LQR control with frequency‐dependent scheduled gain for a semi‐active floor isolation system

Floor isolation is an alternative to base isolation for protecting a specific group of equipment installed on a single floor or room in a fixed‐base structure. The acceleration of the isolated floor should be mitigated to protect the equipment, and the displacement needs to be suppressed, especially under long‐period motions, to save more space for the floor to place equipment. To design floor isolation systems that reduce acceleration and displacement for both short‐period and long‐period motions, semi‐active control with a newly proposed method using the linear quadratic regulator (LQR) control with frequency‐dependent scheduled gain (LQRSG) is adopted. The LQRSG method is developed to account for the frequency characteristics of the input motion. It updates the control gain calculated by the LQR control based on the relationship between the control gain and dominant frequency of the input motion. The dominant frequency is detected in real time using a window method. To verify the effectiveness of the LQRSG method, a series of shake table tests is performed for a semi‐active floor isolation system with rolling pendulum isolators and a magnetic‐rheological damper. The test results show that the LQRSG method is significantly more effective than the LQR control over a range of short‐period and long‐period motions. Copyright © 2013 John Wiley & Sons, Ltd.     

Petrographic study of the Miocene Mizunami Group,Central Japan: Detection of unrecognized volcanic activity in the Setouchi Province

A petrographic study of sandstones from the Miocene Mizunami Group in Central Japan has been performed on core samples from a single borehole, in order to evaluate the provenance of the sedimentary rock. Evaluation of the provenance is based on bulk mineral, heavy mineral and plagioclase contents and on whole rock chemical compositions. The sandstones studied are divisible into three types; the first type is characterized by the occurrence of biotite and plagioclase ranging from albite to oligoclase, the second type is characterized by the dominance of amphibole and labradorite with pyroxene (clinopyroxene > orthopyroxene), and the third type is characterized by the dominance of pyroxene (orthopyroxene > clinopyroxene) and andesine with lesser labradorite, bytownite and anorthite. The first type is interpreted to be derived from the basement granite, whereas the others were derived mostly from volcanic ash, judging from their mineral compositions. The volcanic activity that supplied the volcanic ash to the Mizunami Basin occurred in two phases, distinguishable by variations in their mineralogical and geochemical compositions, an indication of change in character of the volcanic activity. This petrographic study of the sandstones in the Mizunami Group suggests that unrecognized volcanic activity occurred around the Mizunami Basin, even though potential provenance of the volcanic ash has not yet been identified.     

Observation of cosmic soft X-rays

Cosmic soft X-rays in the energy range between 0.14 and 7 keV were observed with thin polypropylene window proportional counters on board a sounding rocket. The field of view crossed the galactic plane in the Cygnus-Cassiopeia region at a large angle and reached the galactic latitudes of –55° and +30°. Referring also to the result with Be window counters, we obtained the energy spectrum of Cyg XR-2, the flux from the Cas A region and the distribution of the intensity of diffuse X-rays over the scanned region. The turn-over of the Cyg XR-2 spectrum at about 1 keV indicates that the distance of the Cyg XR-2 source lies between 600 and 800 pc, if the turn-over is due entirely to interstellar absorption. The flux from the Cas A region is obtained as 0.23±0.05 photons cm^–2 sec^–1 in the energy range between 1.1 and 4.1 keV. The intensity of diffuse soft X-rays depends on the galactic latitude more weakly than expected from the interstellar absorption of extragalactic X-rays and shows asymmetry with respect to the galactic equator, thus suggesting a contribution of galactic X-rays. The spectrum of extragalactic X-rays is approximately represented by a power lawE ^–1.8.     

