The development of the synchrosqueezing techniques has been a very active research area in recent years. The synchrosqueezing transform (SST) is a powerful tool for time-frequency analysis of multicomponent nonstationary signals. The local maximum synchrosqueezing transform (LMSST) was recently introduced as a variant of SST. It directly uses the local maximum to reassign the time-frequency coefficient in order to obtain an energy-concentrated representation. However the theoretical foundation of LMSST is yet to be properly established. In this paper, we study the linear-chirp model based LMSST with the objective of obtaining its error bounds of mode separation/retrieval under noise-free and noisy conditions. Our theorems demonstrate the applicability of LMSST mathematically. Numerical experiments including synthetic and real-world signals are provided to illustrate the general theory.

Constant common mode voltage (CCMV) is critical in Solar Photovoltaic (SPV) systems. Maintaining CCMV further eliminates the transitions in the voltage across the parasitic capacitance of the PV panel, making it a low-frequency output voltage waveform. This further reduces the leakage and common mode current in the SPV systems. This paper presents a five-level switched capacitor PV inverter configuration that maintains the CCMV with low-frequency terminal voltage while considering the effect of switch device junction capacitance (SDJC). Thus, the proposed configuration eliminates the transitions in the common mode and terminal voltage due to the SDJC. The two switched capacitors employed in the proposed configuration are charged and discharged within each switching period, supporting the self-charge balance feature apart from low voltage and RMS current rating, reducing its size and value. Further, the proposed system effectively utilizes the PV source as switch capacitors, and the dc-bus buffer capacitor across the PV source comes in parallel in each switching period. The proposed configuration can also operate at lower modulation indices generating three levels with CCMV. A new mathematical analysis considering SDJC is given for the derivation of expression for terminal voltage and CMV, which is further verified using simulation and experimental results

On-body Sensor-based Human Activity Recognition provides an excellent opportunity to monitor a person’s movement unobtrusively. This is commonly used to provide recommendations about physical activity and similar, while our main motivation was to streamline the time-consuming process of documenting nursing activities. In this paper we lay out the methodology for building activity recognition models, and present a mobile application that provides comprehensive support for it: from data acquisition and annotation, to on-device training and execution of activity recognition in real time. Furthermore, we developed a novel approach to determine the optimal sensor placement to achieve higher classification accuracy through real-time video data conversion into pose estimations. We demonstrate our systems’ various features in a feasibility study. For this purpose, we conducted a study with inertial sensors on nursing activities with ten subjects on 13 activities. We demonstrate that our vision-based approach for the optimal sensor placement achieved a high correspondence with a trained model. Furthermore, we achieved a higher accuracy for a multimodal and on-device trained model approach compared to a single-modality and offline variant. This evaluation showed that the system served essential support in a realistic environment.

This paper proposes a DC-Decoupled based multi-phase inverter (DCDMI) topology for electronic-pole changing (E-PC) induction motor drive (IMD) system. The given E-PC IMD with flexible wide speed and torque ranges, operates in two modes: 3-φ,12-pole and 9-φ, 4-pole mode. However, during the operation of E-PC IMD, a prevalent torque ripple exists, especially in high-pole mode. Torque ripples are due to harmonics in the stator periphery. The harmonics are attributed to common-mode current (CMC) produced by common-mode voltage (CMV) of the inverter that drives IMD. Consequently, this results in deterioration of drive performance. This paper proposes a new inverter topology for E-PC IMD, which minimizes the transitions in CMV both operating modes. Subsequently, the CMC in the stator periphery is minimized in both the modes. The required analysis and derivation for the expression of terminal voltages are also presented in this manuscript. Further, in the proposed system an interleaved switching strategy which is integrated with the DC-decoupling technique improves the over-all performance in 3-φ mode. To validate the results of the proposed DCDMI topology, a 36-slot 3.0HP 3-φ, 12-pole IM is designed, simulated, and analyzed in an Ansys environment. Simulated results are corroborated by experimentation.

