Pure OAI Repository
Not a member yet
421387 research outputs found
Sort by
In situ acoustic characterization of a porous layer backed by a large air cavity
No full textThe in situ measurement of acoustic surfaces presents a significant challenge in room acoustics, as it is often impractical to conduct laboratory measurements of already installed materials. In a former study, the in situ analysis of porous samples that react locally when supported by a solid wall demonstrated a good degree of accuracy. Nevertheless, when a porous layer is supported by a large air cavity (depth >100 mm), a situation commonly seen in suspended ceiling designs, the air cavity exhibits a non-locally reacting behavior; thus, the local reaction cannot be reliably assumed. This study introduces a method to characterize such a non-locally responding system through in situ PU probe measurements, utilizing an inverse technique to fit the parameters of the impedance model of a porous layer that is backed by an infinite air layer, based on the measured reflection coefficient. The precision of the approach was confirmed through 2D numerical simulations, indicating that the method produced reliable results for air cavities of 200 mm or deeper. The method was then experimentally validated on systems comprising several porous layers supported by air cavities of varying depths. Good agreement was obtained between the parameters measured experimentally using the proposed technique and the references, even in cases where the air cavity was less than 200 mm deep. Additionally, the proposed method demonstrated more precise characterization results compared to those achieved by fitting the parameters of an impedance model based on a standard multilayer model.</p
Supporting Analysts and Managers to Utilize Prescriptive Process Monitoring:A User Interface Design and Evaluation
No full textPrescriptive process monitoring techniques recommend actions in an ongoing case of a business process to maximize its success rate. Different techniques have been proposed that focus on the efficiency and precision of recommendations. In contrast, little attention has been given to presenting the outputs of techniques to end users. In this study, we design an interface for prescriptive process monitoring outputs following the design science research methodology. As an artifact, we develop a web-based tool Kairos. We evaluate the tool with end users (operational managers and process analysts). Based on the findings, we derive suggestions for designing prescriptive process monitoring interfaces
Embodied computation for emergent goal-oriented behavior in soft robots
No full textIn living organisms, directed behavior arises from repeated rhythmic (oscillatory) motions whose sequence and timing are robustly coordinated. This coordination is typically in part or even fully distributed throughout the organism: animals employ central pattern generators within their nervous systems, plants utilize distributed mechanoreceptors, and fungi leverage expansive mycelial networks. Such decentralized orchestration offloads computation from a central brain to the body, allowing behaviors to emerge naturally through interactions between the body and its environment. This thesis explores alternatives to centralized control inspired by decentralized systems in nature. It identifies sequences and timing of oscillations leading to directed locomotion in soft robots. We aim to embody directed behavior in the physical system so that purposeful actions emerge from local body-environment interactions and feedback. Through an exploratory study spanning design, simulation, and hardware, we demonstrate how soft robotic systems can leverage their embodied mechanical intelligence using embodied computation to achieve complex autonomous behaviors without a centralized processor. As a start, we draw inspiration from the physiology and decentralized nervous system of echinoderms (e.g., sea urchins, brittle stars, feather stars, and sea cucumbers) to examine how decentralized feedback can facilitate directed locomotion towards a light source (phototaxis) in limbed soft robots. We build a modular system where each limb is a self-contained module that stochastically optimizes its behavior with a feedback loop based on limited sensing, short-term memory, and computation. By harnessing the inherent mechanical intelligence of soft pneumatic actuators, cyclic on-off inputs to a pump at a fixed frequency are converted into complex bending and stepping motions. By physically connecting multiple limbs and letting each limb independently learn the phase of its oscillating motion, coordination between the limbs emerges. We show that, similar to echinoderms, such as sea stars, interactions of the individual limbs with the environment guide the robot toward coordinated movement patterns without relying on comprehensive full-body representations or complex algorithms. The soft robot dynamically re-coordinates its movement in response to changing conditions (changes in actuators and damage) without any central controller. Resilient, whole-body locomotion thus emerges from the interplay of many basic units, each with limited memory and no body awareness, demonstrating a route of adaptable goal-directed movement sequencing in soft robotics through embodied computation. To gain a better understanding of how this coordination emerges, we build a second modular system of self-contained units. In this system, the modules use the same strategy for sensing and processing, but we limit the actuation, making them immobile on their own. Instead, they expand and contract their connections to the other physically connected modules on a two-dimensional plane. When interconnected in two-dimensional grid configurations, the system as a whole can break the symmetry of the friction to achieve