Procter & Gamble (United Kingdom)
GraFar - Repository of the Faculty of Civil EngineeringNot a member yet
3934 research outputs found
Sort by
Comparative Analysis of Machine Learning Models for Predicting Interfacial Bond Strength of Fiber-Reinforced Polymer-Concrete
This study presents a detailed analysis of various machine learning models for
predicting the interfacial bond strength of fiber-reinforced polymer (FRP) concrete, including
multiple linear regression, Multigene Genetic Programming (MGGP), an ensemble of
regression trees, Gaussian Process Regression (GPR), Support Vector Regression (SVR),
and neural networks. The evaluation was based on their predictive accuracy. The optimal
model identified was the GPR ARD Exponential model, which achieved a mean absolute
error (MAE) of 1.8953 MPa and a correlation coefficient (R) of 0.9658. An analysis of this
optimal model highlighted the most influential variables affecting the bond strength. Additionally,
the research identified several models with lower expression complexity and
reduced accuracy, which may still be applicable in practical scenarios
Kalibracija hidroloških modela za potrebe analize uticaja klimatskih promena
Jedan od najvažnijih zadataka hidrologije jeste pouzdana
procena uticaja klimatskih promena na vodne resurse.
Ove analize se zasnivaju na klimatskim projekcijama,
koje predstavljaju ulazne podatke za hidrološke modele.
Uticaj klimatskih promena se kvantifikuje poređenjem
budućih vrednosti hidroloških veličina, i njihovih vrednosti u referentnom (istorijskom) periodu. Jedna od najznačajnijih veličina u ovim analizama je unutargodišnja raspodela protoka koja pokazuje sezonske varijacije oticaja
na slivu. Iako predstavlja jedan od najznačajnijih hidroloških potpisa, efikasnost modela u reprodukovanju unutargodišnje raspodele protoka se veoma retko razmatra u
inženjerskoj praksi. U ovom radu je hidrološki model
sliva reke Ribnice kalibrisan prema nekoliko kriterijumskih funkcija, čime je kreirano nekoliko njegovih verzija.
Svaka verzija modela je detaljno analizirana sa aspekta
mogućnosti da reprodukuje unutargodišnju raspodelu
protoka. Rezultati pokazuju velike varijacije u efikasnosti
modela, ali i da verzije modela koje su kalibrisane prema
kompozitnim kriterijumskim funkcijama imaju najbolje
slaganje sa unutargodišnjim raspodelama osmotrenih
protoka. Rezultati takođe pokazuju da ni jedan pokazatelj
(odnosno, kriterijumska funkcija u kalibraciji modela) ne
odražava u potpunosti različite aspekte efikasnosti modela. Stoga je odabir kriterijumske funkcije veoma važan
korak u izradi hidrološkog modela, i treba da se rukovodi
konačnom namenom modela (eng. fitness for a said purpose, [1])
3D Building Models Generation from Airborne Lidar and Photogrammetry Data
This paper considers procedures and software tools for generating 3D models of buildings using airborne laser scanning and aerial photogrammetry data. Various software tools, such as Agisoft Metashape and DJI Terra, can be used for generating 3D point clouds and 3D meshes using photogrammetry and computer vision techniques. 3D point cloud can be used as a geometric reference for 3D modelling of buildings. Other solutions are based on extracting building footprints and calculating building heights necessary for 3D buildings extrusion. Advanced algorithms enable more or less automatic generation of 3D building models with varying Levels of Detail (LOD), with LOD2 providing sufficient representation for urban planning and visualization without capturing intricate architectural details. Additional editing of obtained models is required if high-quality models are required. The research is focused on assessment of different procedures and tools in terms of their efficiency and the quality of the reconstructed 3D models. Applicability of different approaches for generating city-wide 3D models of buildings is also considered. Therefore, the main objective of this study is the assessment of the effectiveness and efficiency of various tools in automating the process of generating 3D building models from diverse data sources. This assessment is based on processing the data captured for selected study areas by using different tools and procedures. Data preprocessing involved point cloud classification, with a focus on the roof and terrain classes, which are critical for model generation. Following classification, points were clustered to represent individual buildings, resulting in the creation of 3D building models
