132571 research outputs found
Sort by
Exploring Training Pair-Generation Strategies for Deep Metric Learning for Floor Plan Retrieval
No full textExisting content-based image retrieval models work well for natural photos, but not for images of architectural floor plans. Previous work on floor plan retrieval has focused on graph-based methods, rather than image-based floor plans.Training a CNN-based representation learning framework on segmented floor plan images with standard image augmentations does not result in semantically meaningful retrievals.This work shows that a CNN-based representation learning model can learn features for retrieving floor plans that have similar graphs given the right training signal. Two methods were investigated here: GeomPerturb, a data augmentation that perturbs the underlying geometry of a floor plan, and a weakly supervised method with labels based on the graph edit distance between a pair of floor plans. The results show that while GeomPerturb learns representations that are correlated with the floor plan graph, training with GED labels leads to better retrievals both in terms of the floor plan graph and with respect to room shapes.Computer Scienc
The erosion process of cohesive soil due to a submerged inclined water jet
No full textThis thesis investigates the erosion process of cohesive soil due to a submerged, moving, inclined water jet. This study provides a visualization and description of the failure mechanics of cohesive soil due to a submerged inclined water jet and develops a new equation for estimating the erosion process of cohesive soil while including the angle of the water jet as a parameter. Figures providing support to this new equation are also provided in Chapter 7. The primary variables tested are jetting angle, stand-off distance, forward velocity, and jetting velocity. The study is guided by two primary research objectives. First, to visually and descriptively understand the failure mechanics of cohesive soil subjected to a submerged inclined water jet. Second, to develop a method of predicting the erosion process of cohesive soil, considering variable jet angles, stand-off distances, forward velocity, and jet velocity of a submerged inclined water jet.Experimental testing was conducted in the "Dredging Lab" of Delft University of Technology. The experiment involved eroding cohesive soil blocks using half of a circular nozzle placed along the wall of a flume. This nozzle configuration allowed for the nozzle to be visible during testing. I designed several half nozzles that were tested in a flume provided by TU Delft. Nozzles with a variety of nozzle diameters were designed to have either a 25, 45, 65, 90, 115, 135, or 155 degree jetting angle. Seventy-two clay blocks with a known undrained shear strength were used as the cohesive test soil, allowing for immediate replacement after each test. Detailed experimental procedures are outlined in Section 5.3, offering insights into the design and execution of the tests. The data analysis provides evidence that the jetting angle has a notable impact on the erosion process of cohesive soil. This includes a “deflecting jet” failure mode, the formation and prediction of the sediment plume, and the estimation of the erosion cavity depth. The failure mode of the soil can be seen in Section 5.5 and data analysis figures are provided in Chapter 6. Section 7.1.2 provides a new estimation for the erosion depth of cohesive soil using an inclined water jet (Equation 7.17). This study contributes to our understanding of cohesive soil erosion and methods of estimating erosion processes. The findings emphasize the crucial role of the jetting angle and provide a foundation for future research aimed at refining erosion prediction models and exploring additional parameters influencing the process. Practical applications may include improved design considerations for projects involving water jet erosion such as deep-sea mining, water injection dredging, trailing suction hopper dredgers, as well as other dredging processes involving water jetting.Offshore and Dredging Engineerin
Neural Autoencoder-Based Structure-Preserving Model Order Reduction and Control Design for High-Dimensional Physical Systems
No full textThis letter concerns control-oriented and structure-preserving learning of low-dimensional approximations of high-dimensional physical systems, with a focus on mechanical systems. We investigate the integration of neural autoencoders in model order reduction, while at the same time preserving Hamiltonian or Lagrangian structures. We focus on extensively evaluating the considered methodology by performing simulation and control experiments on large mass-spring-damper networks, with hundreds of states. The empirical findings reveal that compressed latent dynamics with less than 5 degrees of freedom can accurately reconstruct the original systems' transient and steady-state behavior with a relative total error of around 4%, while simultaneously accurately reconstructing the total energy. Leveraging this system compression technique, we introduce a model-based controller that exploits the mathematical structure of the compressed model to regulate the configuration of heavily underactuated mechanical systems.Learning & Autonomous Contro
Measurements of morphodynamics of a sheltered beach along the Dutch Wadden Sea
