Forward Topology Research Articles

Time Series Classification Based on Forward Echo State Convolution Network

The Echo state network (ESN) is an efficient recurrent neural network that has achieved good results in time series prediction tasks. Still, its application in time series classification tasks has yet to develop fully. In this study, we work on the time series classification problem based on echo state networks. We propose a new framework called forward echo state convolutional network (FESCN). It consists of two parts, the encoder and the decoder, where the encoder part is composed of a forward topology echo state network (FT-ESN), and the decoder part mainly consists of a convolutional layer and a max-pooling layer. We apply the proposed network framework to the univariate time series dataset UCR and compare it with six traditional methods and four neural network models. The experimental findings demonstrate that FESCN outperforms other methods in terms of overall classification accuracy. Additionally, we investigated the impact of reservoir size on network performance and observed that the optimal classification results were obtained when the reservoir size was set to 32. Finally, we investigated the performance of the network under noise interference, and the results show that FESCN has a more stable network performance compared to EMN (echo memory network).

WITHDRAWN: Application of Data Mining Technology in Analysis of Influencing Factors of Smart Electric Power Emergency Rescue

Because electric emergency rescue is affected by many factors, there are certain problems in the rescue process. In order to improve the power emergency rescue effect, based on data mining technology, this paper selects the neural network algorithm as the main analysis algorithm of the model, and uses the multi-layer forward neural network topology to perform modeling and correlation analysis. Moreover, this paper establishes the functional architecture of the power system emergency plan deduction system based on the analysis of the demand for the power system emergency plan deduction system. In addition, this paper combines the power emergency rescue system requirements to build the model function module, proposes a mobile model based on the emergency rescue scene, and analyzes its realization process. Finally, in order to study the system performance, this paper designs experiments to analyze the model performance, and combines mathematical statistics to draw charts and score the system model. The research results show that the power rescue influence factor analysis model constructed in this paper is effective.

Efficient Optimization approach in WSN using Multicast Forwarding Scheme in 6LowPAN Environment

The rapidly growing network of devices that are connected known as the Internet of Things, also referred to as IoT, is designed to collect and exchange data. These devices are integrated with sensors, software, and other technologies. An IPv6-based routing system called 6LowPAN was created especially for Internet of Things devices with little resources. However, because multicast packets must be replicated to every multicast group member, the conventional multicast forwarding technique in 6LowPAN may be wasteful. This can result in excessive energy usage and network traffic, particularly in large-scale Internet of Things implementations. An optimization process is presented here using a multicast forwarding scheme. In the proposed forwarding scheme, nodes can send multicast packets up and down at Low Power and Lossy networks over the 6LowPAN environment. Should the sink node be on the root node, it should immediately request that other nodes join the multicast messaging network of things. If otherwise, an ICMPv6 packet will be sent towards the root node in the Destination-oriented Directed-Acyclic Graphs (DODAG) tree, which expands the multicast messages in the first source's interest. In multi-bounce forward straight-line topology, when the DODAG network root is the source of multicast congestion, it exhibits a similar ratio of packet delivery and delay from end to end with a minimal memory overhead. When traffic via multicast originates from the most important rank node in a backward straight-line topology, this paper focuses on the efficiency of the multicast network at 6LowPAN.

Nacre-like block lattice metamaterials with targeted phononic band gap and mechanical properties