There is an increasing need for flexibility in power systems worldwide, giving rise to European policy documents outlining how Distribution System Operators (DSOs) should procure flexibility and include flexibility in planning and operation of their electricity distribution grids. This implies a remarkable change from today’s situation where DSOs rely on investments in grid assets that they have full control of, to a new regime where DSOs should rely on flexibility provided by third parties. The objective of this work has been to gain a better understanding of the feasibility of and barriers to wide-spread utilization of flexibility in planning and operation of electricity distribution grids. Building upon a previous literature review and a taxonomy for classifying and characterizing power system flexibility, we propose frameworks for i) classifying flexibility resources and flexibility enablers in grid operation and planning, and ii) classifying and understanding barriers to utilizing them in terms of a flexibility value chain. These theoretical frameworks are tested against empirical data collected in semi-structured in-depth interviews with a representative selection of Norwegian DSOs. Mapping the findings to the frameworks gives a systematic overview of the flexibility situation in Norway and presents both country-specific and general insights about barriers to utilization of flexibility.

In 6G networks, unmanned aerial vehicles (UAVs) can serve as aerial flying base stations (AFBS) with aerial mobile edge computing (AMEC) server capabilities. AFBS is an increasingly popular solution for delivering time-sensitive applications, extending network coverage, and assisting ground base stations in the healthcare systems for remote areas with limited infrastructure. Furthermore, the UAVs are deployed in the healthcare system to support the Internet of medical things (IoMT) devices in data collection, medical equipment distribution, and providing smart services. However, ensuring the privacy and security of patients’ data with the limited UAV resources is a major challenge. In this paper, we present a federated deep reinforcement learning framework for resource allocation in UAV-enabled healthcare systems, where IoMT devices send their trained model parameters without transmitting sensitive raw data to the AMEC server. In the proposed framework, the IoMT device is associated with AFBS  based on the quality of the data and its demand in order to maximize learning efficiency and accuracy. This work aims to minimize the computation costs of the IoMT devices while considering UAV resources and the fairness of UAV coverage. Simulation results prove that our proposed algorithm outperforms other baseline algorithms in learning accuracy and computational cost.

As a result of the introduction of new infectious illnesses, key infection prevention measures were implemented. Now, a new coronavirus (SARS-CoV-2) epidemic has expanded swiftly, causing the coronavirus illness 2019 (COVID-19). Many microorganisms spread illness via hospital surfaces due to environmental pollution. This virus has been associated to close contact between persons in tight situations such as houses, hospitals, assisted living, and residential institutions. Aside from health care settings, public buildings, faith-based community centers, marketplaces, transportation, and corporate environments are prone to COVID-19 transmission. Physical contact with the sanitizer device may cause for spread Covid virus. That’s why we have proposed an automatic fogger mechanism-based hand sanitizer that may be able to reduce covid risk. Disinfectant fog will flow when an object will pass through the machine. This project will save cost, time, and wastage along with Covid spreading risk. This project is about designing a good healthcare system. In recent years, sophisticated automation has influenced the health industry. Health care in poor nations is costly. So, the project is an attempt to tackle this issue.

Portland General Electric (PGE), an investor-owned electric utility serving nearly 900,000 customers in 51 Oregon cities, joined Western Energy Imbalance Market (EIM) – a California Independent System Operator (CAISO) platform for sharing cleaner and more efficient generation resources among multiple states in the Western U.S. While participation in EIM brings a new set of economic opportunities for PGE by capturing flexibility needs in the market, it also changes operational patterns (e.g., more frequent start/stop, off-nominal power output, and ramping) of PGE’s conventional generation fleet, including hydropower units. With 492 MW of hydropower capacity, which constitutes 15% of PGE’s generation mix, it is important for PGE (as any other EIM-participant utility) to understand how participation in EIM impacts its hydropower assets, and how a techno-economically sustainable operation of those assets could be accomplished. In this paper, a case study is performed with PGE’s hydropower generation facilities to assess and quantify EIM?participation driven changes in hydropower operational patterns

Many emerging sensor network applications operate in challenging environments wherein the base station is not available. Data generated from such intermittently connected sensor networks (ICSNs) therefore must be stored inside the network for some unpredictable period of time before uploading opportunities become available. Consequently, sensory data could overflow the limited storage capacity available in the entire network, making discarding valuable data inevitable. To overcome such overall storage overflow in ICSNs, we propose and study a new algorithmic framework called data aggregation for overall storage overflow (DAO2). Utilizing spatial data correlation that commonly exists among sensory data, DAO2 employs data aggregation techniques to reduce the overflow data size while minimizing the total energy consumption in data aggregation. At the core of our framework are two new graph theoretical problems that have not been studied. We refer to them as traveling salesmen placement problem (TSP2) and quota traveling salesmen placement problem (Q-TSP2). Different from the well-known multiple traveling salesman problem (mTSP) and its variants, which mainly focus on the routing of multiple salesmen initially located at fixed locations, TSP2 and Q-TSP2 must decide the placement as well as the routing of the traveling salesmen. We prove that both problems are NP-hard and design approximation, heuristic, and distributed algorithms. Our algorithms outperform the state-of-the-art data aggregation work with base stations by up to 71.8% in terms of energy consumption. The building blocks of our framework are two novel graph structures called aggregation network and minimum q-edge forest that accurately capture the information needed for energy-efficient data aggregation in ICSNs.

