Power systems are undergoing rapid transitions to incorporate renewable sources of generation and to combat climate change. Next stage of transitions will lead to a shift from large-scale, centralized systems to networks of smallsized, distributed electricity systems which require distributed or decentralized ledgers for database management for efficient transactions. Distributed Ledger Technology (DLT) are a form of decentralized ledgers where the transactions (energy, information and money) among various entities are maintained. One such DLT is blockchain technology which offers several advantages. Data recorded in blockchains are difficult to tamper with; have privacy protection; facilitate fast, accurate and real-time settlement of financial transactions. Contemporary research has started focusing on their possible applications in energy systems. State-of-the-art suggests that while business and market aspects have been extensively discussed, the electrical constraints and implementation methodologies have not been adequately addressed. Furthermore, all the reviewed projects have implemented only peer-to-peer transactions that are not scalable. To incorporate the new entities like prosumers, intermicrogrid transactions and interactions with the legacy power grid, new structural and operational frameworks are necessary. The proposed research explores the possibility of developing blockchain enabled smart microgrids (BSMG) with the above frameworks. It aims to build a conceptual framework of BSMG, including the transaction protocols and process flows. It proposes inclusion of network constraints in a three-levelled transaction setup which is executed over a four-layered architecture. Another practical challenge is that BSMGs may be set up on different blockchain platforms. Hence, this paper also proposes implementing Inter-Blockchain Protocol for the first time to include interoperability and communication between different platforms. Finally the performance metrics that will be used to validate the BSMGs are outlined.

Surrogate-assisted evolutionary algorithms (SAEAs) have attracted considerable attention for reducing the computation time required by an EA on computationally expensive optimization problems. In such algorithms, a surrogate model estimates the solution evaluation with a low computing cost and is used to obtain promising solutions to which the accurate evaluation with an expensive computation cost is then applied. This study proposes a novel pairwise ranking surrogate model called the Extreme Learning-machine-based DirectRanker (ELDR). ELDR integrates two machine learning models: extreme learning machine (ELM) and DirectRanker (DR). ELM is a single-layer neural network capable of fast learning, whereas DR uses pairwise learning to rank using a neural network developed mainly for information retrieval. To investigate the effectiveness of the proposed surrogate model, this study first examined the estimation accuracy of ELDR. Subsequently, ELDR was incorporated into a state-of-the-art SAEA and compared with existing SAEAs on well-known real-valued optimization benchmark problems. The experimental results revealed that ELDR has a high estimation accuracy even on high-dimensional problems with a small amount of training data. In addition, the SAEA using ELDR exhibited the highest search performance compared with other existing SAEAs, especially on high-dimensional problems.

Robotics is widely seen as a key enabling technology for the society of tomorrow. This work examines the role of robotics and intelligent medical devices in Intensive Care Medicine. Demographics predict that more elderly patients will need to be treated with fewer healthcare personnel, calling for innovations in ICU management.  Robotics is a key enabling technology for the 21st century and may help to smooth the foreseeable workload/manpower disparity in medicine. Studying the application of robotics in the ICU in a manner beneficial for patients and accepted by intensive care team is desirable. Financial sustainability is essential for the introduction of these new technologies to the ICU. This study therefore assesses the state of the art in robotics in intensive care medicine and identifies opportunities for progress, using an observational approach in a teaching hospital ICU in combination with an in-depth review of the literature and a survey of the market. Tasks amenable to robotics are identified, their acceptability to patients and caregivers are examined and their quantitative contribution to future management of an intensive care unit is assessed.

Code smells are code structures that harm the software’s quality. An obstacle to developing automatic detectors is the available datasets’ limitations. Furthermore, researchers developed many solutions for Java while neglecting other programming languages. Recently, we created the code smell dataset for C# by following an annotation procedure inspired by the established annotation practices in Natural Language Processing. This paper evaluates Machine Learning (ML) code smell detection approaches on our novel dataset. We consider two feature representations to train ML models: (1) code metrics and (2) CodeT5 embeddings. This study is the first to consider the CodeT5 state-of-the-art neural source code embedding for code smell detection in C#. To prove the effectiveness of ML, we consider multiple metrics-based heuristics as alternatives. In our experiments, the best-performing approach was the ML classifier trained on code metrics (F-measure of 0.87 for Long Method and 0.91 for Large Class detection). However, the performance improvement over CodeT5 features is negligible if we consider the advantages of automatically inferring features. We showed that our model exceeds human performance and could be helpful to developers. To the best of our knowledge, this is the first study to compare the performance of automatic smell detectors against human performance.