Strategic contribution of the water management sector to the sustainable development of irrigation under climate change, as a basis for intensive and reliable agricultural production
Klimatski trendovi pokazuju veoma izraženu vremensku i prostornu neravnomijernost padavina. Ova pojava se iskazuje kao „resursni paradoks“ u slučaju kada su padavine najoskudnije u vegetacionom periodu i u područjima u kojima su koncentrisani najkvalitetniji i najveći zemlјišni resursi za poljoprivrednu proizvodnju. Zbog toga se stvaraju oštra ograničenja razvoja agrarnog kompleksa, od primarne proizvodnje, do najviših nivoa finalizacije. Zaostajanja u agrarnom kompleksu djeluju destabilizirajuće na ostale oblasti privređivanja, što stvara ekonomski, društveni i tehnički okvir za razvoj polјoprivrede u uslovima uređenih i upravlјanih vodnih režima. U navedenim uslovima, navodnjavanje postaje jedan od klјučnih razvojnih, ne samo polјoprivrednih, već i društvenoekonomskih cilјeva i prioriteta.
Uređeni i upravljani vodni režimi podrazumijevaju obezbjeđenje vodnog resursa za potrebe navodnjavanja. To znači da voda za navodnjavanje treba da ima odlike socijalne, ekonomske i ekološke kategorije, odnosno pored tri atributa vode prisutne na slivu, koja se definiše matričnom strukturom V = L,Q,K, (L - lokacija, Q - količina i K - kvalitet vode), mora da posjeduje i četvrti, izuzetno važan - postojanje uslova za zahvatanje, korišćenje i zaštitu voda. Pri tome je veoma važno ispuniti složenu ciljnu strukturu višenamjenskog korišćenja voda, kao i potrebne doprinose zaštite životne sredine.
U integralnom upravljanju vodama sektor voda treba da prepozna raspoložive kapacitete vodnog resursa za potrebe navodnjavanja i ponudi moguća tehničko-tehnološka rješenja sektoru poljoprivrede. Uz pouzdane prostorne identifikacije perspektivnih poljoprivrednih površina, koordinisani pristup će nedvosmisleno pokazati da pojavu „resursnog paradoksa“ treba rješavati integrisano sa drugim sektorima, odnosno planiranjem mikro, manjih, srednjih ili velikih akumulacija, koje će imati jednonamjensku ili višenamjensku ulogu.
U ovom radu daje se analiza strateške podrške sektora voda, razvoju navodnjavanja. Konkretni primjeri trebalo bi da budu stručna osnova u formiranju ozbiljnog strateško-planskog dokumenta, koji će prije svega definisati raspoložive potencijale i strategiju jačanja kapaciteta intenzivne poljoprivredne proizvodnje u Bosni i Hercegovini u uslovima klimatskih promjena primjenom navodnjavanja.Climate trends show an extreme temporal and spatial unevenness of precipitation. This phenomenon manifests as a “resource paradox” in cases where precipitation is scarcest during the growing season and in areas with concentrated most fertile and extensive land resources for agricultural production. As a result, severe constraints arise for the development of the agricultural sector, from primary production to the highest levels of processing. Lagging in the agricultural sector has a destabilizing effect on other sectors of the economy, creating an economic, social, and technical framework for developing agriculture within regulated and managed flow regimes. Under such conditions, irrigation becomes one of the key development priorities, not only for agriculture but also for broader socioeconomic progress.
Regulated and managed flow regimes involve the provision of water resources for irrigation needs. This means that irrigation water should be perceived as a social, economic, and ecological asset. In addition to the three attributes of water within a river basin (defined by the matrix structure V = ⟨L, Q, K⟩, where L stands for location, Q for quantity, and K for water quality), it must also include a fourth, equally important attribute: the existence of conditions that allow for the abstraction, use, and protection of water resources. It is crucial to address the diverse objectives of multipurpose water management and to provide the necessary support for environmental protection.
