Reconfigurable Processor Research Articles

Realisation of responsive and sustainable reconfigurable manufacturing systems

There is a lack of a design method for the manufacturing system reconfiguration to cope with the changing demand and evolving production technologies while minimising energy consumption. The key drivers for the new industrial paradigm are flexibility and sustainable manufacturing, which have been studied independently in the prior research. The aim of this research is to study two drivers simultaneously by designing robust models and analysing manufacturing system configurations to achieve feasible solutions in any scenario that may arise due to evolving, incomplete, and unforeseen production requirements, while minimising energy usage during product manufacture. To achieve this goal, this research develops a robustly validated pre-emptive decision engineering framework (DEF) for the manufacturing system reconfiguration process to manage future uncertainty of future conditions and identifies current production vulnerabilities and alternative production portfolios. In this research, a robust RMS reconfiguration strategy is designed using a compromise decision support problem (cDSP), and decentralised decision-making designs are explored through the use of game theory. The findings provide a new production system for adaptable, responsive, and sustainable manufacturing processes in the dynamic global economy. These results can empower stakeholders to make timely design decisions that lead to significant cost savings and sustainable manufacturing.

Unraveling segregation behavior of inactive secondary phase driven by ion-competition reaction for perovskite-2D PbI2 heterojunction solar cells

The undesired segregation of inactive secondary phase that randomly distributed across the perovskite film is detrimental to charge-carrier transport dynamics and light stability. However, in-depth studies for understanding the underlying chemical reaction mechanism and the relationship between phase structural configuration and physical properties are still lacking. Herein, the segregation behavior of the inactive secondary phase and the underlying microcosmic mechanism regarding ionic chemical replacement reaction and lattice reconfiguration process are systematically elaborated. A high-quality perovskite-PbI2 heterojunction film is constructed via converting the three-dimensional PbI2 photoactive phase into a two-dimensional inactive phase, alleviating the energetic disorder of carrier dynamics and photochemical dissociation. The resultant perovskite solar cells achieve an efficiency of 24.23% with excellent light-resistance. The dissociation energetics of phase dimensional structure is theoretically analyzed. This work provides new insights into crystal structure design for constructing high-quality perovskite heterojunction devices.

Study on Meta-heuristics techniques for shade dispersion to enhance GMPP of PV array systems under PSCs

Reconfiguring the photovoltaic modules in an array is one option for reducing the influence of partial shadowing circumstances, which shows the shade dispersion property on the entire photovolatic array. This paper proposes a novel meta-heuristic optimization technique named Vommi that resolves the reconfiguration process of shaded modules in a photovoltaic array. Using the proposed objective function, the major goal of this unique optimization technique is to maximize the power extracted from the shaded PV array during shading conditions. The proposed meta-heuristic Vommi optimization algorithm is compared to conventional total cross-tied, and particle swarm optimization algorithm-based configurations in terms of power at the global maximum power point, improved fill factor, power losses, execution ratio, etc. The global maximum power point of the Vommi optimization algorithm configuration delivers such as 990 W, 1283 W, 1181 W, and 1173 W, which are higher than conventional approaches during all four partial shodowing circumstances and smoother power-voltage characteristics..

Simultaneous Network Reconfiguration and Wind Power Plants Participation in Day‐Ahead Electricity Market Considering Uncertainties

Recently, the participation of wind sources in electricity markets has become a severe challenge due to their intermittent nature. Reconfiguration of power systems can effectively reduce the negative effects of uncertainties. So, this article presents a new method for participating in wind power plants and uncertain customers in a day‐ahead electricity market considering the reconfiguration process. This method tries to maximize social welfare through a two‐level optimization problem. To this end, uncertainties are modeled using the empirical cumulative distribution function and the Monte–Carlo method, and a probabilistic analysis of the market is performed. Then, by defining some indices to evaluate the participants’ satisfaction and using the analytic hierarchy process method (AHP), a new objective function is proposed so that its minimization leads to planning the system configuration and market participants to optimize mentioned indices. The proposed methodology also assumes that the participation of uncertain participants in the spot market will eliminate the imbalances caused by uncertainties. The simulations are implemented using real data on an 8‐bus sample network. The results confirm the efficiency of the proposed method in significantly reducing power producers’ and customers’ costs along with increasing total income and profit from the sale of energy.

