Researchers have developed an innovative approach to predict the working status of high-formwork support systems (HFSS) by combining finite element model (FEM) simulations with deep learning and large language models (LLMs). Published in *Smart Construction*, this study addresses the challenges of structural health monitoring (SHM) by leveraging a genetic algorithm (GA)-optimized FEM to generate training data for a convolutional neural network (CNN) classifier. The framework also integrates a Retrieval-Augmented Generation (RAG) model with a knowledge graph (KG) to automatically generate reasonable SHM reports for HFSS, demonstrating superior performance over common methods.

image: The proposed framework integrates FEM simulations, CNN classification, and an RAG model for automated SHM reporting, achieving high accuracy in structural status prediction and report generation for HFSS.

Credit: Linlin Zhao/Beijing University of Technology, Jasper Mbachu/Bond University, Siyu Liu/Beijing University of Technology, Rongtian Zhang/Beijing University of Technology

High-formwork support systems (HFSS) are critical in construction but prone to collapses due to inadequate monitoring. Common methods rely on expensive and complex experiments, limiting their practicality. This study proposes a data-driven solution using FEM simulations and deep learning to predict HFSS working statuses—normal, local instability, and fully unstable—with high accuracy.

The research begins with developing and optimizing an FEM of an HFSS using a genetic algorithm (GA) to minimize discrepancies between simulated and experimental data. The optimized FEM generates datasets for training a CNN classifier, which achieves an impressive accuracy in predicting structural statuses. Experimental validation on a full-scale HFSS confirms the CNN's superiority over support vector machines (SVM), with the CNN outperforming SVM in classification tasks.

To streamline SHM reporting, the study introduces a Retrieval-Augmented Generation (RAG) model, combining GPT-4 with a domain-specific knowledge graph (KG). The RAG model generates detailed SHM reports, evaluated using metrics like BLEU, ROUGE, and cosine similarity, demonstrating its effectiveness in producing accurate and contextually relevant reports compared to standalone GPT-4.

Key contributions of the study include:

1. A GA-optimized FEM for generating reliable training data under varied HFSS conditions.

2. A CNN classifier with high accuracy in predicting structural statuses.

3. An RAG model that automates SHM report generation, reducing reliance on subjective expertise.

The framework's potential applications extend to complex structures, with future work focusing on integrating multimodal models and edge computing for broader deployment.

This paper, "Working Status Prediction for a High-Formwork Support System Using Finite Element Model-Informed Deep Learning and GPT-Aided Method," was published in Smart Construction.

Zhao L, Mbachu J, Liu S, Zhang R. Working Status Prediction for a High-Formwork Support System Using Finite Element Model-Informed Deep Learning and GPT-Aided Method. Smart Constr. 2025(2):0015, https://doi.org/10.55092/sc20250015.

Source from [https://www.eurekalert.org/news-releases/1090156].

Researchers have developed a revolutionary non-invasive method combining 3D laser scanning technology and sensor data time-series analysis for identifying defective water treatment filters, achieving significant reductions in inspection time and labor costs while eliminating operational disruptions. Published in Smart Construction, this breakthrough has the potential to transform water treatment facility maintenance, ensuring safer and more efficient water processing by detecting subsurface structural defects through surface geometric changes and operational anomalies.

image: The multi-dimensional data interpretation framework integrates upside-down 3D laser scanning, SCADA sensor data analysis, and CFD simulation validation, tested on actual water treatment filters during backwash operations. The framework features non-invasive detection through geometric feature analysis and time-series pattern recognition, achieving reliable identification of subsurface structural defects. This system could be applied in smart water treatment facilities for automated filter health monitoring, predictive maintenance, and optimized backwash operations.