In integrated water management, the water sector should identify the available capacities of water resources for irrigation purposes and provide potential technical and technological solutions to the agricultural sector. Combined with reliable spatial identification of prospective agricultural areas, a coordinated approach will clearly demonstrate that the phenomenon of the “resource paradox” must be addressed in an integrated manner with other sectors, that is, through the planning of micro, small, medium, or large reservoirs, which may serve specific or multipurpose functions.
This paper presents an analysis of the strategic support provided by the water sector for the development of irrigation. The specific examples discussed are intended to serve as a professional foundation for the preparation of a comprehensive strategic and planning document, which will primarily define the available potentials and outline a strategy for strengthening the capacity of intensive agricultural production in Bosnia and Herzegovina under climate change conditions through the implementation of irrigation
RC frames with isolated/decoupled masonry infills: advantages and benefits for practice
Reinforced concrete (RC) frame structures with masonry infills are commonly used in earthquake-prone areas. Infill walls experience both in-plane and out-of-plane loads during seismic activity and are often damaged as a result. Recent studies have shown that the combination of these loads poses significant risks to masonry infills, but structures as well. Traditional mortar-based connections tend to fail under such conditions, leading to the collapse of the infill walls and in-crease of shear forces acting to the RC columns. To address this issue, a new system called INODIS was developed, which decouples the infill from the frame using recycled rubber strips to connect the infill wall to the RC frame. This system was tested on full-scale specimens under quasi-static cyclic in-plane and out-of-plane loading. Initial cracks appeared at 2% drift, indicating the system's potential to minimize infill damage and lower repair costs. INODIS system was further investigated under the real earthquake excitation in the shaking-table test, where 2.6% of drift was reached without any damage in the infill walls. Comprehensive experimental campaign confirms that INODIS system offers a solution for enhancing seismic resilience and economic benefits by reducing the amounts of needed reinforcement and concrete for structural elements. Straightforward design concept was developed that enables simple application in the practice, suitable for every brick type and size
INODIS – A Novel Decoupling System to Improve the Earthquake Resistance of Masonry Infills
Masonry infills are widely used as internal and external partitions in reinforced concrete frame structures. However, recent earthquakes have revealed that the interaction between masonry infills and RC frames often leads to severe, uncontrolled damage, highlighting their high seismic vulnerability. Such damage not only results in significant economic losses but also poses serious risks to human safety. The poor seismic performance of masonry infills underscores the urgent need for effective mitigation strategies. This contribution introduces the innovative decoupling system INODIS designed to isolate masonry infills from the surrounding RC frame. The proposed solution involves inserting recycled elastomeric strips between the infills and the frame, thereby delaying the activation of the infills under in-plane seismic loading while still offering support against out-of-plane forces. Cyclic static experimental tests were conducted on RC frames with both rigidly connected and decoupled infills, including configurations with and without openings. The results demonstrate a significant improvement in the seismic performance of masonry infills when the decoupling system is applied. Furthermore, the system was validated by shaking table tests under real earthquake excitation. Finally, a simple and safe design concept based implemented in the second generation of Eurocode 8 is presented and applied to a typical RC frame structure
Modeling of the vertical gravity gradient in the territory of the Republic of Kazakhtan: Methods, Comparison and Evaluation of Accuracy