locomotion, similar to earthworms that expand and contract segments. By combining simulations and experiments, we gain an understanding of how this decentralized strategy can follow locally optimal sequences solely from the implicit communication facilitated through their physical connection (as the system moved toward the light, the connected units all increased their light intensity). The simulations also provide insight into how the sequences that the system produces are linked to the potential behaviors of the system and how these change with different configurations and in dynamic, unstructured environments. These results not only demonstrate that robust, directed locomotion in soft robots can emerge entirely from local environmental interactions but also show the profound link between the coordination strategy and the body morphology. They also illustrate the dynamic nature of the learning process as it adapts to changing, partially observable environments. While the work mentioned above focuses on reducing the hardware and complexity of algorithms needed to coordinate the sequences of movements starting from random behaviors, it still requires many electronic components to make the individual modules. Therefore, we next aim to embody sequences without relying on electronics, by harnessing soft fluidic circuits with integrated magnetic components. By designing a fluidic relaxation oscillator that produces an oscillating output for a fixed input flow, we can encode the rhythmic inflation-deflation cycles into a single component. We implement directional air-driven coupling between the relaxation oscillators to emulate biological central pattern generators, orchestrating the rhythmic motions without electronics. By altering the fluidic coupling between them, we demonstrate rapid and reversible reprogramming of the oscillation sequences and timings. Such physically embodied control paves the way for soft robotic systems equipped with decentralized locomotion primitives, eliminating dependence on complex electronics and centralized controllers. Lastly, the approaches above start with predetermined morphologies, whereas natural organisms demonstrate how body morphology and embodied computation evolve synergistically over longer time scales. Inspired by this co-evolutionary principle, we simulate coupled oscillator networks as mentioned above, integrated within evolving soft robotic morphologies. We show that oscillator networks with minimal complexity (number of oscillators and number of connections), when co-designed with the body morphology, can enable robots to transition spontaneously between distinct behaviors, such as climbing or running, in response to environmental feedback. These findings emphasize how thoughtful morphological and feedback co-design can embed rich, context-specific behaviors into relatively simple physical structures. Collectively, this work contributes to the broader vision of autonomous soft robots with distributed intelligence, where sophisticated, goal-directed behaviors emerge from the continuous interplay of body and environment rather than explicit centralized command. These results, using embodied computation, pave the way towards soft robots that harness their mechanical intelligence to complete tasks autonomously in real-world scenarios
Cooperative Multi-Agent Advice Exchange via Topological Graph Learning
No full textAdvice exchange (AE) is a commonly used approach to enhance the performance of multiagent reinforcement learning (MARL). It refers (requesting) agents to beneficial behaviors of (target) agents and thus facilitates efficient policy learning in a multiagent system (MAS). However, traditional AE approaches often depend on polling all agents, causing substantial communication costs and computational effort. Moreover, they adopt manually designed rules to select teacher agents, which ignore the natural topology in MAS and limit policy learning. In this article, we propose a cooperative multiagent advice exchange via topological graph learning (ToGAE), which entails the similarity of knowledge domains among agents in cooperative MAS. ToGAE enables agents to select their corresponding target agents with the largest knowledge domain similarity for AE. The knowledge domain similarity is extracted by a two-stage graph attention network with Jensen–Shannon divergence-based training loss and favorably reflects the functional relationship between two agents. In addition, we design a clarity-based advice acceptance scheme to avoid the blind execution of any advice and thus further boost the efficiency of MARL. Extensive experiments show that ToGAE significantly outperforms the baseline methods in terms of the efficiency of policy learning and performance on different MARL tasks.</p
Evaluation of Human Visual Privacy Protection:A Three-Dimensional Framework and Benchmark Dataset
No full textRecent advances in AI-powered surveillance have intensified concerns over the collection and processing of sensitive personal data. In response, research has increasingly focused on privacy-by-design solutions, raising the need for objective techniques to evaluate privacy protection. This paper presents a comprehensive framework for evaluating visual privacy-protection methods across three dimensions: privacy, utility, and practicality. In addition, it introduces HR-VISPR, a publicly available human-centric dataset with biometric, soft-biometric, and non-biometric labels to train an interpretable privacy metric. We evaluate 11 privacy protection methods, ranging from conventional techniques to advanced deep-learning methods, through the proposed framework. The framework differentiates privacy levels in alignment with human visual perception, while highlighting trade-offs between privacy, utility, and practicality. This study, along with the HR-VISPR dataset, serves as an insightful tool for selecting privacy-protection methods and offers a structured evaluation framework applicable across diverse contexts
