The core of the problem lies in a mismatch between what traditional seismic metrics measure and what actually breaks in an earthquake. For decades, engineers have gauged ground motion severity using quantities such as peak ground acceleration, or PGA, and correlated them with the inter-story drift ratio, the sideways deformation between adjacent floors. That pairing works reasonably well when the question is whether the structural skeleton will collapse. But the modern philosophy of seismic design, increasingly framed around functional recovery, asks a different question: after the shaking stops, can the building still be used? Answering that question means caring about nonstructural elements, from elevator machinery to medical equipment, and those components respond not to drift but to the acceleration and velocity of the floor slabs they sit on.

This distinction matters enormously in super-tall buildings, where the dynamic behavior is dominated by higher vibration modes. A building with a fundamental period of several seconds does not simply sway back and forth as a single unit; its floors oscillate in complex patterns, and accelerations can amplify dramatically at upper levels in ways that bear little relationship to the peak acceleration recorded at the ground surface. Scaling ground motions to match PGA, a common practice in time-history analysis, therefore produces a poor and inefficient predictor of floor-level demands. Engineers compensating for this scatter must run many more analyses, each computationally expensive for a detailed model of a 489-meter tower, driving up the cost and time of every performance assessment.

Cao and Qu’s solution is a scalar intensity measure rooted in the spectral acceleration of the ground motion, but weighted and combined in a way that explicitly reflects both peak floor acceleration and peak floor velocity demands. Spectral acceleration describes the maximum response of a single-degree-of-freedom oscillator of a given period to a ground motion, and averaged spectral quantities have proven powerful for collapse assessment in prior research. The authors extended this idea by calibrating the key parameters of their measure against an enormous body of simulation data: approximately 100,000 time-history analyses of building models whose fundamental periods ranged from one second to ten seconds, covering the full span of tall and super-tall structural behavior.

The calibration campaign is what gives the new measure its teeth. By regressing floor-level demands against candidate intensity measures across this vast library of ground-motion-and-building combinations, the researchers could identify the spectral weighting that minimizes the dispersion between predicted and actual demands. In the vocabulary of performance-based earthquake engineering, the resulting measure exhibits high efficiency, meaning less scatter in the demand prediction, and sufficiency, meaning it captures the demand well regardless of the magnitude, distance, or spectral character of the earthquake that produced it. These two properties are the gold standards for any intensity measure intended for probabilistic seismic demand analysis, because they determine how many ground motion records an analyst must run to obtain a trustworthy estimate of risk.

The efficiency gains reported in the study are striking. When the proposed intensity measure was tested against a 489-meter super-tall building archetype, a structure representative of the tallest class of towers now common in Chinese metropolitan skylines, only seven ground motion records were needed to achieve a stable average response. Conventional scaling based on peak ground acceleration required twenty-three records to reach the same stability, a 70 percent reduction in the number of analyses. For a nonlinear model of a super-tall building, where a single time-history simulation can consume hours of computation, that reduction translates directly into days or weeks saved on every vulnerability study, and into the practical feasibility of running the many analyses required by modern performance-based design codes.

To demonstrate the practical payoff, the authors applied their intensity measure to a seismic vulnerability assessment of rocking freestanding indoor contents, the unanchored furniture, cabinets, and equipment that populate real floors and are notorious for sliding and overturning during earthquakes. Freestanding bodies rock and topple in response to both the acceleration and the velocity of the floor beneath them, which is precisely why a measure capturing both quantities outperforms acceleration-only or drift-based alternatives. The study builds on a long line of research into rigid-body overturning, from classic criteria for earthquake-induced rocking through recent experimental work on electrical cabinets and hospital equipment, and shows how a better ground-motion descriptor flows through the entire fragility pipeline to yield more reliable estimates of content damage.

The implications reach beyond the laboratory into codes and practice. Performance-based seismic design standards, including the FEMA P-58 framework in the United States and a family of Chinese specifications for tall buildings, increasingly demand quantified predictions of nonstructural damage and downtime. An intensity measure that targets the demands actually experienced by nonstructural components gives engineers a sharper instrument for those calculations, and its computational efficiency makes routine fragility assessment of super-tall towers far more tractable. The work also aligns with a broader trend in earthquake engineering: rather than treating the building as an isolated skeleton, designers are learning to evaluate the full functional system, structure, contents, and occupants together, because a building that stands but cannot operate has failed the recovery test.