No full textA field campaign was carried out at a sheltered sandy beach with the aim of gaining new insights into the driving processes behind sheltered beach morphodynamics. Detailed measurements of the local hydrodynamics, bed-level changes and sediment composition were collected at a man-made beach on the leeside of the barrier island Texel, bordering the Marsdiep basin that is part of the Dutch Wadden Sea. The dataset consists of (1) current, wave and turbidity measurements from a dense cross-shore array and a 3 km alongshore array; (2) sediment composition data from beach surface samples; (3) high-temporal-resolution RTK-GNSS beach profile measurements; (4) a pre-campaign spatially covering topobathy map; and (5) meteorological data. This paper outlines how these measurements were set up and how the data have been processed, stored and can be accessed. The novelty of this dataset lies in the detailed approach to resolve forcing conditions on a sheltered beach, where morphological evolution is governed by a subtle interplay between tidal and wind-driven currents, waves and bed composition, primarily due to the low-energy (near-threshold) forcing. The data are publicly available at 4TU Centre for Research Data at: https://doi.org/10.4121/19c5676c-9cea-49d0-b7a3-7c627e436541 (Van der Lugt et al., 2023).Coastal EngineeringLab Hydraulic EngineeringEnvironmental Fluid MechanicsCivil Engineering & Geoscience
Integrating geospatial, remote sensing, and machine learning for climate-induced forest fire susceptibility mapping in Similipal Tiger Reserve, India
No full textAccurately assessing forest fire susceptibility (FFS) in the Similipal Tiger Reserve (STR) is essential for biodiversity conservation, climate change mitigation, and community safety. Most existing studies have primarily focused on climatic and topographical factors, while this research expands the scope by employing a synergistic approach that integrates geographical information systems (GIS), remote sensing (RS), and machine learning (ML) methodologies for identifying and assessing forest fire-prone areas in the STR and their vulnerability to climate change. To achieve this, the study employed a comprehensive dataset of forty-four influencing factors, including topographic, climate-hydrologic, forest health, vegetation indices, radar features, and anthropogenic interference, into ten ML models: neural net (nnet), AdaBag, Extreme Gradient Boosting (XGBTree), Gradient Boosting Machine (GBM), Random Forest (RF), and its hybrid variants with differential evolution algorithm (RF-DEA), Gravitational Based Search (RF-GBS), Grey Wolf Optimization (RF-GWO), Particle Swarm Optimization (RF-PSO), and genetic algorithm (RF-GA). The study revealed high FFS in both the northern and southern portions of the study area, with the nnet and RF-PSO models demonstrating susceptibility percentages of 12.44% and 12.89%, respectively. Conversely, very low FFS zones consistently displayed susceptibility scores of approximately 23.41% and 18.57% for the nnet and RF-PSO models. The robust mapping methodology was validated by impressive AUROC (>0.88) and kappa coefficient (>0.62) scores across all ML validation metrics. Future climate models (ssp245 and ssp585, 2022–2100) indicated high FFS zones along the northern and southern edges of the STR, with the central zone categorized from low to very low susceptibility. Boruta analysis identified actual evapotranspiration (AET) and relative humidity as key factors influencing forest fire ignition. SHAP evaluation reinforced the influence of these factors on FFS, while also highlighting the significant role of distance to road, distance to settlement, dNBR, slope, and humidity in prediction accuracy. These results emphasize the critical importance of the proposed approach for forest fire mapping and provide invaluable insights for firefighting teams, forest management, planning, and qualification strategies to address future fire sustainability.Geo-engineerin
Impact of calibrating a low-cost capacitance-based soil moisture sensor on AquaCrop model performance
No full textSensor data and agro-hydrological modeling have been combined to improve irrigation management. Crop water models simulating crop growth and production in response to the soil-water environment need to be parsimonious in terms of structure, inputs and parameters to be applied in data scarce regions. Irrigation management using soil moisture sensors requires them to be site-calibrated, low-cost, and maintainable. Therefore, there is a need for parsimonious crop modeling combined with low-cost soil moisture sensing without losing predictive capability. This study calibrated the low-cost capacitance-based Spectrum Inc. SM100 soil moisture sensor using multiple least squares and machine learning models, with both laboratory and field data. The best calibration technique, field-based piece-wise linear regression (calibration r2 = 0.76, RMSE = 3.13 %, validation r2 = 0.67, RMSE = 4.57 %), was used to study the effect of sensor calibration on the performance of the FAO AquaCrop Open Source (AquaCrop-OS) model by calibrating its soil hydraulic parameters. This approach was tested during the wheat cropping season in 2018, in Kanpur (India), in the Indo-Gangetic plains, resulting in some best practices regarding sensor calibration being recommended. The soil moisture sensor was calibrated best in field conditions against a secondary standard sensor (UGT GmbH. SMT100) taken as a reference (r2 = 0.67, RMSE = 4.57 %), followed by laboratory calibration against gravimetric soil moisture using the dry-down (r2 = 0.66, RMSE = 5.26 %) and wet-up curves respectively (r2 = 0.62, RMSE = 6.29 %). Moreover, model overfitting with machine learning algorithms led to poor field validation performance. The soil moisture simulation of AquaCrop-OS improved significantly by incorporating raw reference sensor and calibrated low-cost sensor data. There were non-significant impacts on biomass simulation, but water productivity improved significantly. Notably, using raw low-cost sensor data to calibrate AquaCrop led to poorer performances than using the literature. Hence using literature values could save sensor costs without compromising model performance if sensor calibration was not possible. The results suggest the essentiality of calibrating low-cost soil moisture sensors for crop modeling calibration to improve crop water productivity.Water Resource
District heating with complexity: Anticipating unintended consequences in the transition towards a climate-neutral city in the Netherlands