Creating Symphonies:Coordinating Capacity Limitation and Redispatch Instruments in Power Distribution Systems
No full textClimate change concerns and the limited supply of fossil fuels are driving the shift towards a more sustainable energy system. The rapid growth of renewables and the electrification of demand put increasing strain on power distribution systems. As a consequence, distribution system operators (DSOs) are struggling to prevent network overloading, also known as congestion. In many countries, congestion already has several major consequences, such as slowing down the energy transition, hampering economic development, and threatening the affordability of electricity. Traditionally, DSOs mitigated congestion only through grid reinforcements. Today, the increasing controllability of devices and improved network observability enable flexibility-based solutions as a complementary measure. By changing the power import/export of electrical devices, peak flows in the distribution system can be reduced, thereby better utilizing the existing network capacity. To achieve this in practice, DSOs can apply various contractual, financial, or technical measures, called congestion management instruments. Several pioneering countries have been implementing such instruments over the last decade, and while there are differences, a representative setting emerges in e.g. Germany, the United Kingdom, and the Netherlands. In this setting, a DSO can use a capacity limitation instrument before the majority of electricity is bought and sold (day-ahead), and a redispatch instrument afterwards (intraday). With a capacity limitation instrument, DSOs can temporarily reduce the connection capacity of a network user below their physical connection capacity. This type of instrument can help DSOs to prevent electricity market outcomes that could cause congestion. With redispatch, the DSO compensates flexibility service providers(FSPs) for the amount of energy they deviate from a baseline. This instrument can be applied to correct market outcomes resulting in congestion. The implementation of capacity limitation and redispatch marks a shift in the role of DSOs: beyond grid reinforcement, they can now manage congestion with multiple instruments. This results in a new question for DSOs to answer: how to coordinate capacity limitation instruments and redispatch instruments for effective and efficient congestion management? To address this challenge, this thesis first develops new theory and modeling approaches to study different aspects of the coordination problem in detail. These methods are then applied in various simulation studies to obtain practical insights for DSOs. Methodological Contributions In terms of the developed methodology, this thesis presents several key contributions: The first contribution is a general mathematical framework for studying congestion management with multiple instruments. We formalize congestion management with capacity limitation instruments and redispatch instruments as a stochastic dynamic game between the DSO, FSPs, and network users. This framework extends existing academic approaches to multi-instrument congestion management and functions as the mathematical foundation for all modeling in this thesis. Beyond this thesis, it offers other researchers a structured approach to present or understand modeling work on the topic. The second contribution is a new simulation tool for time-discrete distributed agent-based simulations in energy systems. DOTS-energy is a Python-based simulation framework designed for scalability and for providing full control over computational resources. The tool can be used to quickly explore the interactions between congestion management instruments in a wide variety of conditions and distribution networks. This third contribution is a new perspective on congestion management with multiple instruments under uncertainty. During operations, DSOs face various types of uncertainty. We propose that, in the face of this uncertainty, the risks associated with a congestion management instrument can be characterized by a risk profile. We then formulate congestion management with multiple instruments as a multi-stage risk-aware resource allocation problem, where coordinating instruments is viewed as balancing the risk profiles of the respective instruments .We apply this approach to the coordination of capacity limitation and redispatch instruments. The fourth contribution is a new defender-attacker-defender model to study how capacity limitation instruments can be applied to mitigate increase-decrease gaming in redispatch markets. A key concern with redispatch is that FSPs may manipulate their redispatch baselines to increase future redispatch income. Such strategic behavior can worsen congestion and raise redispatch costs for DSOs. While the literature has suggested that combining redispatch with capacity limitation could reduce this risk, its potential has never been quantified before. This thesis provides the first tools to do so, formulating the interaction between the DSO and a FSP as a defender-attacker-defender game. In addition, we develop solution methods for the resulting trilevel optimization problem, enabling DSOs to determine how best to apply capacity limitation instruments for this purpose. Practical Insights Applying these methods in simulation studies results in several key insights related to the main research question of this thesis. DSOs can apply these insights to improve congestion management strategies involving capacity limitation and redispatch instruments. First, we identify that applying capacity limitation and redispatch instruments is not always the most effective and