The research, funded by the Natural Science Foundation of China and the Institute of Engineering Mechanics of the China Earthquake Administration, arrives at a moment when super-tall construction and seismic risk are converging in densely populated regions. As cities push towers ever higher, the floors near the top will experience acceleration demands that ground-level instruments cannot foreshadow, and the contents of those floors, from data centers to emergency supplies, will determine how quickly a shaken city returns to normal. By recalibrating the language engineers use to describe earthquake intensity around the quantities that govern nonstructural damage, Cao and Qu offer a deceptively simple but consequential upgrade: fewer simulations, tighter predictions, and a clearer path to buildings that not only survive great earthquakes but keep working afterward.

Subject of Research: A spectral-acceleration intensity measure for predicting peak floor acceleration and velocity demands in super-tall buildings

Article Title: A spectral-acceleration intensity measure for peak floor acceleration and velocity demands in super-tall buildings

Article References: Cao, Y., & Qu, Z. (2026). A spectral-acceleration intensity measure for peak floor acceleration and velocity demands in super-tall buildings. Bulletin of Earthquake Engineering. https://doi.org/10.1007/s10518-026-02695-3

Image Credits: AI Generated

DOI: 10.1007/s10518-026-02695-3

Keywords: earthquake engineering, intensity measure, super-tall buildings, peak floor acceleration, peak floor velocity, nonstructural elements, spectral acceleration, time-history analysis, seismic vulnerability, performance-based design, functional recovery, fragility assessment

The appeal of friction dampers lies in their physics. Conventional steel frames absorb earthquake energy by yielding, which means the structure itself accumulates damage with every loading cycle, a phenomenon engineers call cyclic deterioration or pinching. An RFD behaves differently. Governed by Coulomb’s law of dry friction, it produces stable, rectangular hysteretic loops whose enclosed area, and therefore the energy dissipated per cycle, remains constant no matter how many times the ground shakes. The device consists of a central rotating plate sandwiched between side plates with friction discs, all clamped by a pre-tensioned bolt. When inter-story drift forces the plates to rotate against one another, friction resists the motion with a constant moment, protecting the primary frame. Experimental work dating back to Mualla and Belev in 2002 showed reductions of up to 89 percent in lateral displacement and 41 percent in base shear under optimized configurations.

Simulating such devices accurately, however, is notoriously difficult. The abrupt transition between sticking and slipping is mathematically discontinuous, and the two common workarounds both introduce error. Regularization smooths the discontinuity with an artificial curve, adding fake compliance exactly where the response should be sharp, while equivalent-damping approaches replace the device with a viscous dashpot that cannot reproduce the rate-independent, rectangular hysteretic loop. Both require case-specific calibration, which would be hopeless at the scale of hundreds of thousands of simulations. The authors instead built on the Non-Smooth Dynamics method, reformulating Coulomb friction as a Linear Complementarity Problem solved at each timestep with a Projected Successive Over-Relaxation algorithm. This formulation reproduces the exact friction law with no tuning parameters, and because each step typically converges in fewer than five iterations, sweeping thousands of structural configurations becomes computationally feasible. Crucially, it also means the training labels fed to the machine learning models contain pure physics rather than artifacts of a smoothing choice.

The dataset behind the study is staggering in scope. The team generated 1,000 shear-type frames ranging from 3 to 20 stories, with randomized Latin Hypercube sampling covering fundamental periods from 0.3 to 3.0 seconds, floor masses from 150 to 350 tonnes, and four different mass and stiffness distributions. Each frame was fitted with roughly 50 different damper configurations, varying the number of devices, their vertical placement, and friction moments between 25 and 500 kilonewton-meters. Eight real seismic records, including El Centro 1940, Kobe 1995, and Northridge 1994, were applied at their original historical intensities. In total, 433,984 nonlinear simulations were run, each quantified with a composite index called CR2 that combines peak roof displacement, peak roof acceleration, and peak base shear, each normalized by the uncontrolled response, so that lower values mean better performance.