ADAS: A High Computational Utilization D ynamic Reconfigurable Hardware A ccelerator for S uper Resolution

Super-resolution (SR) based on deep learning has obtained superior performance in image reconstruction. Recently, various algorithm efforts have been committed to improving image reconstruction quality and speed. However, the inference of SR contains huge amounts of computation and data access, leading to low hardware implementation efficiency. For instance, the up-sampling with the deconvolution process requires considerable computation resources. In addition, the sizes of output feature maps of several middle layers are extraordinarily large, which is challenging to optimize, causing serious data access issues. In this work, we present an all-on-chip hardware architecture based on the deconvolution scheme and feature map segmentation strategy, namely ADAS, where all the generated data by the middle layers are buffered on-chip to avoid large data movements between on- and off-chip. In ADAS, we develop a hardware-friendly and efficient deconvolution scheme to accelerate the computation. Also, the dynamic reconfigurable process element (PE) combined with efficient mapping is proposed to enhance PE utilization up to nearly 100% and support multiple scaling factors. Based on our experimental results, ADAS demonstrates real-time image SR and better image reconstruction quality with PSNR (37.15 dB ) and SSIM (0.9587). Compared to baseline and validated with the FPGA platform, ADAS can support scaling factors of 2, 3, and 4, achieving 2.68 ×, 5.02 ×, and 8.28 × speedup.

Study of the performance of fault-tolerant multi-level inverter included in shunt active power filter

<p><span lang="EN-US">Nowadays, the large number of shunt active power filters (SAPF) is installed in many grid networks to eliminate the source currents harmonics and enhance power quality. These filters are installed in different places according to the filtration requirements. The connection between SAPF and grid network has a negative effect during the open-circuit fault of the insulated gate bipolar transistor (IGBT) switch of the SAPF. This paper proposes the application of the new diagnostic method based on the trigonometric circle and mean value variations techniques to the early detection and precise location of the open-circuit fault of the IGBT switches, and the inclusion of the modified reconfigurable inverter topology to allow the perfect continuity of the filter currents, and improve the diagnostic of the open-circuit fault. A single-sided amplitude spectrum technique (SSAS) is applied on the source currents to get the THDi% value. The obtained simulation results prove, the great success of the proposed diagnostic method, the ability of the modified reconfigurable inverter to be adapted to the grid network, the short response time between the diagnosis and the reconfiguration process is about 7 ms which is very sufficient to guarantee the rapid continuity of the shunt active power filter.</span></p>

Reconfiguration of Nematic Disclinations in Plane-Parallel Confinements

We study numerically the reconfiguration process of colliding m=1/2 strength disclinations in an achiral nematic liquid crystal (NLC). A Landau–de Gennes approach in terms of tensor nematic-order parameters is used. Initially, different pairs m1,m2 of parallel wedge disclination lines connecting opposite substrates confining the NLC in a plane-parallel cell of a thickness h are imposed: {1/2,1/2}, {−1/2,−1/2} and {−1/2,1/2}. The collisions are imposed by the relative rotation of the azimuthal angle θ of the substrates that strongly pin the defect end points. Pairs {1/2,1/2} and {−1/2,−1/2} “rewire” at the critical angle θc1=3π4 in all cases studied. On the other hand, two qualitatively different scenarios are observed for {−1/2,1/2}. In the thinner film regime h<hc, the disclinations rewire at θc2=5π4. The rewiring process is mediated by an additional chargeless loop nucleated in the middle of the cell. In the regime h>hc, the colliding disclinations at θc2 reconfigure into boojum-like twist disclinations.