Explainable Artificial Intelligence (XAI) is an increasingly important field of research required to bring AI to the next level in real-world applications. Global sensitivity analysis methods play an important role in XAI, as they can provide an understanding of which (groups of) parameters have high influence in the predictions of machine learning models and the output of simulators and real-world processes. In this paper, we conduct a survey into global sensitivity methods in an XAI context and present both a qualitative and a quantitative analysis of these methods under different conditions. In addition to the overview and comparison, we propose an open source application, GSAreport, that allows you to easily generate extensive reports using a carefully selected set of global sensitivity analysis methods depending on the number of dimensions and samples, to gain a deep understanding of the role of each feature for a given model or data set. We finally present the methods discussed in a complex real-world application of genomic prediction and draw conclusions about when to use which GSA methods.

The dataset has been done at Gilan University of Medical Sciences. It included 210 clinical images of oral mucosa, 154 images of which were related to the three studied lesions: 50 cases of erosive lichen planus and ulcerative, 53 cases of leukoplakia, 51 cases of oral SCC, and 56 images were related to healthy mucosa. The dataset contained 30 features.

This study was acquited on the mining area located under the Saranda area of West Singhbhum district in Jharkhand and part of the state of Odisha. Here, GST data is used at the 30 m resolution of Hyperion, a hyperspectral remote sensing satellite. In this study, different types of image processing exemplar are used to empower and extract the Hematite and Manganese ore mining class. The present method was created to improve the mining class with a spectral signature associated with the features class. In the present study, an index, that is, a mineral index created to distinguish the mineral ore as seen on Earth in open mining areas. A decree analysis is to clarify the eubstance of this index.In this way, the multifarious process & generated mineral-index applied to wedded frequency, strength to enhance the mineral class & comparative evaluation done over the data sets.The diverse supervised classification has been done to identify the mineral class under the GUA Mining range of Hyperion data. Also, the NDVI used to excerption the open cast mining class after the virtual removal of vegetation class on the scale of (1-NDVI).Further THOR, ICA, ANN used here to single out the minerals, & also by this way the new index probed.

While   global   navigation   satellite   system   (GNSS)technologies have always been the go-to solution for localizationproblems, they may not be the best choice for some Internet-of-Things (IoT) applications due to the incurred power consumptionand cost. In this paper, we present an alternative satellite-basedlocalization method exploiting the signature of Doppler shifts andangle  of  arrival  measurements  as  seen  by  a  receiving  satellite.We first derive the joint likelihood function of the measurements,which  is  represented  as  a  combination  of  three  Gaussian  distri-butions.  Then,  we  show  that  the  maximum  likelihood  problemreduces  to  a  more-efficient  mean  squared  error  minimizationin  the  Gaussian  case  as  inferred  from  real  measurements  wecollected  from  low-earth-orbit  satellite  using  a  tracking  groundstation.  Thus,  we  propose  utilizing  a  stochastic  optimizer  tosearch for the global minimum of the mean squared error whichrepresents the location of the ground IoT device as estimated bythe  satellite  platform.  The  emulated  results  show  that  the  IoTdevice localization, under such realistic model, can be performedwith  sufficient  accuracy  for  IoT  applications.

This paper proposes a fast-processing algorithm for finding endmembers in hyperspectral image data and it’s FPGA architecture. The proposed algorithm named as Improved Real-Time Fast Simplex Growing Algorithm (IRT-FSGA) removes redundant computation for the pixels which do not qualify as endmember replacements and thereby achieves a significant reduction in computation time over the RT-FSGA algorithm. The improvement obtained is without compromising on the accuracy of the endmembers found. FPGA implementation results of the designed architecture show that the proposed IRT-FSGA achieves 33× improvement in performance at the cost of only a small area overhead. Thus, it is potentially useful in real-time applications which have stringent throughput and power requirements.