This paper proposes synchronous grant-free non-orthogonal multiple access (GF-NOMA) frameworks that effectively integrate UE clustering and low-complexity power control to facilitate power reception disparity required by the power domain NOMA. While single-level GF-NOMA (SGF-NOMA) designates an identical transmit power for all user equipments (UEs),  multi-level GF-NOMA (MGF-NOMA) groups UEs into partitions based on the sounding reference signals strength and assign partitions with different identical power levels. Based on the objective of interest (e.g., max-sum or max-min rate), the proposed UE clustering scheme iteratively admits UEs to form clusters, whose size is dynamically determined based on the number of UEs and available resource blocks (RBs). Once the UEs are acknowledged with power levels and allocated RBs through random access response (RAR) messages, UEs can transmit anytime without grant acquisition. Numerical results show that the proposed GF-NOMA frameworks can compute clusters in the order of milliseconds for hundreds of UEs. The MGF-NOMA can reach up to 96-99% of the optimal benchmark max-sum rate, the SGF-NOMA reaches 87% of the optimal benchmark max-sum rate at the same power consumption. Since the MGF-NOMA and optimal benchmark enforces the strongest and weakest channel UEs to transmit at maximum and minimum transmit powers, respectively, the SGF-NOMA also offers a significantly higher energy consumption fairness and network lifetime as all UEs consume equal transmit powers. While the MGF-NOMA delivers an inferior max-min rate performance, the SGF-NOMA is shown to reach 3x106 MbpJ energy efficiency compared to 1x107 MbpJ

Non-homogeneous, bi-directional glass fabric reinforced thermoplastic (GFRT) composites are used in the automobile industry to produce economic, environmentally friendly and recyclable vehicles. Despite the fact that GFRT materials have superior mechanical characteristics and lower manufacturing cost compared with traditional materials, they are prone to significant deformation and defect emergence during the forming/manufacturing of parts from prefabricated laminates. Some defects, especially hidden ones like delamination and consolidation defects, can significantly reduce the mechanical properties of the final parts. It is essential to use 3D non-destructive testing (NDT) technology to detect and identify such hidden defects. This research investigates the ability of 3D THz imaging to detect hidden defects (i.e., delamination and consolidation) in GFRT composite materials. Frequency modulated continuous wave (FMCW) technique is used to generate 3D-THz cross-sectional images. Non-defective and defective samples with a thickness of 1.5 mm have been measured with two 3D THz imaging systems operating in the frequency range from 238 to 316 GHz and from 499 to 733 GHz. In addition, a simulation of the reflected FMCW spectrum is presented to analyze the measurements and intercompare the performance of the two imaging systems. A systematic analysis in a large sample set demonstrates that a delamination defect with a thickness of 125 µm as well as a consolidation defect with a difference in thickness less than 0.4 mm can be successfully detected by the 700-GHz imaging system.

We investigate the re-entrant undercut profile resulting from Au wet etching for patterning micron range thick films using an aqua regia-based solution in comparison with an iodine-iodide-based commercial etchant. Our work discriminates between two undercutting mechanisms: galvanic acceleration of etch rate at the Au adhesion or barrier layer, and delamination-based undercutting. We tracked etch outcomes of feature size reduction from photoresist size, undercut Au in cross-section and lift-off of small (5-10 μm) features. Results indicate that galvanic undercutting is well-mitigated by the aqua regia solution compared to commercial etchant results. Good Au adhesion eliminates undercut for 500 nm-thick Au and mitigates it by ~80% for 1 μm-thick Au. We discuss the electrochemical origin of this mitigated galvanic undercut.

Manuscript submitted to Transactions on Intelligent Transportation Systems for review.

In many emerging Internet of Things (IoT) applications, the freshness of the data collected by the sensor nodes is an important design criterion. Age of Information (AoI) quantifies the freshness of the received information or status update. This work considers a setup of deployed IoT devices in an IoT network; multiple unmanned aerial vehicles (UAVs) serve as mobile relay nodes between the sensors and the base station. We formulate an optimization problem to jointly plan the UAVs’ trajectory, while minimizing the AoI of the received messages and the devices’ energy consumption. The solution accounts for the UAVs’ battery lifetime and flight time to recharging depots to ensure the UAVs’ green operation. The complex optimization problem is efficiently solved using a deep reinforcement learning algorithm. In particular, we propose a deep Q-network, which works as a function approximation to estimate the state-action value function. The proposed scheme is quick to converge and results in a lower ergodic age and ergodic energy consumption when compared with benchmark algorithms such as greedy algorithm (GA), nearest neighbour (NN), and random-walk (RW).