Six regression models were then trained to predict seismic performance from the design variables, and the comparison itself delivered a scientific finding. Linear models achieved a cross-validated coefficient of determination of only 0.64 for the typical scenario, while XGBoost reached 0.907 and Random Forest 0.904, cutting prediction error nearly in half. The gap proves that the relationship between damper design and seismic response is fundamentally nonlinear, with threshold effects and interactions that simple equations cannot capture. The nonlinearity deepens under severe ground motion: for the conservative 90th-percentile scenario, linear models managed only 0.31, while Random Forest reached 0.786. To prevent data leakage, the team used structure-grouped cross-validation, ensuring models were always evaluated on buildings they had never seen, and verified accuracy on 200 entirely held-out structures with calibrated 90 percent prediction intervals.

The most striking practical insight concerns coverage density, the fraction of floors equipped with dampers. Feature importance analysis identified this single variable as the dominant predictor of performance, and the response surface revealed clear diminishing returns. By formulating the threshold as an explicit optimization over the trained surrogate, the researchers found that the median coverage at which 80 percent of the maximum achievable gain is reached sits at about 0.83, or 0.86 when read from the empirical envelope across the whole population. Resolved by building height, the picture inverts common intuition: low-rise frames of 3 to 5 stories require full coverage, mid-rise frames of 6 to 12 stories need at least 86 percent, and high-rise frames of 13 to 20 stories need only about 73 percent, with none requiring full instrumentation. The mechanical explanation is elegant, since tall buildings concentrate drift demand in a subset of stories that can be covered selectively, whereas the flat drift profile of a low-rise frame leaves no inexpensive stories to omit.

The analysis also uncovered a counterintuitive failure mode. A small number of oversized dampers, high friction moments paired with low coverage, can leave a frame worse off than no dampers at all. In this lock-up regime, the devices never reach their slip threshold under the imposed demand, act as rigid links, and locally stiffen the structure, amplifying floor accelerations and base shear. These configurations account for just 1.4 percent of the dataset but nearly 10 percent of the total prediction error, and more than 80 percent of them occur in frames of five stories or fewer. The practical lesson is a lower bound on coverage for any given device capacity: coverage cannot be traded freely against friction moment. The study identifies an optimal friction moment window of roughly 300 to 460 kilonewton-meters for mid- and high-rise frames, with a lower optimum near 250 kilonewton-meters for short buildings.

Robustness across seismic scenarios proved to be another strength of the framework. When the researchers compared the optimal damper count recommended under typical conditions with that recommended under the conservative 90th-percentile scenario, 67.9 percent of the 1,000 structures received identical recommendations, and in the remaining cases the median discrepancy was just one damper. The conservative scenario never suggested fewer devices, only occasionally one or two more, meaning designers can optimize for typical demand and then verify against severe conditions with minimal adjustment. This multi-record approach, summarizing performance with 10th-percentile, median, and 90th-percentile statistics consistent with the FEMA P-58 framework, addresses a long-standing gap in earlier studies that optimized dampers against a single ground motion.

The authors are candid about the limits of their results. The structural models are elastic shear frames, and a substantial fraction of the simulated demand levels actually lie at or beyond the yield limit of real steel frames, so the coverage recommendation should be read as an upper bound for quasi-elastic serviceability design rather than a validated rule for ductile response. The design space is a factorial sweep rather than a representative building stock, and the eight-record suite carries sampling uncertainty of about 0.10 in CR2 units, which exceeds the surrogate’s own prediction error of 0.070 by roughly half. That last point may be the study’s most consequential methodological message: enlarging the ground motion database, not refining the machine learning model, is what would most sharpen future design guidance. Future extensions include soil-structure interaction, three-dimensional frames with torsional coupling, nonlinear material behavior, floor-specific friction moments, and cost-based optimization.