The extraordinary quasi-static mechanical properties of nacre-like composite metamaterials, such as high specific strength, stiffness, and toughness, are due to the periodic arrangement of two distinct phases in a “brick and mortar” structure. It is also theorized that the hierarchical periodic structure of nacre structures can provide wider band gaps at different frequency scales. However, the function of hierarchy in the dynamic behavior of metamaterials is largely unknown, and most current investigations are focused on a single objective and specialized applications. Nature, on the other hand, appears to develop systems that represent a trade-off between multiple objectives, such as stiffness, fatigue resistance, and wave attenuation. Given the wide range of design options available to these systems, a multidisciplinary strategy combining diverse objectives may be a useful opportunity provided by bioinspired artificial systems. This paper describes a class of hierarchically-architected block lattice metamaterials with simultaneous wave filtering and enhanced mechanical properties, using deep learning based on artificial neural networks (ANN), to overcome the shortcomings of traditional design methods for forward prediction, parameter design, and topology design of block lattice metamaterial. Our approach uses ANN to efficiently describe the complicated interactions between nacre geometry and its attributes, and then use the Bayesian optimization technique to determine the optimal geometry constants that match the given fitness requirements. We numerically demonstrate that complete band gaps, that is attributed to the coupling effects of local resonances and Bragg scattering, exist. The coupling effects are naturally influenced by the topological arrangements of the continuous structures and the mechanical characteristics of the component phases. We also demonstrate how we can tune the frequency of the complete band gap by modifying the geometrical configurations and volume fraction distribution of the metamaterials. This research contributes to the development of mechanically robust block lattice metamaterials and lenses capable of controlling acoustic and elastic waves in hostile settings.

Towards Full Forward On-Tiny-Device Learning: A Guided Search for a Randomly Initialized Neural Network

In the context of TinyML, many research efforts have been devoted to designing forward topologies to support On-Device Learning. Reaching this target would bring numerous advantages, including reductions in latency and computational complexity, stronger privacy, data safety and robustness to adversarial attacks, higher resilience against concept drift, etc. However, On-Device Learning on resource constrained devices poses severe limitations to computational power and memory. Therefore, deploying Neural Networks on tiny devices appears to be prohibitive, since their backpropagation-based training is too memory demanding for their embedded assets. Using Extreme Learning Machines based on Convolutional Neural Networks might be feasible and very convenient, especially for Feature Extraction tasks. However, it requires searching for a randomly initialized topology that achieves results as good as those achieved by the backpropagated model. This work proposes a novel approach for automatically composing an Extreme Convolutional Feature Extractor, based on Neural Architecture Search and Bayesian Optimization. It was applied to the CIFAR-10 and MNIST datasets for evaluation. Two search spaces have been defined, as well as a search strategy that has been tested with two surrogate models, Gaussian Process and Random Forest. A performance estimation strategy was defined, keeping the feature set computed by the MLCommons-Tiny benchmark ResNet as a reference model. In as few as 1200 search iterations, the proposed strategy was able to achieve a topology whose extracted features scored a mean square error equal to 0.64 compared to the reference set. Further improvements are required, with a target of at least one order of magnitude decrease in mean square error for improved classification accuracy. The code is made available via GitHub to allow for the reproducibility of the results reported in this paper.

A 5 Level Multilevel Inverter

Scheduled Inverter could be a kind of unused multilevel inverter topology utilized in HVDC framework and medium voltage applications like acceptance engine drive. Measured Multilevel inverter can work at tall voltages and current with less constant and basic channels, and without expanding the number of switches. This paper basically centers on straight forward topology for measured multilevel inverter in order to decrease the number of switches additionally boost the voltage of the system by utilizing the capacitor, Diode and a switch. And after that utilizing of lesser switches and by employing a capacitor the inverter gets to be the self-balancing inverter. In this we are employing a Battery as a putting away gadget. The complete circuit is carried out/Simulated in a MATLAB.

Dynamic Optimization Method of Knowledge Graph Entity Relations for Smart Maintenance of Cantilever Roadheaders

The fault maintenance scenario in coal-mine equipment intelligence is composed of videos, images, signals, and repair process records. Text data are not the primary data that reflect the fault phenomenon, but rather the secondary processing based on operation experience. Focusing on the difficulty of extracting fault knowledge from the limited textual maintenance process records, a forward static full-connected topology network modeling method based on domain knowledge from four dimensions of physical structure, internal association, condition monitoring, and fault maintenance, is proposed to increase the efficiency of constructing a fault-maintenance knowledge graph. Accurately identifying the intrinsic correlation between the equipment anomalies and the faults’ causes through only domain knowledge and loosely coupled data is difficult. Based on the static full-connected knowledge graph of the cantilever roadheader, the information entropy and density-based DBSCAN clustering algorithm is used to process and analyze many condition-monitoring historical datasets to optimize the entity relationships between the fault phenomena and causes. The improved DBSCAN algorithm consists of three stages: firstly, extracting entity data related to fault information from the static fully connected graph; secondly, calculating the information entropy based on the real dataset describing the fault information and the historical operating condition, respectively; and thirdly, comparing the entropy values of the entities and analyzing the intrinsic relationship between the fault phenomenon, the operating condition data, and the fault causes. Based on the static full-connected topology storage in the Neo4j database, the information entropy and density-based DBSCAN algorithm is computed by using Python to identify the relationship weights and dynamically display optimized knowledge graph topology. Finally, an example of EBZ200-type cantilever roadheader for smart maintenance is studied to analyze and evaluate the forward and four-mainlines knowledge graph modeling method and the dynamic entity relations optimization method for static full-connected knowledge graph.