efficient solution in the first place. Since these instruments can result in operational costs for DSOs, they should be applied to incidental, rather than structural congestion. In low-voltage networks, a well designed network tariff can often already reduce the need for active congestion management when the available flexibility is limited. In such cases, applying the capacity limitation and redispatch instruments does not warrant the additional effort of DSOs. Second, we show how various factors such as flexible asset types, instrument parameters, available flexibility, and market conditions impact the trade-off between capacity limitation and redispath instruments. We especially find that the impact of electricity market dynamics is considerable. This is not only because market conditions are expected to have an increasing influence on power flows in distribution networks, but also because procuring redispatch through redispatch markets exposes DSOs to market risks when applying congestion management. At the same time, FSPs will be active on multiple electricity markets, thereby shaping their willingness to participate in congestion management in general. Third, we find that, when applied correctly, capacity limitation and redispatch instruments can serve as complementary tools, hedging the operational risks inherent to either instrument type. Capacity limitation instruments typically do not involve market risks, but have to be applied earlier when forecasts of congestion are less accurate. Redispatch can be used closer to real time delivery, reducing the risk of over- or underprocuring flexibility. However, redispatch procured through redispatch markets brings both price and liquidity risks for DSOs. When timing redispatch, this results in a trade-off. From the DSO’s perspective, redispatch should be applied as late as possible to maximize the accuracy of congestion forecasts. However, applying it a few hours earlier reduces the risks of not being able to procure enough flexibility in the redispatch markets due to short trading windows. Fourth, we find that capacity limitation instruments, like alternative connection agreements, can be used to partially mitigate the negative consequences of increase-decrease gaming associated with redispatch instruments. We observe in a case study that even if the DSO does not anticipate the strategic gaming behavior of the FSP, significant redispatch cost reductions can be achieved when the capacity limitation is applied at times of low expected day-ahead market prices. The resulting impact on the revenues of the FSP does not need to be large and can even be positive, as the DSO can prevent the FSP from applying the risky gaming strategy at the wrong time
Dual-Polarized Dielectric-filled Cavity Antenna with Air-Gap-Free Metasurface Loading for LTCC-based 5G-and-Beyond Antenna-in-Package Phased Arrays
No full textThis work introduces a dual-polarized (DP) slot-excited dielectric-filled cavity antenna phased array designed for 5G and beyond AiP solutions using low-temperature co-fired ceramic (LTCC) technology. The primary challenge is to maximize a two-dimensional (2-D) beamsteering range with a compact form factor in a DP operation. Most reported LTCC AiP arrays have ≤ 55◦ beamsteering range at 5G mmWave frequencies and lack experimental validation for DP 2-D beamsteering operation. In contrast, the proposed antenna element in the infinite array enables ±60◦ 2-D beamsteering in the 5G n261 (27.5-28.35 GHz) band with the form factor as small as 5mm×5mm×1.3mm. This performance is achieved by introducing a wide-angle impedance matching (WAIM) structure directly at the antenna aperture, comprising a dual-layer metasurface formed by periodically arranged electrically small patches in a staggered configuration. The antenna is jointly optimized with a DP feeding network to minimize parasitics in the multilayer LTCC stack and a co-integrated printed circuit board (PCB) carrier to mitigate package-carrier coupling often neglected in previous works. It is fabricated in a single LTCC process without air gaps to ensure precise layer-to-layer alignment and eliminate losses from additional support structures inherent to conventional air-gap WAIM stacks. A custom equivalent-circuit model enables efficient analysis of scan-dependent impedance matching for the optimization. A 128-element AiP prototype employs 4×4 subarrays of four DP elements on the PCB carrier and is packaged with ball grid arrays (BGAs) to facilitate integration. This modular approach allows scaling to large arrays while considering the critical effects of the package stack-up and DP feeding. The simulated active reflection coefficient is < −8.5 dB within ±60◦ (with a scan loss marginally above 3 dB) at 27.5-29.5 GHz. Measurements confirm simulations for all relevant performance metrics in ±60◦ 2D-range.</p
Loss-Optimal Modulation for an MMC-Based Triple-Active-Bridge Topology
No full textThis paper presents a loss-optimal modulation strategy for an MMC-based triple-active-bridge (TAB) converter that guarantees zero-voltage switching (ZVS). A switching pattern that blends penta phase-shift (PPS) and pulse-amplitude control (PAC) is proposed, and a generalized harmonic approximation (GHA)-based optimization is formulated to minimize total losses using device-calibrated switching and conduction loss models under power-balance and ZVS constraints. The optimization is solved offline over a grid spanning the operating space (power setpoints and dc-link voltages), and the resulting angles are stored in look-up tables for online deployment. Simulations demonstrate up to 44 % lower total losses than Phase-Shift Modulation and PPS, as well as lower conduction losses than PAC. In addition, the method expands the feasible operating region of the converter to power points that are not supported by PAC