Even with those caveats, the work marks a genuine shift in how earthquake protection devices might be designed. Instead of running expensive simulations case by case, engineers can now query a trained surrogate that exposes the entire response surface across damper placement, capacity, and building geometry in milliseconds, complete with calibrated uncertainty bounds. The finding that taller buildings need proportionally fewer instrumented floors, that a coverage threshold near 86 percent captures most of the achievable benefit for mid-rise construction, and that a handful of oversized devices can actively harm a structure, together give practitioners an evidence-based starting point that previously required months of computation. As machine learning surrogates spread through earthquake engineering, this study offers both a template, exact physics in the labels, grouped validation, honest uncertainty, and a reminder that the biggest gains sometimes come from questioning design intuition.

Subject of Research: Machine learning surrogate modeling of nonlinear seismic response of steel moment frames with rotational friction dampers

Article Title: A machine learning-based surrogate model for analyzing the nonlinear seismic response of steel moment frames with rotational friction dampers

Article References: Ponce-López, E., & Yáñez-Godoy, H. (2026). A machine learning-based surrogate model for analyzing the nonlinear seismic response of steel moment frames with rotational friction dampers. Machine Learning with Applications, 26, Article 101030. https://doi.org/10.1016/j.mlwa.2026.101030

Image Credits: AI Generated

DOI: 10.1016/j.mlwa.2026.101030

Keywords: rotational friction dampers, machine learning, seismic response, steel moment frames, XGBoost, non-smooth dynamics, earthquake engineering, surrogate model, coverage density, structural optimization, energy dissipation, time-history analysis

The team focused on a configuration familiar to anyone who has walked through Mediterranean or Asian cities: the pilotis building, in which the ground storey is left open, its columns rising without the confining embrace of masonry infill walls. Buildings with this open ground floor have collapsed repeatedly in past earthquakes, including the Yushu earthquake in China, because the abrupt stiffness discontinuity between the open base and the stiff upper storeys concentrates deformation there. Yet pilotis buildings continue to be constructed because open ground floors are architecturally desirable, leaving engineers to assess the risk with tools of uncertain reliability.

To probe that risk systematically, the researchers built finite element models of reinforced concrete frames at three heights—three, six and twelve storeys—in three configurations each: bare frames, fully infilled frames, and frames with ground-floor pilotis. The plan geometry was held constant across every model, with four bays of 4.0 metres and five bays of 3.0 metres, and uniform 3.0-metre storey heights, so that only the infill arrangement and the building height varied. Reinforcement detailing followed Eurocode 2 and Eurocode 8 practice for medium-ductility frames, with longitudinal reinforcement ratios of roughly 1.7 percent in beams and 2.7 percent in columns, and confined concrete cores represented through the Mander constitutive model.

The modelling strategy balanced fidelity against computational practicality. Reinforced concrete beams and columns were represented with fibre-based force-formulated frame elements, which subdivide each cross-section into fibres whose nonlinear uniaxial stress-strain behaviour is integrated along the member, capturing the spread of inelasticity in both space and time. Masonry infills were represented through a macro-model based on Crisafulli’s double-strut formulation, in which each panel is replaced by six components: two parallel axial struts in each diagonal direction and shear springs that capture bed-joint sliding. This equivalent-strut approach accounts for four failure mechanisms—diagonal tension, sliding shear, corner compression and compression at the panel centre—without the prohibitive cost of detailed micro-modelling, making it well suited to large parametric studies.

Nonlinear time-history analyses were performed in SeismoStruct using the HHT-alpha integration algorithm with a fixed timestep of 0.05 seconds, while eigenvalue analyses employing the Lanczos algorithm established the modal properties of each frame. Three recorded earthquake motions—Landers 1992, Northridge 1994 and Kobe 1995—were deliberately selected because their response spectra peaked near the natural periods of the structures, creating challenging resonant conditions. Critically, the records were applied without scaling or spectrum matching, so the results reflect genuine recorded events, though the authors caution that the limited number of motions means the findings should be read as indicative of behavioural mechanisms rather than statistically robust design predictions.