Dual Outputs Switch Mode Power Supply (DOSMPS) Utilizing Forward Converter Topology

Experimental works of dual, 5 volt DC outputs switch mode power supply (DOSMPS) is presented in this paper. To design, construct and evaluate the DOSMPS using forward converter topology are the objectives of this article. In this work, a 556 timer was picked as the pulse width modulator. The 556 timer was picked because it is widely used, cheap and capable of producing stable dual pulse width modulations. Since, this work targeted dual DC outputs, two forward converter circuits were built. However, switching frequencies at two transistors were controlled by only one 556 timer. The complete circuit of the DOSMPS consist of full bridge rectifier, smoothing capacitor, two high frequency transformers and two forward converter circuits. At the first stage of this work, the DOSMPS circuit was designed using Proteus software. Later, the complete DOSMPS circuit was constructed on two breadbroads after all of the electronic components and devices were available. Important parameters such as the input and output voltages and duty cycles were measured using digital oscilloscope and multimeter. Results from these measurements have shown that the designed circuit can convert single-phase 24 Vrms 50 Hz AC voltage input into dual and stable 5±0.1 V DC outputs. Also from this work, it was observed that, pulsed DC voltages with peak value of 2 V and frequency less than 20 kHz are sufficient in controlling the switching frequencies of both BJT transistors. This allows the regulation of the output voltage within the desired range. Analysis carried out after the measurements has shown close correlation between the theoretical and experimental results, where the calculated and measured switching frequencies had only deviated by 1.1%.

Modelling and analysis of voltage balancing topology for series-connected supercapacitor energy storage system

ABSTRACT This paper proposes a forward conversion topology using a multi-winding transformer known as global modular balancing (GMB) for multiple supercapacitor cells connected in a series of high-voltage applications. The proposed topology can easily be modularized and is not limited to the number of supercapacitors existing in the modules. Moreover, the proposed GMB needs merely one MOSFET switch for each supercapacitor, and a one Pulse width modulation circuit that can control all the switches in the modules. Furthermore, the proposed topology is employed to achieve voltage balancing in a supercapacitor string. The proposed topology balancing performance is validated through MATLAB/Simulink Software.

Deep learning for the design of phononic crystals and elastic metamaterials

Abstract The computer revolution coming by way of data provides an innovative approach for the design of phononic crystals (PnCs) and elastic metamaterials (EMs). By establishing an analytical surrogate model for PnCs/EMs, deep learning based on artificial neural networks possesses the superiorities of rapidity and accuracy in design, making up for the shortcomings of traditional design methods. Here, the recent progresses on deep learning for forward prediction, parameter design, and topology design of PnCs and EMs are reviewed. The challenges and perspectives in this emerging field are also commented.

A High Efficiency Modified Forward Converter for Solar Photovoltaic Applications

In this paper, a modification of a conventional forward converter and a new mixed-stranded transformer winding technique are proposed for the development of a simple and high efficiency isolated dc-dc boost converter for solar photovoltaic applications. The conventional single switch forward converter topology is modified for simplification and elimination of losses due to the output filter components. A new mixed-stranded transformer winding technique is proposed to minimize the leakage inductance between the primary and tertiary windings for reducing losses. A pulse frequency modulation technique (PFM) is employed for reducing switching losses at low power operation. An experimental prototype of 300 W with a voltage step-up ratio of 1:12 has been built to obtain dc output voltage between 360-480 V for the dc input voltage of 30-40 V. Through the experimental results, it is shown that the converter can boost the input voltage according to the step-up ratio and exhibits a remarkably high dc-dc power conversion efficiency with a flat response curve. The average efficiency of around 95.5 % and peak efficiency of 97.6% are obtained.