NASA+: Neural Architecture Search and Acceleration for Multiplication-Reduced Hybrid Networks

Multiplication is arguably the most computation-intensive operation in modern deep neural networks (DNNs), limiting their extensive deployment on resource-constrained devices. Thereby, pioneering works have handcrafted multiplication-free DNNs, which are hardware-efficient but generally inferior to their multiplication-based counterparts in task accuracy, calling for multiplication-reduced hybrid DNNs to marry the best of both worlds. To this end, we propose a Neural Architecture Search and Acceleration (NASA) framework for the above hybrid models, dubbed NASA+, to boost both task accuracy and hardware efficiency. Specifically, NASA+ augments the state-of-the-art (SOTA) search space with multiplication-free operators to construct hybrid ones, and then adopts a novel progressive pretraining strategy to enable the effective search. Furthermore, NASA+ develops a chunk-based accelerator with novel reconfigurable processing elements to better support searched hybrid models, and integrates an auto-mapper to search for optimal dataflows. Experimental results and ablation studies consistently validate the effectiveness of our NASA+ algorithm-hardware co-design framework, e.g., we can achieve up to 65.1% lower energy-delay-product with comparable accuracy over the SOTA multiplication-based system on CIFAR100. Codes are available at https://github.com/GATECH-EIC/NASA.

Electrospinning-based nanofiber architectures for outstanding CO2 capture

The worsening climate change has made the development of alternative and sustainable technologies to achieve carbon neutrality a pressing issue. Numerous strategies for carbon dioxide (CO2) capture/conversion and relevant material preparation have been dealt with on a case-by-case basis. Therefore, a reconfigurable process to secure optimal procedures and materials is urgently required to reduce CO2 emissions in a convenient, rapidly implementable manner. On this account, this study attempted to construct polyethylenimine (PEI)/cerium oxide (CeO2)/activated carbon nanofibers (ACNFs) as a reliable CO2 adsorbent from an electrospinning (ES)-based architecting as a modular process to accelerate their field applications because adsorption has outstanding advantages for CO2 capture, such as process simplicity, wide operational range, low cost, and stable performance. By selecting ES with commercially available materials (PEI (providing CO2 chemisorption), cerium nitrate hexahydrate (CO2 selectivity), and polyacrylonitrile (base of porous nanofibers)), a composite nanofibrous architecture (PEI/CeO2/ACNFs) with outstanding CO2 capture can be achieved by simple process manipulation (manipulating the electrical polarity applied to the nozzle and concentration of the cerium nitrate for the ES). This study may provide a reconfigurable manufacture to ensure a reliable adsorbent for CO2 capture and base materials for CO2 conversion from electro- and photocatalysis.

Adaptation of FPGA architecture for accelerated image preprocessing

The work is devoted to the topical problem at the intersection of communications theory, digital electronics and numerical analysis, namely the study of image processing methods implementation time on different architectures of computational devices, which are used for software and hardware acceleration. The subject of this article is the investigation of reconfigurable FPGA processing systems in the image processing area. The goal of this work is to create a reconfigurable FPGA-based image processing system and compare it with existing processing architectures. Task. To fulfill the requirements of this work, it is necessary to prepare a practical experiment as well as theoretical research of the proposed architecture; to investigate the process of creating a ZYNQ SoC-based image processing system; and to develop and benchmark the speed of execution for the given set of algorithms with the specific range of the picture resolution. Methods used: FPGA simulation, C++ parallel programming with OpenMP, NVIDIA CUDA, performance analysis tools. The result of this work is the development of a resilient SoC Zynq7000–based computing system with programmable logic and the possibility to load images to FPGA RAM using the resources of ARM core for further processing and output via HDMI video interface, which enables the change of PL configuration at any time during the processing process. Conclusions. The efficiency of the FPGA approach was compared with a parallel image processing method implementation with OpenMP and CUDA. An overview of the ZYNQ platform with specific details related to media processing is presented. The analysis of algorithm speed testing findings based on various outputs proved the advantage (of over 60 times) of hardware acceleration of image processing over software analogs. The obtained results may be used in the development of embedded SoC-based solutions that require acceleration of big data processing. Also, the achieved findings can be used during the process of finding a suitable embedded platform for a certain image-processing task, where high data throughput is one of the most desired requirements.

Interplay between alpha and theta band activity enables management of perception-action representations for goal-directed behavior

Goal-directed behavior requires integrated mental representations of perceptions and actions. The neurophysiological underpinnings of these processes, however, are not yet understood. It is particularly undetermined, which oscillatory activities in which brain regions are involved in the management of perception-action representations. We examine this question with a focus on response inhibition processes and show that the dynamics of perception-action representations reflected in theta band activity (TBA) are particularly evident in the supplementary motor area and the occipito-temporal cortex. Mental representations coded in alpha band activity (ABA) during perception-action integration are associated with the occipito-temporal cortex. Crucially, perception-action representations are exchanged between theta and alpha frequency bands. The results imply that ABA functions as dynamic top-down control over binding, retrieval and reconfiguration processes during response inhibition, which in turn are reflected by TBA. Our study thus highlights how the interplay of oscillatory activity enables the management of perception-action representations for goal-directed behavior.