Distribution of iron on Vesta

We have completed a mapping study of 7.6 MeV gamma rays produced by neutron capture by Fe at the surface of the main belt asteroid 4 Vesta as measured by the bismuth germanate scintillator of the Gamma Ray and Neutron Detector (GRaND) on the Dawn spacecraft. The procedures used to determine Fe counting rates are presented, along with a global map, constituting the necessary initial step to quantify Fe abundances. While the final calibration of orbital data to absolute concentrations has not been determined, the range of fully corrected Fe counting rates is compared with that of Fe in howardites. We find that the global distribution of corrected Fe counting rates is generally consistent with mineralogy and composition determined independently by other instruments on the Dawn spacecraft, including measurements of pyroxene absorption bands by the Visible and Infrared Spectrometer and Framing Camera, and an index of diogenitic materials provided by neutron absorption measurements by GRaND. In addition, there is a distinctive low Fe region in the western hemisphere that was not reported by reflectance or optical observations, possibly indicating the presence of a cumulate eucrite component in Vesta's regolith.     

Yellow fluorescence from baghdadite and synthetic Ca3(Zr,Ti)Si2O9

Baghdadite from Fuka, Okayama Prefecture, Japan shows a bright yellow fluorescence under UV (Hg 253.7 nm) excitation. The photoluminescence (PL) spectrum at 300 K consists of one large band near 580 nm and two small UV bands at 318 and 397 nm. The optical excitation spectrum of the bright yellow fluorescence consists of two bands near 220 and 250 nm. The temperature dependence of the PL intensity exhibits linear thermal quenching. To reveal the origin of the bright yellow fluorescence from baghdadite, powder Ca₃(Zr,Ti)Si₂O₉ crystals are synthesized. Synthetic Ca₃(Zr,Ti)Si₂O₉ shows luminescence spectra similar to those of baghdadite, and the intensity of the yellow fluorescence is markedly increased by titanium addition. The origin of the bright yellow fluorescence from baghdadite is ascribed to the existence of titanium.     

Cohesive zone length of metagabbro at supershear rupture velocity

We investigated the shear strain field ahead of a supershear rupture. The strain array data along the sliding fault surfaces were obtained during the large-scale biaxial friction experiments at the National Research Institute for Earth Science and Disaster Resilience. These friction experiments were done using a pair of meter-scale metagabbro rock specimens whose simulated fault area was 1.5 m?×?0.1 m. A 2.6-MPa normal stress was applied with loading velocity of 0.1 mm/s. Near-fault strain was measured by 32 two-component semiconductor strain gauges installed at an interval of 50 mm and 10 mm off the fault and recorded at an interval of 1 MHz. Many stick-slip events were observed in the experiments. We chose ten unilateral rupture events that propagated with supershear rupture velocity without preceding foreshocks. Focusing on the rupture front, stress concentration was observed and sharp stress drop occurred immediately inside the ruptured area. The temporal variation of strain array data is converted to the spatial variation of strain assuming a constant rupture velocity. We picked up the peak strain and zero-crossing strain locations to measure the cohesive zone length. By compiling the stick-slip event data, the cohesive zone length is about 50 mm although it scattered among the events. We could not see any systematic variation at the location but some dependence on the rupture velocity. The cohesive zone length decreases as the rupture velocity increases, especially larger than \( \sqrt{2} \) times the shear wave velocity. This feature is consistent with the theoretical prediction.     

Characteristics of an atypical large-meander path of the Kuroshio current south of Japan formed in September 2017

Since September 2017, the Kuroshio has taken a large-meander (LM) path in the region south of Japan. We examined characteristics of the 2017–present LM path in comparison with previous LM paths, using tide gauge, altimetric sea surface height, and bottom pressure data. The 2017–present LM path was formed from a path passing through a channel south of Hachijo-jima Island, while a typical LM path originated from a path through a channel north of Miyake-jima Island. The meander trough of this atypical path was found to be shifted far to the east and to vary on a timescale of months. These characteristics are different from those of a typical LM path but they are similar to those of the 1981–1984 LM path. Therefore, we identified two types of LM path; a stable and unstable LM paths. The 2017–present unstable type large meander has a zonal scale greater than that of the 2004–2005 stable type large meander and protrudes from the eastern boundary of the Shikoku Basin, i.e., Izu-Ogasawara Ridge. No significant bottom pressure depression was observed, associated with the formation of the 2017–present LM path, indicating that baroclinic instability was not important in the formation of this LM path. Due to no significant bottom steering, even during the 2017–present LM period, a mesoscale current path disturbance occurred southeast of Kyushu, propagated eastward, and amplified the offshore displacement of the Kuroshio.