The Analysis, Modeling, and Control of the Forward DC/DC Converter for the Electric Vehicle

This article deals with the analysis, modeling, and control of the DC/DC converters. Although it is considered with the analysis of the power semiconductor converters in general however it can be applied also to sophisticated electronic devices perfectly suitable for on-board applications of a huge variety of transport vehicles. Specifically, as an example, the Forward type topology is used which offers the advantage of the galvanic isolation of output. The task is to obtain a state space description of the system, the transfer function of the control, and design controller parameters using MATLAB. The state space model and the transfer function is implemented for the ideal and nonideal type of topology, where it is considered with at least some parasitic parameters of circuit elements to get closer to real results of the physical sample and to compare obtained transfer functions based on their frequency characteristics.

A Boost-Type Three-Port Resonant Forward Converter With Flexible Power Flow Path Optimization for PV Systems

This brief proposes a three-port resonant forward topology for low power applications. Specifically, a modified boost converter, which adds a controllable freewheeling path, is combined with a resonant forward converter to simultaneously fulfill PV voltage regulation and battery charging management. Then, by taking the benefit of more flexible power flow adjustment, the PV and the battery can adaptively share the load current with reduced power conversion stages. Therefore, in comparison to other LLC-type three-port converters, the proposed scheme achieves the better compromise between cost and power efficiency.

Fault Location Estimator Design for Power Distribution System Using Artificial Neural Network

Fault location in distribution system is critical issue to increase the availability of power supply by reducing the time of interruption for maintenance in electric utility companies. Fault location estimator for power distribution system using artificial neural network is developed for line to ground, line to line, line to line to ground and three phases to ground faults in distribution system. To develop this estimator is one of rural radial power distribution feeder in Ethiopia, south west reign, Aba substation Tarcha line feeder is used as a test feeder. This feeder is simulated using ETAP software to generate data for different fault condition, with different fault resistance and loading conditions, which is the fault phase voltage and current. The generated data is preprocessed and put as an input for neural network to be trained. MATLAB R2016a neural network toolbox to train ANN and programming toolbox is used to develop graphic user interface for fault estimator. The feed forward multi-layer network topologies of neural network with improved back propagation, Liebenberg Marquardt learning algorithm is used to train the network. After the network is trained the mean square error performance, regression plot and error histogram analysis was made and found to have an excellent performance with regression coefficient 0.99929 , validation performance of 0.000102 and error histogram range 0.015 to 0.019. In this thesis for practical implementation the fault records at the test feeder is handled by intelligent electronic device (IED) installed at the substation feeders. The fault record of IED can be read by PCM600 tool using laptop or manually using IEDs human machine interface, this fault recorded data feed to the graphic user interface to estimate the fault location as well as the fault type. Finally it is found that artificial neural networks are one of the alternate options in fault estimator design for distribution system where sufficient distribution network data are available with narrow fault location distance range from the substation. This has benefits in assisting for maintenance plan, saving efforts in fault location finding and economic benefits by reducing interruption time. Keywords : Power distribution, artificial intelligence, neural network, feed forward network DOI: 10.7176/JETP/12-5-01 Publication date: November 30 th 2022

Cryogenic DC/DC Converter for Superconducting Magnet Application

Recent advances in superconductors have led to their increased application in magnets and a requirement to carry ever-larger currents. This article describes the development and implementation of a cryogenic DC/DC converter for supplying a DC excitation current of 500 A for superconducting magnet applications. Practical aspects are addressed and the rationale behind selecting an interleaved forward converter topology is elucidated. Experimental results are given and a ripple content in the load current of less than 5% is estimated.