Identifying organizations’ dynamic capabilities for sustainable digital transformation: A mixed methods study

Digital transformation has become a top priority for organizations. As more and more organizations undergo digital transformation, sustainability alignment is gaining attention as it is considered the next big challenge. However, little is currently known about the nexus between digital transformation and environmental sustainability. This study is designed to fill this gap by employing mixed methods to develop a framework for environmentally sustainable digital transformation and explore dynamic capabilities for organizations. To begin with, we review extant literature and adopt the grounded theory approach to define and develop a framework for environmentally sustainable digital transformation (SDT). The framework identifies sustainable digital transformation as a reconfiguration process of an organization's core strategy to align novel digital technologies with sustainability goals. Using a qualitative meta-synthesis procedure (195 full-length articles screened) and a survey questionnaire (n = 63), we identify dynamic capabilities for SDT. Finally, we employ the modified Delphi method to build a consensus of experts (n = 52) on the identified capabilities. This study initially identified 28 capabilities based on the mixed method. However, consensus (>70% agreement) was achieved for 19 capabilities, classified into sustainable seizing, sensing, and transforming capabilities for organizations. The dynamic capabilities outlined in this paper provide insights for organizations seeking sustainable digital transformation. This study contributes to the literature by defining the novel concept of SDT through a conceptual framework and has many practical and theoretical implications.

Anion assisted completely reconfigured manganese oxides with optimal proton adsorption for boosting acidic hydrogen evolution reaction

The optimal proton adsorption design of non-noble metal based catalysts is critical to the realization of efficient hydrogen evolution reactions (HER) in acidic media, but it is also challenging. Herein, sulfur anion-assisted completely reconfigured MnOx (R-S-MnOx-CC) with self-supported nanostructure was prepared on carbon cloth by in situ phase transformation method. The reconfiguration process results in spinel-structured Mn3O4 which can enhance charge transport and expose more active sites through charge disproportionation. The theoretical calculation indicates that the reconfiguration can increase the electron state density of the Mn d band near the Fermi level, promote the right shift of the d-band center, enhance the electron transport capacity of the catalyst to achieve the optimal H+ adsorption. The R-S-MnOx-CC catalyst has excellent acidic HER performance with an overpotential of only 43 mV and 234 mV at 10 mA cm−2 and 400 mA cm−2, respectively, a low charge transfer resistance Rct (0.29 Ω) and stably maintained for 100 h without decay (∼300 mA cm−2). This strategy provides a feasible way to design efficient and stable non-noble metal-based electrocatalysts for the HER industrialization of proton exchange membrane (PEM) water electrolyzers.

An enhanced reconfiguration approach for mitigating the shading effect on photovoltaic array using honey badger algorithm

The series–parallel configuration is one of the widely utilized schemes in the photovoltaic arrays. The existence of shaded or failed modules in the string of the array has a significant impact on the array-produced power. Therefore, this work introduce an innovative objective function and a novel approach of honey badger algorithm (HBA) as a first time to dynamically reconfigure the series–parallel interconnected photovoltaic array under the partial shading and random failure situations to boost the harvested array’s power. The reconfiguration process is performed via electrically reconnecting the modules among the strings via a set of switches. The performance of the proposed approach is examined with two series of analyses. The first one focuses on handling the impact of partial shading phenomena while the second concerns optimizing the array output in case of existing random failed modules inside the array besides partial shading conditions. Two arrays are considered, including a symmetric 9 × 9 array and an asymmetric 10 × 8 photovoltaic one. The proposed optimizer is compared with a set of recently proposed meta-heuristic approaches, including wild horse optimizer, artificial gorilla Troops optimizer, and particle swarm optimizer to clarify the reason for selecting the HBA optimizer. The harvested global maximum power, the mismatch power loss, and enhancement are quality measures performed to affirm the proposed approach’s ability to enhance the array output power. Moreover, energy-saving and income payback are investigated over a day and a year for two systems using symmetric 9 × 9 array and asymmetric 10 × 8 array. The comparisons and analyses showed that, the reconfigured arrays based on HBA provide higher values of maximum global power with smoother photovoltaic characteristics. Moreover, the 9 × 9 and 10 × 8 PV systems-based HBA save income by the percentage of 18.12%, 24.24% compared with series–parallel systems without reconfiguration per year, respectively.

