Intelligent Infrastructure Crack Detection Using MSEDBO-Optimized Deep Learning

 

Infrastructure surface crack detection is a vital task in structural health monitoring, directly influencing the safety, durability, and serviceability of civil engineering assets. Although deep learning methods have achieved notable success in automated crack detection, their performance is often constrained by inefficient hyperparameter tuning, susceptibility to local optima, and suboptimal feature extraction. This study addresses these limitations by proposing an intelligent optimization-driven crack detection framework.

Limitations of Conventional Deep Learning-Based Crack Detection

Traditional deep learning models rely heavily on manual or heuristic-based hyperparameter selection, which can lead to unstable training outcomes and reduced generalization performance. Moreover, commonly used optimization techniques may become trapped in local optima, resulting in inaccurate crack localization and increased false positive rates, particularly when dealing with complex backgrounds and diverse infrastructure materials.

Multi-Strategy Enhanced Dung Beetle Optimizer (MSEDBO)

The proposed framework integrates a Multi-Strategy Enhanced Dung Beetle Optimizer (MSEDBO) to systematically optimize critical parameters within the crack detection pipeline. MSEDBO incorporates Latin Hypercube Sampling with elite population initialization, an improved sigmoid-based nonlinear control factor, sine–cosine algorithm integration, and multi-population mutation strategies. These enhancements collectively strengthen global exploration and local exploitation capabilities.

Integration with Deep Learning Models

By embedding MSEDBO into deep learning-based crack detection models, the framework enables adaptive optimization of network parameters and feature extraction processes. This synergy improves convergence behavior, enhances robustness against local optima, and ensures efficient learning across varying crack patterns and surface conditions in civil infrastructure.

Experimental Validation and Benchmark Datasets

The proposed approach was validated using multiple benchmark datasets, including CrackTree200, CFD, GAPs, and SDNET2018, covering a wide range of materials such as concrete pavements, asphalt roads, and bridge surfaces. Comparative experiments demonstrate that the MSEDBO-optimized framework consistently outperforms conventional optimization algorithms and baseline deep learning models.

Performance Gains and Practical Implications

Results show significant improvements, including an 8.7% increase in detection accuracy, a 12.3% improvement in precision, and a 15.6% reduction in false positive rates. The framework maintains computational efficiency while effectively avoiding local optima, making it well suited for real-world deployment. This research advances intelligent infrastructure monitoring by providing a robust optimization strategy to enhance the reliability and accuracy of automated crack detection systems.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#AIinCivilEngineering

#AutomatedInspection

#ConcreteCracks

#RoadSurfaceMonitoring

#BridgeInspection

#MachineLearning

#EngineeringOptimization

#DigitalInfrastructure

#SustainableInfrastructure

#CivilEngineeringResearch

Thermal Feasibility of Ground Source Heat Pump Systems for Civil Defense Facilities

Persistently high cooling loads and strict safety requirements make thermal regulation in civil defense facilities particularly challenging. Conventional cooling systems often struggle to meet long-term efficiency, reliability, and concealment demands. This study evaluates the thermal feasibility of a ground source heat pump (GSHP) system with high concealment for application in the Zhushan civil defense facility in Nanjing, China, aiming to assess its suitability for managing continuous and intensive cooling demands.

Numerical Modeling of Ground Heat Exchange

A three-dimensional numerical model was developed to simulate coupled flow and heat transfer processes within the ground heat exchanger (GHE). The model incorporates unsaturated porous soil conditions, time-varying fluid flow rates, and environmental disturbances, enabling realistic representation of heat transfer between the circulating fluid, grout, and surrounding soil. This modeling framework provides a robust basis for evaluating subsurface thermal behavior.

Coupling with Heat Pump Unit Performance

The ground heat exchanger model was coupled with a heat pump unit model through facility cooling load, fluid outlet temperature, and coefficient of performance (COP). This integrated approach allowed for a comprehensive assessment of overall system efficiency, capturing the interaction between subsurface heat exchange processes and surface-level heat pump performance under realistic operating conditions.

Cooling Load Characteristics and Energy Consumption

Results show that the civil defense facility experiences consistently high cooling loads due to continuous heat gains from equipment operation and occupant metabolic activity. System energy consumption is dominated by the compressor, accounting for approximately 80% of total power usage. A clear seasonal pattern was observed, with energy consumption increasing steadily and peaking in August, corresponding to the highest cooling demand.

Thermal Performance and System Stability

The average fluid outlet temperature of the GHE (Tf,o), heat pump COP (COPhp), and system COP (COPsys) were 24.51 °C, 5.35, and 4.42, respectively. Most values clustered within narrow ranges, indicating stable operation. Minimum values of Tf,o, COPhp, and COPsys were 18.78 °C, 5.17, and 4.22, while maximum values reached 26.43 °C, 5.78, and 4.71, demonstrating both robustness and efficiency across operating conditions.