Dual Voltage Forward Topology for High Efficiency at Universal Mains

This paper introduces a forward converter aimed at the universal mains voltages, i.e., 220–230 Vac and 115 Vac, named the ‘dual voltage forward converter’. The suggested converter has a narrow dynamic range at the universal mains voltages, which results in lower stress on devices, optimal duty cycles, and better overall efficiency. The topology comprises two primary power loops reconfigurable by additional two-state switches and a passive diode, which allows the converter to run in parallel or in series modes and increase the performance over the full universal mains range of 90–265 Vac. The utilization of the devices is better, as they experience lower voltage and current stress by supporting two optimized working points. A converter operating at 100 kHz with an output power of 75 W and output voltage of 12 Vdc was designed and tested. The results were compared with a conventional forward converter with the identical specification. The results at the low mains were similar between the converters; however, at the high mains, the efficiency improvement was between 5% and 23%.

W band solid‐state power amplifier for aerospace usage

Concentrating on the high frequency aerospace demand of solid-state power amplifier (SSPA), this paper presents the methodology and key technique of a W-band 3-W SSPA that achieves high efficiency while maintaining excellent reliability capability. An improved Magic-T power combiner is utilized to achieve high output power, and eliminate the influence of the parallel end-stage gallium-nitride high-electron-mobility-transistor (GaN HEMT) PAs by its good isolation performance. The SSPA's millimetre-wave circuit is implemented by three stages of amplifier, to realized more than 30-dB power gain. The power-supply circuit based on the two-transistor forward topology is proposed for voltage converting and output serious sequence controlling. The SSPA is fabricated and measured, achieving 33-dBm to 34.1-dBm output power within the frequency range of 92–96 GHz, with the power gain of 31–32 dB, and peak efficiency of 7.1%, respectively. The proposed SSPA shows competitive performance for space application.

Modeling and impact analysis of connected vehicle merging accounting for mainline random length tight-platoon

To reveal impacts of tight-platoon on connected vehicle (CV) merging, a merging advisory model is proposed to account for mainline random length tight-platoon. The tight-platoon is modeled by a consensus based algorithm with constant spacing under looking forward communication topologies, and string stabilities are analyzed to ensure robustness against gap disturbances. To ensure traffic safety, a floating merging point speed advisory model (FMP-SAM) based on finite state machine and virtual vehicle mapping is proposed to address merging conflicts. Besides, a rolling horizon control is employed in FMP-SAM to smooth speed trajectory of merging leader during gap adaption. Simulations results indicate that encouraging CV to drive in tight-platoon can improve safety, enhance mobility, conserve fuel consumption, reduce congestion and improve road capacity at higher traffic demand. Besides, FMP-SAM can improve FE performance greatly without sacrificing other performance metrics when compared with baseline merging advisory model.

Research on a new converter topology for energy transmission between spacecrafts

A novel power converter based on dual-transistor forward converter topology is proposed to meet the requirements of high power transmission between spacecrafts. Compared with the traditional two-transistor forward converter, this converter eliminates the secondary filter inductance and the diode of the rectifier. Theoretical analysis and experimental results show that the new topology reduces the use of magnetic elements, improves the power density of the converter, and improves the parasitic oscillation and reverse recovery of the rectifier diode. The new topology can better meet the requirements of power supply system for high reliability, reducing device stress, increasing the availability of key devices, optimizing system structure and so on.

A Family of DC Transformer (DCX) Topologies Based on New ZVZCS Cells With DC Resonant Capacitance

The series resonant soft-switching topology is widely used in high-density DCX products due to its zero switching loss. But, the conventional series resonant converter is a dual end topology, which results in complex circuit structure and high cost. In order to simplify the resonant DCX topology and reduce the cost, a novel resonant cell named input current fed (ICF) cell is proposed in this paper, which can be applied to single-ended forward topologies. The zero-voltage-switching (ZVS) on and quasi-zero-current-switching (ZCS) off for switch are both achieved. According to the duality theory of circuit, another output current sink cell is derived. Based on the proposed ZVZCS ICF cell, two cells are interleaved to construct several dual end cells, which can be extended to push-pull, full bridge, half bridge, and other dual end topologies. Based on these cells, a new family of ZVZCS resonant DCX is derived, which are suitable for high-frequency and high-efficiency applications with various power levels. The operating principle and characteristics of the cells and topologies are verified by simulations and experimental results.