Aging of infrastructures has been a worldwide issue, and cost saving by shifting to preventive maintenance is urgent. Especially, damage detection of concrete bridge decks is one of the most important subjects, because of the significant repair costs due to its complicated structure. Rebars are the fundamental components of bridge decks, and they often become the trigger of bridge decks’ damages. Previous researches have been focused on detecting the locations of rebars in cross section images acquired by single-channel ground penetrating radar (GPR), however, no research has reproduced 3D rebar mesh arrangement from radar volume images acquired by multi-channel GPR.This paper proposes a method that reproduces 3D rebar mesh that contains the data of vertex location and reflection time from radar volume images. Real scale bridge deck specimens were created in this study and reflections of electromagnetic waves were observed utilizing an on-vehicle GPR. In the proposed method, 3D filtering based on the 3D DFT theory for the noise reduction was applied. Also, automation in detection of the rebar depth was achieved focusing on edges of the images. As a result, 3D rebar meshes were successfully reproduced with appropriate distribution rebar spacing. Also, the method was applied to radar data acquired from a bridge in service. The proposed method effectively functioned for in-service bridge data, and automatically reproduced the mesh of 21m length.

A completely automatic and accurate detection algorithm for the delamination on tunnel concrete lining surfaces from laser 3D point cloud data is proposed. A Mobile Mapping System (MMS), which mounts laser sensors and a positioning system, is utilized to measure the geometry of tunnel lining surfaces highspeed. The proposed algorithm consists of 4 steps: removal of tunnel profile components, detection of peaks of anomalies, localization of anomaly areas, classification of delamination and appendages. On tunnel linings, there are many appendages such as cables, lights, signs, and water guides to mask the features of delamination. In the article, a novel 3D feature was introduced to realize the accurate classification. An automatic SVM algorithm was developed using real tunnel lining data and manual inspection results, showing an accurate delamination map.

Job-shop scheduling is an important but difficult combinatorial optimization problem for low-volume and high-variety manufacturing, with solutions required to be obtained quickly at the beginning of each shift. In view of the increasing demand for customized products, problem sizes are growing. A promising direction is to take advantage of Machine Learning (ML). Direct learning to predict solutions for job-shop scheduling, however, suffers from major difficulties when problem scales are large. In this paper, a Deep Neural Network (DNN) is synergistically integrated within the decomposition and coordination framework of Surrogate Lagrangian Relaxation (SLR) to predict good-enough solutions for subproblems. Since a subproblem is associated with a single part, learning difficulties caused by large scales are overcome. Nevertheless, the learning still presents challenges. Because of the high-variety nature of parts, the DNN is desired to be able to generalize to solve all possible parts. To this end, our idea is to establish “surrogate” part subproblems that are easier to learn, develop a DNN based on Pointer Network to learn to predict their solutions and calculate the solutions of the original part subproblems based on these predictions. Moreover, a masking mechanism is developed such that all the predictions are feasible. Numerical results demonstrate that good-enough subproblem solutions are predicted in many iterations, and high-quality solutions of the overall problem are obtained in a computationally efficient manner. The performance of the method is further improved through continuous learning.

Seated postural limit defines the region from beyond which a subject cannot return the trunk to the neutral position without additional external support. The seated postural limits can be used as a reference to provide assistive support to the torso by an appropriately designed robotics, e.g., Trunk Support Trainer (TruST). However, fixed boundary representations of seated postural limits in the literature cannot capture the dynamically changing seated postural limits during training. In this study, we propose a conceptual model of the dynamic boundary of the trunk center using a boundary vector designed to track the postural-goal direction and trunk movement amplitude during a sitting task. We experimented with 20 healthy subjects. The results support our hypothesis that TruST intervention with an assist-as-needed controller based on the proposed dynamic boundary representation could achieve more significant sitting workspace area improvements than a fixed boundary representation. The second contribution of this paper is that we provide an approach to effectively introduce deep learning into TruST’s real-time controller design. We compiled a 3D trunk movement dataset which is currently the largest in the literature. We designed a loss capable of solving the gate-controlled regression problem. We proposed a novel roadmap for the exploration study. Following the roadmap, we developed a deep learning architecture, modified the widely used Inception module, and then obtained a deep learning model capable of accurately predicting the dynamic boundary in real-time. We believe this approach can be extended across other rehabilitation robots towards designing intelligent dynamic boundary-based assist-as-needed controllers.