No full textDistrict heating systems are considered a feasible heating alternative to replace natural gas to mitigate emissions in cities. However, urban transitions are very complex because energy systems often operate in densely populated areas, which gives rise to all kinds of interdependencies in cities. These interdependencies can result in unintended consequences which can indirectly help or hinder urban energy transitions. Understanding these influences the transition to climate neutrality. This research investigates the lessons learned from a project conducted in Rotterdam: a high-density city in the Netherlands which is expanding its district heating systems. We use qualitative system dynamics models to explore the underlying complexity and to recognize indirect consequences of policies. Our results cover both technologically oriented and policy-oriented insights, contributing to the literature on transition governance in cities. On the one hand, the national and urban strategies in the Netherlands activate mechanisms that support cities with district heating systems such as Rotterdam. On the other hand, the same strategies could also lead to a potential rivalry between energy efficiency and energy security, which are both crucial goals in urban transition governance. Participative modeling provides policy-makers with an analytical tool to detect systemic dependencies which can be used to identify synergies and barriers among different energy policy objectives. This helps avoiding potential unintended consequences including the use of carbon-heavy systems and displacing investments from energy efficiency and renewable heating systems.Organisation & Governanc
Multi-criteria design methods in façade engineering: State-of-the-art and future trends
No full textFaçade engineering is facing an era of extraordinary challenge to meet the surge in demand for buildings that are environmentally sustainable and enhance occupant wellbeing. Facades, also known as building envelopes, play a major role in the resource-efficiency of buildings and the quality of its indoor environment. Consequently, the development of effective design approaches is crucial for generating appropriate façade solutions. Façade design is complex and multi-disciplinary involving several and oftentimes conflicting performance criteria. Systematic and holistic design procedures are, therefore, required to achieve optimal trade-offs. Over the last decades, researchers in this field have used computational tools and power to address this challenging problem within the context of multi-criteria design approaches. This paper reviews the existing research in this field, and presents the state-of-the-art review from simple to advanced decision-making procedures currently used at the early design stages, where decisions have a disproportionally large impact on the façade performance. The paper provides a complete description of the design variables and objectives typically involved. Alternative multi-criteria design methodologies regarding discrete decisions and automated optimization are reviewed, each with salient pros/cons, and overall conclusions are drawn. Finally, the paper discusses ongoing trends and research needs, namely, the development of uncertainty-based procedures to enable more informed decision-making; the inclusion of structural/seismic safety considerations in the design process to achieve higher socio-economic benefits; the integration of smart building information modeling and processing technologies to facilitate smarter design decisions; and the adoption of integrated design approaches to promote climate-adaptive solutions that enhance resilience.Architectural Technolog
Timber-glass shear wall stabilised timber modules: Application of a timber-glass shear wall as stabilising element in a mid-rise modular timber building
No full textCurrently, most of the modular building can be fabricated off-site, except for the stability system. Therefore, the question arises: Could modules be designed such that there is no need for an additional stability system? As modules are typically built in a rectangular shape, incorporating stability elements such as bracings along the longer side of the module is not the problem. The main challenge lies in the limited length available for stability elements on the shorter side of the module, combined with the fact that this shorter side is commonly used for windows and openings for door frames. This introduces a conflicting interest between structural capacity and daylight within the module. This conflicting interest becomes more and more prominent as the building height goes up. A possible solution to this problem can be found in the use of a load-bearing window frame with glass infill as a stability element, often referred to in literature as a timber-glass shear wall (TGSW).Therefore, this thesis will answer the main research question: 'To what extent can the structural performance of a timber-glass shear wall as a stability element in a timber module be used to accommodate for the stability of a mid-rise modular timber building?'The approach to answering the main research question consists of several steps. First, a literature study was conducted on modular buildings and TGSW's. The outcome of the study has provided insight into how modular buildings are constructed in general and how relevant aspects such as progressive collapse, fire safety design, and foundation design influence structural design. The study also resulted in an analytical prediction model for the load-bearing capacity and stiffness of the TGSW. Through this prediction model, it became clear how the properties of individual components relate to the load-bearing capacity and stiffness of the total TGSW-system.The second step was to propose a design for a modular timber building composed of timber modules. To save on computational time, the stability elements of the building are modelled using steel diagonals as an equivalent system for the TGSW. The cross-sectional area of the steel diagonals is directly related to the properties of the TGSW. Therefore, the steel diagonals have identical stability properties as the TGSW. In this study, varying the type of adhesive and the spacing of the screws was