Objective: The need for distilling the hemodynamic complexity into clinically relevant quantities has imposed a reductionist approach to investigate aortic flows, resulting in a lossy compression of the information hidden in 4D aortic fluid structures. Aiming at reducing information loss, this study proposes a network-based approach to identify and characterize in vivo the large-scale coherent motion of blood in the healthy human aorta using 4D Flow MRI. Methods: Adopting the quantitative paradigm of the aortic flow as a “social network”, 4D flow MRI acquisitions were performed on forty-one healthy volunteers. Correlations between the aortic blood flow rate waveform at the proximal ascending aorta (AAo), assumed as a main determinant of the aortic flow, and the waveforms of the axial velocity (aligned with the aortic centerline) in the whole aorta were used to build “one-to-all” networks. The impact of the driving flow rate waveform and of the main aortic geometric attributes on the transport of large-scale coherent fluid structures was investigated. Results: The anatomical length of persistence of large-scale coherent motion was the 29.6% of the healthy thoracic aorta length (median value, IQR 23.1%-33.9%). Moreover, it was positively correlated with the average and peak-to-peak AAo blood flow rate values, suggesting a remarkable inertial effect of the AAo flow rate on aortic hemodynamic coherence. No association emerged between the anatomical length of persistence and aortic geometric attributes. Conclusion: The here proposed in vivo approach allowed to quantitatively characterize the large-scale aortic blood motion, strengthening the definition of coherent hemodynamic structures. Significance: The findings on healthy aortas may be used as reference values to investigate the impact of aortic disease or devices implantation in disrupting/restoring the physiological spatiotemporal coherence of large-scale aortic flow.

In this communication, a wideband reconfigurable reflectarray based on reflector-backed active second-order bandpass frequency selective surface (FSS) is presented. The reflector is composed of periodic short-circuited parallel plate waveguide (PPW) and the FSS is composed of stacked non-resonant metallic elements separated by thin dielectric substrates. By integrating microwave varactors in the capacitive layers of FSS, more than 270 o continuous phase tunability is achieved within a factional bandwidth of 14%. A one-dimensional reflectarray prototype operating at C band is fabricated and measured. The experimental results show that it can achieve ±55 o beam scanning coverage. Symmetric beam steering is observed due to the center-fed configuration. With advantages of low cost and simple structure, the proposed reflectarray can be potentially used in wideband wireless communication and radar systems.

Objective: In the last two decades, there has been a growing interest in exploring surgical procedures with statistical models to analyze operations at different semantic levels. This information is necessary for developing context-aware intelligent systems, which can assist the physicians during operations, evaluate procedures afterward or help the management team to effectively utilize the operating room. The objective is to extract reliable patterns from surgical data for the robust estimation of surgical activities performed during operations. The purpose of this paper is to review the state-of-the-art deep learning methods that have been published after 2018 for analyzing surgical workflows, with a focus on phase and step recognition. Methods: Three databases, IEEE Xplore, Scopus, and PubMed were searched, and additional studies are added through a manual search. After the database search, 343 studies were screened and a total of 44 studies are selected for this review. Conclusion: The use of temporal information is essential for identifying the next surgical action. Contemporary methods used mainly RNNs, hierarchical CNNs, and Transformers to preserve long-distance temporal relations. The lack of large publicly available datasets for various procedures is a great challenge for the development of new and robust models. As supervised learning strategies are used to show proof-of-concept, self-supervised, semi-supervised, or active learning methods are used to mitigate dependency on annotated data. Significance: The present study provides a comprehensive review of recent methods in surgical workflow analysis, summarizes commonly used architectures, datasets, and discusses challenges.