     

Responses of polar mesospheric cloud brightness to stratospheric gravity waves at the South Pole and Rothera,Antarctica

We present the first observational proof that polar mesospheric cloud (PMC) brightness responds to stratospheric gravity waves (GWs) differently at different latitudes by analyzing the Fe Boltzmann lidar data collected from the South Pole and Rothera (67.5°S, 68.0°W), Antarctica. Stratospheric GW strength is characterized by the root-mean-square (RMS) relative density perturbation in the 30–45 km region and PMC brightness is represented by the total backscatter coefficient (TBC) in austral summer from November to February. The linear correlation coefficient (LCC) between GW strength and PMC brightness is found to be +0.09 with a 42% confidence level at the South Pole and ?0.49 with a 98% confidence level at Rothera. If a PMC case potentially affected by a space shuttle exhaust plume is removed from the Rothera dataset, the negative correlation coefficient and confidence level increase to ?0.61 and 99%, respectively. The Rothera negative correlation increases when shorter-period waves are included while no change is observed in the South Pole correlation. Therefore, observations show statistically that Rothera PMC brightness is negatively correlated with the stratospheric GW strength but no significant correlation exists at the South Pole. A positive correlation of +0.74 with a confidence level of 99.98% is found within a distinct subset of the South Pole data but the rest of the dataset exhibits a random distribution, possibly indicating different populations of ice particles at the South Pole. Our data show that these two locations have similar GW strength and spectrum in the 30–45 km region during summer. The different responses of PMC brightness to GW perturbations are likely caused by the latitudinal differences in background temperatures in the ice crystal growth region between the PMC altitude and the mesopause. At Rothera, where temperatures in this region are relatively warm and supersaturations are not as large, GW-induced temperature perturbations can drive subsaturation in the warm phase. Thus, GWs can destroy growing ice crystals or limit their growth, leading to negative correlation at Rothera. Because the South Pole temperatures in the mesopause region are much colder, GW-perturbed temperature may never be above the frost point and have less of an impact on crystal growth and PMC brightness. The observed phenomena and proposed mechanisms above need to be understood and verified through future modeling of GW effects on PMC microphysics and ray modeling of GW propagation over the South Pole and Rothera.     

Phase relations and melting of carbonated peridotite between 10 and 20 GPa: a proxy for alkali- and CO<Subscript>2</Subscript>-rich silicate melts in the deep mantle