found to have the most significant impact on the overall structural properties of the TGSW. The horizontal connections are made of steel plates fastened with screws. The vertical connections are realised by shear plate connectors. The entire building was modelled in a 3D FEM programme to assess the structural behaviour of the building and its compliance with building regulations. Several building configurations ranging from 1:1 to 1:3 height-to-width ratio were investigated. For each building configuration, the cross-sectional area of the steel diagonals was adjusted within a specified range. This range corresponds to variations in adhesive type or screw spacing. As a result, design graphs were produced, which present the requirements for the load-bearing capacity and stiffness of the stability system. These can be compared to the load-bearing capacity and stiffness of the TGSW. This comparison can be used as a validation method to determine the viability of the TGSW stability element in a modular building.The results of this study indicate that a modular building can be stabilised by a TGSW up to six stories within the height-to-width ratio of 1:1 to 1:3. The minimum building configurations per story height are: 3 modules high by 5 modules wide, 4 by 8 modules, 5 by 12 modules, and 6 by 18 modules. These slenderness ratios were governed by the strength of the TGSW. The limiting factor in the load-bearing capacity is the shear strength of the adhesive. These slenderness ratios could only be reached with elastic adhesives such as silicones. The next step is to create a more extensive FEM model that could predict the load-bearing capacity and stiffness of the TGSW in a more accurate way compared to an analytical model. Furthermore, exploring the performance of the TGSW under different horizontal loads, such as earthquakes, would give valuable insight.Civil Engineering | Building Engineering - Structural Desig
Co-Creation for Sustainable Energy Transition: A Case Study of Local Energy Cooperatives in the Metropolitan Region Rotterdam The Hague
No full textThe irreversible effects of climate change have led to a significant increase in citizen-led and initiated local energy cooperatives. Playing a crucial role in the shift to renewable energy, these cooperatives are evolving from informal community groups to structured organisations. Their success relies on co-creation, a collaborative process in which stakeholders and citizens join forces to improve efficient decision-making, build trust and promote shared responsibility. This leads to creating effective solutions and achieving collective goals across projects and sectors. This thesis examines the role of co-creation in local energy cooperatives and highlights its importance in promoting sustainable energy practices and empowering citizens. It focuses on investigating co-creation within the energy cooperatives of the Metropolitan Region Rotterdam The Hague (MRDH) and connects theoretical concepts with practical applications in energy transition. The central research question is: ”In what ways does co-creation manifest within local energy cooperatives in the Metropolitan Region Rotterdam The Hague?”A literature review was conducted using the PRISMA method and snowball technique to understand the academic perspectives of co-creation, especially in energy transition. This helped identify the scope of the thesis and gaps in current academic knowledge. The thesis contains two theoretical frameworks. The first, from Puerari et al. (2018), examines the dynamics of co-creation in local communities and identifies five key elements: intended purpose, process type, ownership, motivations and incentives, and spaces and places. This research applied these elements to understand co-creation in MRDH’s local energy cooperatives through qualitative case study analysis. The analysis reveals the cooperatives’ commitment to fossil-free energy and their encouragement of citizen participation through a mix of formal and informal methods, addressing different motivations and adopting a shared ownership model to promote community involvement.The second framework consists of six criteria that define co-creation activities. These criteria are developed in this study, and derived from academic literature. These criteria include shared goals, active participation, equality and inclusiveness, iterative processes, value creation and mutual learning. Subsequently, the study used these criteria to evaluate the activities of energy cooperatives and assess their alignment with the concept of co-creation. Through interviews and observations, four key cooperative activities were identified: advisory services, information generation, renewable energy production and stakeholder engagement. While these activities meet the criteria for co-creation, there are opportunities for improvement in almost all areas of co-creation to fully realise the potential of these activities. The findings suggest that cooperatives should organise regular stakeholder discussions, introduce paid functions, make more effective use of physical spaces, promote diversity and develop digital platforms for knowledge sharing. Policymakers can support these cooperatives by recognising their contributions, encouraging professional development and encouraging flexibility and innovation. Future research recommendations address the limitations of this study such as regional focus, time constraints and possible subjectivity of the qualitative method. Suggestions include expanding the geographical scope, involving a wider range of stakeholders, using mixed methods and testing the recommended strategies in different contexts.In summary, this research is an important step in understanding co-creation manifestation within local energy cooperatives. It suggests strategies through which these cooperatives, in collaboration with policymakers, can effectively contribute to environmental sustainability and climate change mitigation.Complex Systems Engineering and Management (CoSEM
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