy dissipation features, due to the SMA Superelastic effect and the double counteracting wire arrangement;  Predictable external load and multiple stiffness stages, allowing a proper configuration of the device for any particular application;  Control of the transmitted force and wide amplitude range, making it a suitable solution for a wide range of different earthquake categories;  Re-centering capability, to bring back the structural system to its initial configuration when the earthquake action is over. This device is our iteration on the concept, which performed well, both in terms of its mechanical behavior as well as in its role as an energy dissipating device. Future work on this subject will include submitting the device to earthquake simulation tests, devise a solution to dissipate the heat generated during the loading cycle and eventually build the full scale version of the SMA damper. References Dolce, M., Cardone, D., 2001. Mechanical behaviour of shape memory alloys for seismic applications, International Journal of Mechanical Sciences. Dolce, M., Cardone, D., Marnetto, R., 2000. Implementation and testing of passive control devices based on shape memory alloys, EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS. Motahari, S., Ghassemieh, M., Abolmaali, S., 2007. Implementation of shape memory alloy dampers for passive control of structures subjected to seismic excitations, Journal of Constructional Steel Research. Zhang, Y., Zhu, S., 2006. A shape memory alloy-based reusable hysteretic damper for seismic hazard mitigation, SMART MATERIALS AND STRUCTURES.