This study investigates the modeling of the vertical gravity gradient (VGG) in the Republic of Kazakhstan using three non-instrumental methods: global geopotential models (GGMs), a precomputed VGG grid from the Institute of Geophysics and Planetary Physics (IGPP), and a method combining the normal gravity field with digital elevation models (DEMs). The research compares these modeled VGG values with high-precision measurements from 48 outdoor gravity stations, employing Scintrex CG-6 gravimeters for indirect VGG determination. Key findings reveal that topographic modeling is unreliable for stations located in depressions due to unaccounted mass above benchmarks. At the same time, GGMs provide results comparable to the normal gradient but lack resolution for stations on elevated terrain. After excluding problematic stations, modeling accuracy improved to approximately 10 μGal/m. The study concludes that while direct VGG measurements remain essential near the Earth's surface, modeled values can offer better accuracy than theoretical corrections in specific scenarios, depending on the detail of input data. The research highlights the importance of VGG in geodesy and gravimetry, particularly for high-precision applications. It underscores the limitations of current modeling techniques in accounting for local topographic and subsurface density variations. The work contributes to refining methods of gravity data reduction and supports the development of Kazakhstan's geodetic infrastructure
Methods of testing systems for strengthening masonry structures based on cement composites reinforced with glass fiber mesh
The paper presents the testing methods used for the contemporary masonry strengthening system, FRCM (fibre-reinforced cementitious matrix), which consists of glass fabric and mortar matrix. The FRCM system is suitable for strengthening masonry structures since it doesn't significantly change their mass and stiffness but increases the capacity for certain loads, especially ones induced by earthquakes. Following the recommendations, the mechanical properties of component materials and three groups (fibers, fibers in mortar, and masonry system) of specimens were tested. The main parameters necessary for calculating this system were defined and compared with the results found in the literature. It was shown that the chosen geometry for the mechanical test was appropriate for comparing the results within different groups. The importance of the type of fixtures used for the specimens (clevisor clamping-grip) was more pronounced in the second group of specimens. Results of the first and the third group of specimens were in accordance with the literature, with the ultimate tensile stress of the dry fabric reaching 742,2 MPa, and the ultimate tensile strain of the dry fabric reaching 1.53%.The conventional strain obtained from the third group of specimens was 1.28%
Numerical modelling of CFRP internally reinforced glulam beams
Numerous studies have proven that timber structures can be effectively reinforced using fibre-reinforced polymer (FRP) composites. In this paper a nonlinear finite element model was developed to predict bending behaviour of glulam beams reinforced with carbon fibre reinforced polymer (CFRP) plates strategically located in the tension zone between the bottom two laminations. The developed model was validated through comparison with the results of experimental tests for both unreinforced and reinforced beams. To accurately simulate the mechanical
behaviour of hybrid glulam-CFRP members, suitable constitutive relations for each material were utilised in the model. The theory of anisotropic plasticity was implemented to include plastic behaviour of timber laminations in the compression zone. The Hill’s criterion for orthotropic materials was used as a condition for transition to the plastic state. The progressive damage model was introduced to effectively simulate softening behaviour of timber in tension. The FEM results have shown excellent agreement with the experimental results. Nonlinear behaviour of glulam beams internally reinforced with CFRP was achieved in the numerical analysis, demonstrating the accuracy of developed model past the linear-elastic range
Eksperimentalno ispitivanje CLT panela sa modifikovanom orjentacijom poprečnih slojeva
Eksperimentalno ispitivanje je sprovedeno u cilju utvrđivanja naponskog i deforamcijskog stanja panela od unakrsno lepljenog lameliranog drveta (CLT-a) izloženih savijanju van ravni kako bi se analizirali efekti različite orijentacije lamela u okviru poprečnih slojeva panela. Ispitani su paneli sa konvencionalnom orijentacijom lamela poprečnih slojeva pod uglom 90° u odnosu na lamele podužnih slojeva i paneli sa modifikovanom orijentacijom lamela poprečnih slojeva pod uglom ±45° u odnosu na lamele podužnih slojeva. Prednosti modifikovanog CLT-a su poboljšane mehaničke performanse i efikasnije korišćenje resursa uzimajući u obzir svojstva materijala. Ponašanje panela pri ekploatacionom i graničnom opterećenju opisano je sa aspekta veze opterećenje-ugib, oblika loma, nosivosti i krutosti