We determined the melting phase relations, melt compositions, and melting reactions of carbonated peridotite on two carbonate-bearing peridotite compositions (ACP: alkali-rich peridotite + 5.0 wt % CO₂ and PERC: fertile peridotite + 2.5 wt % CO₂) at 10–20 GPa and 1,500–2,100 °C and constrain isopleths of the CO₂ contents in the silicate melts in the deep mantle. At 10–20 GPa, near-solidus (ACP: 1,400–1,630 °C) carbonatitic melts with < 10 wt % SiO₂ and > 40 wt % CO₂ gradually change to carbonated silicate melts with > 25 wt % SiO₂ and < 25 wt % CO₂ between 1,480 and 1,670 °C in the presence of residual majorite garnet, olivine/wadsleyite, and clinoenstatite/clinopyroxene. With increasing degrees of melting, the melt composition changes to an alkali- and CO₂-rich silicate melt (Mg# = 83.7–91.6; ~ 26–36 wt % MgO; ~ 24–43 wt % SiO₂; ~ 4–13 wt % CaO; ~ 0.6–3.1 wt % Na₂O; and ~ 0.5–3.2 wt % K₂O; ~ 6.4–38.4 wt % CO₂). The temperature of the first appearance of CO₂-rich silicate melt at 10–20 GPa is ~ 440–470 °C lower than the solidus of volatile-free peridotite. Garnet + wadsleyite + clinoenstatite + carbonatitic melt controls initial carbonated silicate melting at a pressure < 15 GPa, whereas garnet + wadsleyite/ringwoodite + carbonatitic melt dominates at pressure > 15 GPa. Similar to hydrous peridotite, majorite garnet is a liquidus phase in carbonated peridotites (ACP and PERC) at 10–20 GPa. The liquidus is likely to be at ~ 2,050 °C or higher at pressures of the present study, which gives a melting interval of more than 670 °C in carbonated peridotite systems. Alkali-rich carbonated silicate melts may thus be produced through partial melting of carbonated peridotite to 20 GPa at near mantle adiabat or even at plume temperature. These alkali- and CO₂-rich silicate melts can percolate upward and may react with volatile-rich materials accumulate at the top of transition zone near 410-km depth. If these refertilized domains migrate upward and convect out of the zone of metal saturation, CO₂ and H₂O flux melting can take place and kimberlite parental magmas can be generated. These mechanisms might be important for mantle dynamics and are potentially effective metasomatic processes in the deep mantle.     

Attenuation and dispersion ofSH waves due to scattering by randomly distributed cracks

The effect of randomly distributed cracks on the attenuation and dispersion ofSH waves is theoretically studied. If earthquake ruptures are caused by sudden coalescence of preexisting cracks, it will be crucial for earthquake prediction to monitor the temporal variation of the crack distribution. Our aim is to investigate how the property of crack distribution is reflected in the attenuation and dispersion of elastic waves.We introduce the stochastic property, in the mathematical analysis, for the distributions of crack location, crack size and crack orientation. The crack size distribution is assumed to be described by a power law probability density (p(a) a ^– fora _minaa _max according to recent seismological and experimental knowledge, wherea is a half crack length and the range 13 is assumed. The distribution of crack location is assumed to be homogeneous for the sake of mathematical simplicity, and a low crack density is assumed. The stochastic property of each crack is assumed to be independent of that of the other cracks. We assume two models, that is, the aligned crack model and the randomly oriented crack model, for the distribution of crack orientation. All cracks are assumed to be aligned in the former model. The orientation of each crack is assumed to be random in the latter model, and the homogeneous distribution is assumed for the crack orientation. The idea of the mean wave formalism is employed in the analysis, and Foldy's approximation is assumed.We observe the following features common to both the aligned crack model and the randomly oriented crack model. The attenuation coefficientQ ^–1 decays in proportion tok ^–1 in the high frequency range and its growth is proportional tok ² in the low frequency range, wherek is the intrinsic wave number. This asymptotic behavior is parameter-independent, too. The attenuation coefficientQ ^–1 has a broader peak as increases and/ora _min/a _max decreases. The nondimensional peak wave numberk _p a _max at whichQ ^–1 takes the peak value is almost independent ofa _min/a _max for =1 and 2 while it considerably depends ona _min/a _max for =3. The phase velocity is almost independent ofk in the rangeka _max<1 and increases monotonically ask increases in the rangeka _max>1. While the magnitude ofQ ^–1 and the phase velocity considerably depend on the orientation of the crack in the aligned crack model, the above feature does not depend on the crack orientation.The accumulation of seismological measurements suggests thatQ ^–1 ofS waves has a peak at around 0.5 Hz. If this observation is combined with our theoretical results onk _p a _max, the probable range ofa _max of the crack distribution in the earth can be estimated for =1 or 2. If we assume 4 km/sec as theS wave velocity of the matrix medium,a _max is estimated to range from 2 to 5 km. We cannot estimatea _max in a narrow range for =3.