Credit: Pengkun Liu/Carnegie Mellon University, Jinghua Xiao/Circular Water Solution LLC, Pingbo Tang/Carnegie Mellon University,

Water treatment filters serve as the final physical barrier for removing suspended solids and pathogens, making their structural integrity crucial for effective water treatment and public health protection. However, traditional filter inspection methods require manual disruption of filter media, are time-consuming and labor-intensive, involving at least two personnel for several hours, and risk overlooking defects due to limited sampling areas. Addressing this challenge, Dr. Pengkun Liu and Professor Pingbo Tang from Carnegie Mellon University, in collaboration with researchers from Circular Water Solution LLC, has developed a state-of-the-art multi-dimensional data interpretation framework that revolutionizes water treatment filter monitoring without operational interruptions.

"This framework design marks a critical advancement in water treatment facility maintenance," explains Professor Pingbo Tang. "Our central hypothesis is that when subsurface structural irregularities disrupt normal backwash operations and flow patterns, they produce both visible surface deformations and distinctive anomalies in sensor data, enabling non-invasive defect detection."

The newly developed framework utilizes upside-down installed 3D laser scanners to capture high-precision geometric changes on filter media surfaces before and after backwash processes. Lead researcher Pengkun Liu emphasizes, "The integration of geometric feature analysis—including roughness, curvature, omnivariance, and planarity—with time-series sensor data significantly enhances detection accuracy while eliminating the need for invasive media disruption."

A major challenge in filter inspection has been the inability to detect subsurface defects like uneven gravel support beds, mud ball formation, or underdrain blockages without disrupting operations. The team addressed this by developing a comprehensive four-module architecture: data acquisition through 3D laser scanning and SCADA sensor monitoring, geometric analysis using advanced clustering methods, time-series analysis of operational parameters, and fusion diagnosis validated by computational fluid dynamics simulations.

The framework, tested extensively at the Shenango Water Treatment Plant in Pennsylvania across six filter units, underwent rigorous validation over multiple operational cycles. Led by Jinghua Xiao from Circular Water Solution, the team demonstrated that the system effectively identifies abnormal filters through surface elevation irregularities, geometric feature variations, and operational parameter anomalies. Their measurements confirmed that defective filters exhibit distinctive patterns: surface elevation differences, higher geometric feature values, longer backwash durations, elevated turbidity levels, and reduced water production rates.

In practical applications, the framework successfully identified Filter 2 as defective, showing consistent surface bulging patterns across multiple scans, abnormal geometric characteristics, and suboptimal operational performance. These findings were validated through CFD simulations, which confirmed that subsurface defects like mud balls or dead zones disrupt uniform flow distribution from bottom drainage pipes, causing uneven surface deformations. The repeatability tests demonstrated high consistency in defect detection, proving the system's reliability in real-world water treatment settings.

"This framework has the potential to significantly impact the development of smart water treatment systems and infrastructure monitoring applications," says co-researcher Jinghua Xiao. "Its non-invasive nature and comprehensive multi-modal approach could lead to safer, more cost-effective, and more reliable maintenance practices for water treatment facilities worldwide."

In addition to water treatment applications, the techniques developed in this framework could inspire innovations in other infrastructure monitoring settings requiring precise subsurface defect detection, such as advanced pipeline inspection, industrial filtration systems, and civil infrastructure health monitoring.

The multi-dimensional data interpretation framework achieves exceptional performance through its innovative combination of 3D geometric analysis and sensor data fusion. It operates effectively with millimeter-level precision in surface change detection and real-time operational parameter monitoring, offering both geometric and time-series anomaly detection capabilities. "This approach establishes quantitative relationships between surface irregularities and subsurface defects, setting a new standard for non-invasive infrastructure inspection," notes Professor Pingbo Tang.

While the team acknowledges the need for expanding the dataset to more water treatment facilities and developing real-time monitoring systems, this study represents a critical step toward more efficient, safer, and more reliable water treatment operations. Future research directions include integrating 3D LiDAR sensors directly into filter structures for continuous monitoring and developing adaptive backwash control systems based on real-time surface geometry analysis.