The modal analyses confirmed that masonry infill fundamentally reshapes a building’s dynamics. Infilled frames were consistently stiffer than bare frames, with shorter fundamental periods, while pilotis frames fell between the two, slightly softer than fully infilled counterparts because of the missing ground-storey walls. As height increased, fundamental periods lengthened and the first three modes separated further apart, consistent with classical structural dynamics. Notably, the difference between fully infilled and pilotis models was small, suggesting that total infill quantity, rather than its vertical distribution, governs overall stiffness—a first hint that the notorious soft storey might not be the dominant factor engineers often assume.

The nonlinear time-history results deepened that surprise. Acceleration profiles along the height of the buildings were strikingly non-monotonic: after an initial rise, accelerations decreased and rose again in the six- and twelve-storey frames, a pattern the authors attribute to higher-mode participation shifting as stiffness degrades non-uniformly with accumulating damage. Masonry infill reduced acceleration amplification by a factor of roughly 1.5 across the model families, and the peak amplification period scaled almost linearly with building height—doubling as storey count doubled—indicating that the fundamental mode dominates response even deep into the nonlinear regime. Perhaps most counterintuitively, the twelve-storey bare frame sometimes showed lower maximum displacements than its infilled counterparts, hinting that added mass and nonlinear interaction can offset the benefits of added stiffness.

The height-dependent transition in pilotis behaviour emerged most clearly in the drift and force demands. In the three- and six-storey frames, the expected concentration of inter-storey drift at the open ground floor failed to materialise; the pilotis models behaved much like their fully infilled twins. Only in the twelve-storey building did deformation concentrate at the base, and even there, the fully infilled frame experienced comparable ground-floor drift, suggesting that infill mass rather than soft-storey flexibility alone drove the effect. Column shear forces told a parallel story: in shorter buildings, pilotis columns carried less shear than bare-frame columns, but in the twelve-storey pilotis frame, ground-floor column shear exceeded that of the fully infilled counterpart by about 50 kilonewtons, evidence that the detrimental influence of the open storey strengthens as buildings grow taller.

Comparisons across the three earthquake records revealed why the picture is so record-dependent. Under Landers, bare frames showed far higher peak spectral accelerations than pilotis frames, and the pilotis configuration occasionally proved beneficial, reducing amplification relative to the bare frame. Under Kobe and Northridge, differences between configurations shrank, and in one case the infilled frame’s peak acceleration nearly matched the bare frame’s, shifted to a longer period of about 1.0 second. The researchers traced this variability to nonlinear period elongation: as damage softens a structure, its elongated periods migrate across the frequency landscape of the input motion, sometimes sliding toward spectral peaks and sometimes away from them. Whether pilotis proves harmful therefore depends on where the resonant peak of the shaking happens to fall relative to the structure’s evolving dynamics.

The study’s most consequential finding concerns design codes. Eurocode 8 permits structures with significant infill asymmetry to be analysed as bare frames, on the assumption that this is conservative. The Bath team’s results show that such simplification can be conservative for some response quantities and configurations, but not universally so: in several cases, ignoring the infill underestimated demands or overstated benefits in ways that depended on the interplay of period elongation, spectral content and building height. The authors acknowledge limitations—a small set of unscaled records, macro-modelled infills that cannot capture local cracking, rigid diaphragm assumptions, neglected torsion and soil-structure interaction—but argue that the message stands. Assessing irregular reinforced concrete buildings accurately demands nonlinear dynamic analysis that honours infill-frame interaction, because static or linear thinking simply cannot predict where a soft storey will turn from a nuisance into a catastrophe.

Subject of Research: Nonlinear seismic response of masonry-infilled reinforced concrete frame buildings with pilotis configurations across different building heights

Article Title: Height-dependent nonlinear seismic response of masonry-infilled RC frame buildings with pilotis configurations

Article References: Height-dependent nonlinear seismic response of masonry-infilled RC frame buildings with pilotis configurations. (n.d.). https://doi.org/10.1007/s10518-026-02687-3

Image Credits: AI Generated

DOI: 10.1007/s10518-026-02687-3

Keywords: earthquake engineering, reinforced concrete, masonry infill, pilotis, soft storey, nonlinear analysis, time-history analysis, inter-storey drift, period elongation, Eurocode 8, finite element analysis, building height