Micro-Scale Mechanical Metamaterial with a Controllable Transition in the Poisson's Ratio and Band Gap Formation.

The ability to significantly change the mechanical and wave propagation properties of a structure without rebuilding it is currently one of the main challenges in the field of mechanical metamaterials. This stems from the enormous appeal that such tunable behavior may offer from the perspective of applications ranging from biomedical to protective devices, particularly in the case of micro-scale systems. In this work, a novel micro-scale mechanical metamaterial is proposed that can undergo a transition from one type of configuration to another, with one configuration having a very negative Poisson's ratio, corresponding to strong auxeticity, and the other having a highly positive Poisson's ratio. The formation of phononic band gaps can also be controlled concurrently which can be very useful for the design of vibration dampers and sensors. Finally, it is experimentally shown that the reconfiguration process can be induced and controlled remotely through application of a magnetic field by using appropriately distributed magnetic inclusions.

Two-Phase Virtual Machine Placement in Cloud Computing Data Centers Using Extreme Learning Machine Prediction-Based Triggering Method

Background: Two-phase Optimization of Virtual Machine Placement (VMP) Problem considers both the Online Incremental VMP (iVMP) phase in which the new arrival of dynamic requests of Virtual Machines VMs are attended to and the Offline VMP reconfiguration (VMPr) phase that performs placement recalculation. In the two-phase scheme, the first part of the two-phase approach is the iVMP, where virtual machines (VMs) can be built, changed, or destroyed at runtime. While the second phase focuses on raising the standard of solutions produced by the iVMP, several studies have been done in different literature to solve the VMP problem. However, the methods used tend to be over-forecast and have long runs of a linear trend. This affects the prediction and produces a less optimal solution. Objective: The following four objective functions are optimized using the proposed Extreme Learning Machine Prediction-Based Triggering Method for Virtual Machine Placement in Cloud Computing Datacenters in Two-Phases, which combines the advantages of both online (dynamic) and static (offline) VMP formulations: the length of the reconfiguration process, the amount of energy used, the way resources are used, and the financial expenses. This study suggests a brand-new strategy for deciding when to start the VMP reconfiguration phase. Results: The Method provides more accuracy to the predicted requests as well as reduces the total economic penalties for service Level Agreement(SLA) violations. An experimental comparison with the existing approach is conducted utilizing 400 cases. Conclusion: The results demonstrated that, in comparison to the benchmark approach, the proposed work obtained a minimum cost function with a 10.5% improvement

A NSGA-II based approach for multi-objective optimization of a reconfigurable manufacturing transfer line supported by Digital Twin: A case study

In response to the wide range of customer demands, the concept of reconfigurable manufacturing systems (RMS) was introduced in the industrial sector. RMS enables producers to meet varying volumes of demand over varying time periods by swiftly adjusting its production capacity and functionality within a part family in response to abrupt market changes. In these circumstances, RMS are made to swiftly reconfigure their Reconfigurable Machine Tools (RMTs). RMTs are designed to have a variety of configurations that may be conditionally chosen and reconfigured in accordance with specific performance goals. However, the reconfiguration process is not an easy process, which entails optimization of several objectives and many of which are inherently conflictual. As a result, it necessitates real-time monitoring of the RMS's condition, which may be achieved by digital twinning, or the real-time capture of system data. The concept of using a digital replica of a physical system to provide real-time optimization is known as digital twin. This work considered a case study of discrete parts manufacturing on a reconfigurable single manufacturing transfer line (SMTL). Six manufacturing operations are required to be performed on the parts at six production stages. This work uses the Digital Twin (DT) based approach to assist a discrete multi-objective optimization problem for a reconfigurable manufacturing transfer line. This multi-objective optimization problem consists of four objective functions which is illustrated by using DT-based Non-dominated Sorting Genetic Algorithm-II (NSGA-II). The innovative aspect of the current study is the use of a DT-based framework for RMS reconfiguration to produce the best optimum solutions. The produced real-time solutions will be of great assistance to the decision maker in selecting the appropriate real-time optimal solutions for reconfigurable manufacturing transfer lines.