This paper ”Multi-dimensional data interpretation for defective filter identification” was published in Smart Construction.

Liu P, Xiao J, Tang P. Multi-dimensional data interpretation for defective filter identification. Smart Constr. 2025(2):0014, https://doi.org/10.55092/sc20250014.

Source from [https://www.eurekalert.org/news-releases/1089689].

A team of Italian researchers has uncovered compelling evidence of anomalous radioactive decay in cobalt-57 (Co-57) under ultrasonic stimulation, offering strong experimental support for the Deformed Space-Time (DST) theory. The findings, published by Stefano Bellucci (INFN-Frascati) and Fabio Cardone (ISMN-CNR), suggest that brief ultrasonic exposure can trigger a departure from conventional exponential decay laws, mediated by energy-dependent space-time distortions that violate local Lorentz invariance (LLI).

image: Ultrasound triggers anomalous Co-57 decay via Deformed Space-Time effects. Just nanoseconds of sonication induce non-classical nuclear transformations, offering new evidence for Lorentz Invariance violation and metric-dependent nuclear reactions.

Credit: Stefano Bellucci/INFN-LNF, Italy

In a groundbreaking study exploring the frontier between nuclear physics and space-time geometry, Stefano Bellucci and Fabio Cardone report anomalous radioactive decay behavior in the isotope cobalt-57 (Co-57) when exposed to ultrasound at 2.25 MHz. This unexpected deviation from the standard exponential decay curve—specifically in the 14.4 keV Fe-57 emission line—is interpreted through the lens of Deformed Space-Time (DST) theory, which posits that under certain energy conditions, nuclear processes occur in a locally non-Minkowskian metric.

"Only a few nanoseconds of ultrasonic activation—less than one percent of a single wave cycle—are enough to trigger measurable effects consistent with a deformed space-time," explains Bellucci. These effects include enhanced transformation of Co-57 nuclei without traditional radioactive decay emissions, suggesting an alternative, non-weak-interaction-driven nuclear transformation.

At the heart of the study lies the hypothesis that Ridolfi cavities—microcavities formed under ultrasonic stress—act as "nuclear micro-reactors," enabling strong-interaction pathways not accessible under standard conditions. This dual-path decay mechanism, combining traditional weak decay with DST-induced transformation, offers a radical reinterpretation of nuclear stability and decay in dynamic fields.

Notably, the research draws parallels to previous DST-based experiments conducted by the same team involving isotopes like thorium-228 (Th-228) and nickel-63 (Ni-63), where cavitation led to significant reductions in radioactivity. The new findings from the Hagelstein-type experiment with Co-57 serve as an independent confirmation, revealing persistent metric deformation effects (latency) and energy coupling between fields—phenomena consistent with the so-called Mignani mimicry.

“This work challenges the long-held assumption that radioactive decay is immutable under classical field exposure,” says Cardone. “If space-time itself can deform in response to external stress, our understanding of fundamental interactions—and their constraints—must be revisited.”

The authors emphasize that the observed decay anomalies imply a deeper violation of Local Lorentz Invariance, hinting at broader implications for causality and even the constancy of the speed of light. Future experiments are proposed to determine whether the observed transformations result from an increase in decay rate—or a true metamorphosis of nuclear identity, absent radiation emission.

As a forward-looking application, the authors propose real-time radioactivity monitoring during sonication. “If sonication leads to more radiation, it implies faster decay,” notes Bellucci. “If not, we are looking at a fundamentally different reaction altogether.”

This study opens new avenues in nuclear science, cosmology, and the physics of space-time—where matter, energy, and geometry may interact more dynamically than previously imagined.

This paper ”On anomalous radioactive decay according to the energy metrics formalism in the Deformed Space-Time (DST) theory” was published on 30 June 2025 in ELSP Asymmetry.