Feasibility of GSHP for Civil Defense Applications

The tight clustering of thermal and performance indicators, combined with high COP values and controlled outlet temperatures, confirms that the GSHP system operates under stable and efficient conditions. These findings demonstrate the thermal feasibility of GSHP systems as a reliable and energy-efficient solution for managing persistently high cooling loads in civil defense facilities, supporting their broader adoption in similar high-demand, high-security infrastructures.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#InfrastructureEnergy

#HVACEngineering

#LowCarbonBuildings

#UndergroundStructures

#EngineeringSimulation

#COPAnalysis

#EnergySystems

#ClimateResilientInfrastructure

#CivilEngineeringResearch

#ThermalFeasibility

#SustainableInfrastructure

 

Sustainable Rockfall Protection Using Recycled Tyres for Enhanced Impact Resistance

 

Rockfall hazards present a serious risk to infrastructure in mountainous and hilly regions, demanding mitigation systems that are both mechanically robust and economically viable. Reinforced concrete rockfall barriers are commonly used, yet their performance under high-energy impacts depends heavily on energy-absorbing components. This study investigates innovative and sustainable approaches to enhance rockfall barrier performance, with a particular focus on replacing conventional materials with recycled alternatives.

Numerical Modelling of Rockfall Impact

A detailed finite element model of a double-anchored reinforced concrete rockfall barrier was developed using Abaqus/Explicit. The barrier was subjected to a high-velocity impact of 25 m/s from a 1 m diameter spherical rock, representing severe rockfall conditions. This numerical framework enabled accurate simulation of impact dynamics, deformation behavior, and energy dissipation mechanisms within the barrier system.

Performance of Conventional EPS Foam

The study first assessed EPS-11 foam, a traditional energy-absorbing material, for enhancing the impact resistance of the barrier. While EPS foam demonstrated some capacity to reduce structural response, its performance was limited in terms of deformation control and long-term sustainability. These limitations motivated the exploration of alternative, more resilient materials.

Recycled Tyres as Hyperelastic Energy Absorbers

Used car tyres were investigated as sustainable hyperelastic energy-absorbing layers, modeled using the Arruda–Boyce constitutive model. Three configurations were examined: an unprotected barrier, a barrier with a single tyre layer, and a barrier with double staggered tyre layers. The hyperelastic nature of tyres enabled effective redistribution and absorption of impact energy through large deformations.

Impact Response and Energy Dissipation Analysis

Simulation results revealed that tyre-based protection systems significantly outperformed EPS foam. The double-layer staggered tyre configuration exhibited the highest internal energy dissipation (EVDDEN), demonstrating superior impact energy redirection. Compared to the unprotected barrier, peak deformation was reduced by 16.08% with a single tyre layer and by 51.16% with double layers, while vertical settlement decreased by 10.90% and 48.79%, respectively.

Robustness under Variable Impact Velocities

Additional simulations across impact velocities ranging from 15 to 30 m/s confirmed the robustness of the tyre-based systems. The double-layer configuration maintained high effectiveness under increasing velocities, with optimal performance observed at 25 m/s. Crack propagation and energy transfer to the concrete structure were significantly mitigated, highlighting recycled tyres as a low-cost, durable, and sustainable solution for rockfall protection.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#HyperelasticMaterials

#CivilEngineering

#MountainInfrastructure

#StructuralSafety

#GreenEngineering

#LowCostSolutions

#InfrastructureResilience

#ProtectiveStructures

#SustainableConstruction

#EngineeringSimulation

#DisasterMitigation

Accident Causality Analysis of Steel Structure Collapses During Construction

Accidents in the construction sector are widely studied from the perspective of occupational safety, with emphasis on fatalities and injuries. However, construction accidents also have profound impacts on construction processes, project continuity, and structural integrity. Among these, steel structure collapses during construction represent critical events that require systematic investigation. This study addresses this gap by conducting a comprehensive analysis of the causes of steel structure collapses during construction.

Steel Structure Collapses as Process Failures

Structural collapses during construction are not isolated incidents but complex failures arising from multiple interacting factors. These events often disrupt construction workflows, cause economic losses, and undermine public confidence in engineering practices. Understanding collapses as process-related failures allows for a broader safety perspective beyond individual worker-related incidents.

AcciMap Framework for Causality Analysis

This study employs the AcciMap methodology to identify and analyze causal relationships across six fundamental system levels. AcciMap enables a holistic examination of accidents by linking technical, organizational, managerial, and regulatory factors. Using this framework, the study reveals how decisions and conditions at different system levels interact to produce catastrophic outcomes during construction.