An adaptive switching controller based on dynamic zero moment point for versatile hip exoskeleton is proposed in this work. The linear finite hysteretic state machine is designed to recognize hybrid motion phases. The torque planning strategy based on dynamic zero moment point is deployed to obtain assistant torque adaptively under different locomotion. Experiments are carried out to verify the performance of the controller, confirming the stability and accuracy of the motion phase recognition, which also demonstrates excellent kinematic performance. The net metabolic rate can be reduced by 5.75% while wearing the versatile hip exoskeleton. Compared with existing research, the performance of the proposed controller has significant advantages. The proposed controller is capable of multiple types of locomotion including flat walking, stair climbing, and lifting heavy objects with low complexity and energy consumption. The Biological and Medical Ethics Committee of Beihang University has approved the experiments.

Measuring data utility and privacy risk embedded in synthetic or other de-identified datasets is an increasingly important research area. Existing measures in the data privacy literature however are one-sided in that they either measure utility or privacy risk only. In this paper we propose a new measure that can evaluate both data utility and privacy, a well-known trade-off relationship in data synthesis. The proposed measure employs the notion of relative distance between the synthetic and original datasets at the dataset level, and can identify the optimally balanced position of the synthetic data in terms of both utility and privacy. In addition, we devise a graphical tool that visually reveals the current utility-privacy trade-off position of the synthetic data. Numerical studies show our new measure consistently performs better and offers richer interpretations than other existing global data utility measures, for both simulated and real datasets, confirming its distinctive advantages.

Multimodal Land Cover Classification (MLCC) using the optical and Synthetic Aperture Radar (SAR) modalities has resulted in outstanding performances over using only unimodal data duo to their complementary information on land properties. Previous multimodal deep learning (MDL) methods have relied on handcrafted multi-branch convolutional neural networks (CNN) to extract the features of different modalities and merged them for land cover classification. However, natural images-oriented handcrafted CNN models may not the optimal strategies to handle Remote Sensing (RS) image interpretation problems, duo to the huge difference in terms of imaging angles and imaging ways. Furthermore, few MDL methods have analyzed optimal combinations of hierarchical features from different modalities. In this article, we propose an efficient multimodal architecture search framework, namely Multimodal Semantic Consistency-Based Fusion Architecture Search (M2SC-FAS) in continuous search space with the gradient-based optimization method, which can not only discover optimal optical- and SAR-specific architectures according to the different characteristics of the optical and SAR images, respectively, but also realizes the search of optimal multimodal dense fusion architecture. Specifically, the semantic-consistency constraint is introduced to guarantee dense fusion between hierarchical optical and SAR features with high semantic consistency and then capture the complementary performance on land properties. Finally, the basis of curriculum learning strategy is adopted on the M2SC-FAS. Extensive experiments show superior performances of our work on three broad co-registered optical and SAR datasets.

This paper details the design of a three-phase transmitter-receiver topology for use in a dynamic wireless power transfer (DWPT) system for heavy-duty electric vehicles. The proposed approach and topology eliminate the power oscillation and voltage/current sharing problems that plague planar three-phase DWPT systems. Multi-objective optimization is employed to maximize magnetic coupling and minimize undesired positive–negative sequence coupling, yielding a Pareto-optimal front that delineates the tradeoff between the two objectives for the proposed topology. Selected designs are simulated for a 200-kW/m DWPT system, verifying the elimination of power oscillation and phase imbalance with minimal impact on magnetic performance.

The purpose of this paper is to present in fully quantum terms the fundamentals of quantum Fourier optics in a completely coherent and comprehensive method with several key applications. This paper’s physical and mathematical models for various elements of a typical spatial optical signal processing apparatus create a new paradigm and will be of utmost importance to advance many quantum optical science and engineering fields. The results in this paper will, significantly advance the state of the art of quantum Fourier optics.

In this article, we address the issues of stability and data-efficiency in reinforcement learning (RL). A novel RL approach, Kullback–Leibler divergence-regularized distributional RL (KLC51) is proposed to integrate the advantages of both stability in the distributional RL and data-efficiency in the Kullback-Leibler (KL) divergence-regularized RL in one framework. KLC51 derived the Bellman equation and the TD errors regularized by KL divergence in a distributional perspective and explored the approximated strategies of properly mapping the corresponding Boltzmann softmax term into distributions. Evaluated by several benchmark tasks with different complexity, the proposed method clearly illustrates the positive effect of the KL divergence regularization to the distributional RL including exclusive exploration behaviors and smooth value function update, and successfully demonstrates its significant superiority in both learning stability and data-efficiency compared with the related baseline approaches.