Stefano Bellucci and Fabio Cardone, On anomalous radioactive decay according to the energy metrics formalism in the Deformed Space-Time (DST) theory. Asymmetry 2025(1):0005, https://doi.org/10.55092/asymmetry20250005.

Source from [https://www.eurekalert.org/news-releases/1090173].

Researchers from Nanyang Technological University have developed a novel framework that integrates worker self-reportsaZ with objective expert evaluations to address the challenge of validating human reliability models. They also proposed a new hybrid inference model combining Self-Organizing Maps (SOM) and Bayesian Networks (BN) to more accurately predict worker performance and identify potential failures. Published in Smart Construction, this research transcends the limitations of traditional expert-driven methods, offering a validated, data-driven approach to enhance workplace safety and operational efficiency in the high-risk construction industry.

image: The proposed framework for human performance reliability evaluation consists of three phases. First, data is obtained via subjective worker self-assessments and objective expert evaluations. Second, the data is preprocessed using rank standardization and fuzzy set theory to derive an adjusted reliability score for Common Performance Conditions (CPCs) . Finally, in the prediction phase, a SOM-BN inference engine uses this score to predict the workers' control modes.

Credit: Yamo Cao/Nanyang Technological University, Zeren Jin/Nanyang Technological University, Yuguang Fu/Nanyang Technological University

The construction industry is recognized as one of the most hazardous sectors, with human error being a primary cause of most workplace accidents. To assess and mitigate these risks, Human Reliability Analysis (HRA) methods are employed, with second-generation techniques like the Cognitive Reliability and Error Analysis Method (CREAM) being prominent for their focus on cognitive and contextual factors. However, traditional CREAM models suffer from significant limitations: they rely heavily on subjective expert judgment, lack independent ground-truth for validation, and often oversimplify the complex, non-linear interactions between performance-shaping factors .

To overcome these challenges, Yamo Cao, Zeren Jin, and Assistant Professor Yuguang Fu from Nanyang Technological University proposed a comprehensive framework for data collection and a new hybrid model for prediction. The proposed framework consists of two main components: A Unified Data Collection and Processing Framework and A Hybrid SOM-BN Inference Engine.

The Unified Data Collection and Processing Framework brings together subjective worker self-assessments and objective expert evaluations. To process subjective data, the researchers introduced the Contextual Human Reliability Score (CHRS), which applies fuzzy logic and rank-standardization to worker self-reports, effectively reducing bias and creating a robust input metric. The objective data, consisting of expert observations of worker performance, serves as the independent ground truth for model validation.

The Hybrid SOM-BN Inference Engine uses a two-stage process. First, a Self-Organizing Map (SOM), an unsupervised clustering technique, analyzes the CHRS data to identify complex and non-linear patterns among the various performance conditions. Second, these empirically derived clusters are fed into a Bayesian Network (BN), which performs transparent probabilistic inference to predict the worker's performance control mode (e.g., effective or ineffective).

To validate the framework, the researchers conducted a case study on a construction site, collecting data from 35 workers. The results were compelling: the proposed SOM-BN model demonstrated high accuracy (88.57%) and specificity in its predictions. In stark contrast, traditional CREAM models, when applied to the same dataset, were unable to effectively distinguish between safe and unsafe performance, highlighting their limitations in this real-world context. Furthermore, the use of explainability tools like SHAP provided clear insights into individual worker performance, making the model's predictions actionable.

In the future, the research team plans to expand the dataset to include a more diverse worker population and apply the framework to other industries, further demonstrating its broad applicability in improving human reliability and safety.

This paper ”A new hybrid inference model for human performance reliability prediction: a case study of construction workers” was published in Smart Construction.

Cao Y, Jin Z, Fu Y. A new hybrid inference model for human performance reliability prediction: a case study of construction workers. Smart Constr. 2025(3):0016, https://doi.org/10.55092/sc20250016.

Source from [https://www.eurekalert.org/news-releases/1090734].