Case Study: Steel Dome Collapse

A steel dome structure that collapsed during construction was analyzed as a representative case. Accident causality factors were identified through structured expert consultations, ensuring domain-specific accuracy and practical relevance. The case study provides a detailed mapping of contributing factors across the construction lifecycle.

Pareto Analysis and Key Contributing Factors

Based on expert evaluations, a Pareto analysis was conducted to determine the most influential causes of the collapse. A total of 33 causal factors were identified, of which 15 were found to have originated before construction commenced. This finding highlights that many construction-stage failures are rooted in pre-construction decisions, such as design deficiencies, planning errors, and organizational shortcomings.

Implications for Prevention and Safety Management

The results demonstrate that chains of errors during construction are often triggered by systemic factors established during the pre-construction phase. By revealing these hidden causal pathways, the study provides a practical framework for preventing steel structure collapses. This contribution supports improved safety management, risk mitigation, and decision-making throughout the lifecycle of steel construction projects.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#StructuralFailure

#ConstructionManagement

#PreConstructionPlanning

#SystemsEngineering

#WorkplaceSafety

#EngineeringFailures

#CivilEngineeringResearch

#SafetyFramework

#ProjectRisk

#InfrastructureSafety

#BuiltEnvironment

#ConstructionAccidents

Phosphate Biomineralisation for Sustainable Biocementation in Civil and Geoenvironmental Engineering

Over the past two decades, biocementation has emerged as a promising technique for ground improvement, crack healing, and the restoration of building materials and heritage stones. Most existing research has focused on calcite-based biocements produced through urea-hydrolysis pathways. However, the generation of ammonia by-products and limited durability in acidic environments restrict the large-scale and sustainable application of these methods. This review introduces phosphate biomineralisation as a novel and environmentally favorable alternative.

Limitations of Conventional Calcite-Based Biocementation

Microbially induced carbonate precipitation (MICP) and enzymatically induced carbonate precipitation (EICP) rely heavily on urea hydrolysis, which produces ammonia that must be treated or removed. This not only increases environmental risk but also raises operational costs. Additionally, carbonate-based biocements exhibit reduced durability under acidic conditions, limiting their applicability in aggressive soil and environmental settings.

Phosphate Biomineralisation Mechanisms

Phosphate biomineralisation represents an alternative biocementation pathway that avoids ammonia generation. This review examines both microbially induced phosphate precipitation (MIPP) and enzymatically induced phosphate precipitation (EIPP), detailing their biochemical mechanisms, reaction pathways, and mineral formation processes. These mechanisms offer improved chemical stability and environmental compatibility compared to carbonate-based systems.

Applications in Ground Improvement and Soil Stabilisation

Phosphate-based biocements show strong potential for ground improvement, including soil stabilisation and remediation. Their ability to form durable mineral bonds enhances soil strength, reduces permeability, and improves resistance to chemical degradation. These properties make phosphate biomineralisation particularly suitable for challenging geoenvironmental conditions.

Construction Materials and Heritage Conservation

Beyond geotechnical applications, phosphate biocements have promising uses in construction materials and the conservation of heritage stones. Their improved resistance to acidic environments and compatibility with existing materials enable effective crack healing and surface consolidation without compromising material integrity or historical value.

Environmental Sustainability and Future Potential

Phosphate biocementation offers several environmental advantages, including low pH sensitivity, absence of harmful by-products, and alignment with green chemistry principles. Its potential contribution to the United Nations Sustainable Development Goals further supports its viability as a sustainable material. Continued research and development are essential to fully realize the potential of phosphate biomineralisation in civil and geoenvironmental engineering.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#EIPP

#MicrobialEngineering

#EnvironmentalSustainability

#HeritageConservation

#ConstructionMaterials

#LowCarbonMaterials

#CivilEngineeringResearch

#UNSDGs

#Biominerals

#SustainableInfrastructure

#FutureMaterials

 

Numerical Wavefield Simulation Methods for Ultrasonic Analysis in Civil Engineering

Interpreting ultrasonic waveforms in civil engineering structures is inherently complex, particularly when tilted boundaries cause multiple reflections, mode conversions, and overlapping echoes. Wavefield simulations have become an essential tool for analyzing such complex signal behavior and supporting non-destructive testing (NDT) applications. This study focuses on evaluating the accuracy and computational efficiency of commonly used numerical simulation methods for ultrasonic wave propagation in civil engineering contexts.

Challenges in Ultrasonic Waveform Interpretation

Ultrasonic inspections of large civil engineering structures often involve heterogeneous materials and complex geometries that distort waveforms. Tilted interfaces and internal defects introduce signal interference that complicates interpretation. Accurate numerical simulations are therefore critical for understanding wave behavior, validating experimental measurements, and improving defect detection reliability.