Going beyond RF: A survey on how AI-enabled multimodal beamforming will shape the NextG standard

Incorporating artificial intelligence and machine learning (AI/ML) methods within the 5G wireless standard promises autonomous network behavior and ultra-low-latency reconfiguration. However, the effort so far has purely focused on learning from radio frequency (RF) signals. Future standards and next-generation (nextG) networks beyond 5G will have two significant evolution over the state-of-the-art 5G implementations: (i) massive number of antenna elements, scaling up to hundreds-to-thousands in number, and (ii) inclusion of AI/ML in the critical path of the network reconfiguration process that can access sensor feeds from a variety of RF and non-RF sources. While the former allows unprecedented flexibility in ‘beamforming’, where signals combine constructively at a target receiver, the latter enables the network with enhanced situation awareness not captured by a single and isolated data modality. This survey presents a thorough analysis of the different approaches used for beamforming today, focusing on mmWave bands, and then proceeds to make a compelling case for considering non-RF sensor data from multiple modalities, such as LiDAR, Radar, and GPS for increasing beamforming directional accuracy and reducing processing time. This so called idea of multimodal beamforming will require deep learning based fusion techniques, which will serve to augment the current RF-only and classical signal processing methods that do not scale well for massive antenna arrays. The survey describes relevant deep learning architectures for multimodal beamforming, identifies computational challenges and the role of edge computing in this process, dataset generation tools, and finally, lists open research challenges that the community should tackle to realize this transformative vision of the future of beamforming.

Achieving batch-size-of-one production model in robot flexible assembly cells

Manufacturing industry is facing new challenges in that fast-changing demands for products and services from customers push manufacturers to be more flexible and adaptive. The concept of batch-size-of-one production is presented in this paper, which defines a fully automated, highly customised, and short lead time production model. The desired batch-size-of-one production model is a promising solution for the above challenges in manufacturing industry, especially for highly customised or families of similar products like in the mobile phone industry. Along with the concept, we introduce a novel control method that enables the desired batch-size-of-one production model in operation of robots in manufacturing and assembly systems. The strategy was developed for robot control based on a distributed system to enable industrial robots to receive job commands on the fly and to conduct different jobs without the need for reconfiguration and reprogramming and without overheads. The aim of the research is to create the basis for a fully automated robot flexible assembly cell to perform batch-size-of-one assembly tasks with minimal human involvement by eliminating interruptions from the reconfiguration and reprogramming processes. The proposed strategy has been validated in practice in a multi-robot, multi-product flexible assembly cell.

Distribution Network Reconfiguration Based on Hybrid Golden Flower Algorithm for Smart Cities Evolution

Power losses (PL) are one of the most—if not the most—vital concerns in power distribution networks (DN). With respect to sustainability, distribution network reconfiguration (DNR) is an effective course of action to minimize power losses. However, the optimal DNR is usually a non-convex optimization process that necessitates the employment of powerful global optimization methods. This paper proposes a novel hybrid metaheuristic optimization (MO) method called the chaotic golden flower algorithm (CGFA) for PL minimization. As the name implies, the proposed method combines the golden search method with the flower pollination algorithm to multiply their benefits, guarantee the best solution, and reduce convergence time. The performance of the algorithm has been evaluated under different test systems, including the IEEE 33-bus, IEEE 69-bus, and IEEE 119-bus systems and the smart city (SC) network, each of which includes distributed-generation (DG) units and energy storage systems (ESS). In addition, the locations of tie-switches in the DN, which used to be considered as given information in previous studies, are assumed to be variable, and a branch-exchange adaption is included in the reconfiguration process. Furthermore, uncertainty analysis, such as bus and/or line fault conditions, are studied, and the performance of the proposed method is compared with other pioneering MO algorithms with minimal standard deviations ranging from 0.0012 to 0.0101. The case study of SC is considered and the obtained simulation results show the superiority of the algorithm in finding higher PL reduction under different scenarios, with the lowest standard deviations ranging from 0.012 to 0.0432.