Numerical Methods for Wavefield Simulation

This study compares three widely used numerical approaches for ultrasonic wave simulations: the Elastodynamic Finite Integration Technique (EFIT), the Finite Element Method (FEM), and the Spectral Element Method (SEM). EFIT was implemented in Fortran, while FEM and SEM simulations were conducted using COMSOL and Salvus, respectively. These methods differ in their mathematical formulation, numerical accuracy, and computational demands.

Case Study I: Two-Layered Material Validation

In the first case study, simulation results are benchmarked against analytically derived reflection and transmission coefficients for a two-layered material system. The comparison demonstrates that COMSOL and Salvus achieve lower numerical errors than EFIT, although all three methods maintain acceptable relative error levels. This case validates the fundamental accuracy of each numerical approach.

Case Study II: Circular Void Reflection Analysis

The second case study investigates wave reflections from a circular void in a two-dimensional setting, comparing simulated waveforms with analytical solutions. Simulations using progressively refined grids reveal that EFIT and COMSOL require longer computation times to reach the same accuracy level achieved by Salvus. This highlights the superior computational efficiency of the spectral element approach for high-accuracy simulations.

Case Study III: Ultrasonic Echo Array Simulation

The third case study focuses on simulating ultrasonic echo array measurements in a PMMA block containing an internal void. While Salvus and EFIT successfully performed full 3D simulations with comparable accuracy, COMSOL was limited to a 2.5D simulation due to hardware constraints. The results demonstrate that all methods can achieve accurate simulations, but their feasibility depends strongly on computational resources and model dimensionality.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#UltrasonicWaves

#StructuralHealthMonitoring

#ComputationalEngineering

#SimulationAccuracy

#LargeScaleStructures

#EngineeringAnalysis

#WavePropagation

#InfrastructureInspection

#CivilEngineeringResearch

#AdvancedModeling

 

Sustainable Civil Engineering Applications of Non-Metallic Fractions of Waste Printed Circuit Boards (NMF-WPCBs)

The rapid growth of electronic waste has made the recycling of non-metallic fractions of waste printed circuit boards (NMF-WPCBs) a pressing environmental challenge. Due to their complex composition, NMF-WPCBs are often landfilled or incinerated, leading to secondary pollution and loss of valuable resources. This review addresses the urgent need for sustainable reuse pathways by examining the potential of NMF-WPCBs as functional materials in civil engineering applications.

Multiscale Composition and Material Characteristics

NMF-WPCBs are composed of polymers, glass fibers, and residual fillers arranged in a multiscale structure that governs their mechanical and chemical behavior. Understanding these characteristics is essential for their effective integration into construction materials. The review analyzes their physical, thermal, and microstructural properties, which influence bonding, durability, and overall performance in civil engineering systems.

Enhancement Methods and Theoretical Frameworks

To improve compatibility with construction matrices, various enhancement and modification strategies for NMF-WPCBs are reviewed. Theoretical analyses indicate that NMF-WPCBs can accelerate pozzolanic reactions, inhibit alkali–silica reactions, enhance microstructural densification, and restrict crack initiation and propagation. These mechanisms provide a scientific foundation for their performance-enhancing role in construction materials.

Feasibility, Processability, and Safety Assessment

The feasibility of applying NMF-WPCBs in civil engineering materials is evaluated in terms of processability, economic viability, durability, and toxicity. Studies suggest that with appropriate processing and treatment, NMF-WPCBs can be safely incorporated without compromising material integrity or environmental safety. These assessments are crucial for large-scale implementation and industrial acceptance.

Performance in Cementitious and Road Materials

Experimental evidence demonstrates that NMF-WPCBs can significantly improve strength, toughness, and durability in organic cementitious materials. In inorganic materials such as asphalt and road construction composites, they contribute to enhanced pavement performance, aging resistance, and crack resistance. These findings highlight the versatility of NMF-WPCBs across diverse civil engineering applications.

Sustainability Benefits and Future Outlook

In the context of global carbon reduction and circular economy goals, NMF-WPCB-based construction materials offer substantial environmental, economic, and industrial benefits. By transforming electronic waste into value-added construction resources, these materials support sustainable development and resource efficiency. NMF-WPCBs are poised for widespread adoption, positioning themselves as a key component in advancing future sustainable civil engineering practices.

🏗️ Civil Engineering Awards  

👉 Visit our Website: civilengineeringawards.com

#EnvironmentalEngineering

#SustainableMaterials

#CarbonReduction

#EcoFriendlyConstruction

#ConstructionInnovation

#MaterialRecycling

#FutureInfrastructure

#IndustrialSustainability

#CivilEngineeringResearch

#EnvironmentalSustainability