Integrated energy management converter based on maximum power point tracking for photovoltaic solar system

This paper presents an integrated power control system for photovoltaic systems based on maximum power point tracking (MPPT). The architecture presented in this paper is designed to extract more power from photovoltaic panels under different partial obscuring conditions. To control the MPPT block, the integrated system used the ripple correlation control algorithm (RCC), as well as a high-efficiency synchronous direct current (DC-DC) boost power converter. Using 180 nm complementary metal-oxide-semiconductor (CMOS) technology, the proposed MPPT was designed, simulated, and layout in virtuoso cadence. The system is attached to a two-cell in series that generates a 5.2 V average output voltage, 656.6 mA average output current, and power efficiency of 95%. The final design occupies only 1.68 mm2.

The Strong Scaling Advantage of FPGAs in HPC for N-body Simulations

N-body methods are one of the essential algorithmic building blocks of high-performance and parallel computing. Previous research has shown promising performance for implementing n-body simulations with pairwise force calculations on FPGAs. However, to avoid challenges with accumulation and memory access patterns, the presented designs calculate each pair of forces twice, along with both force sums of the involved particles. Also, they require large problem instances with hundreds of thousands of particles to reach their respective peak performance, limiting the applicability for strong scaling scenarios. This work addresses both issues by presenting a novel FPGA design that uses each calculated force twice and overlaps data transfers and computations in a way that allows to reach peak performance even for small problem instances, outperforming previous single precision results even in double precision, and scaling linearly over multiple interconnected FPGAs. For a comparison across architectures, we provide an equally optimized CPU reference, which for large problems actually achieves higher peak performance per device, however, given the strong scaling advantages of the FPGA design, in parallel setups with few thousand particles per device, the FPGA platform achieves highest performance and power efficiency.

xDNN: Inference for Deep Convolutional Neural Networks

We present xDNN, an end-to-end system for deep-learning inference based on a family of specialized hardware processors synthesized on Field-Programmable Gate Array (FPGAs) and Convolution Neural Networks (CNN). We present a design optimized for low latency, high throughput, and high compute efficiency with no batching. The design is scalable and a parametric function of the number of multiply-accumulate units, on-chip memory hierarchy, and numerical precision. The design can produce a scale-down processor for embedded devices, replicated to produce more cores for larger devices, or resized to optimize efficiency. On Xilinx Virtex Ultrascale+ VU13P FPGA, we achieve 800 MHz that is close to the Digital Signal Processing maximum frequency and above 80% efficiency of on-chip compute resources. On top of our processor family, we present a runtime system enabling the execution of different networks for different input sizes (i.e., from 224× 224 to 2048× 1024). We present a compiler that reads CNNs from native frameworks (i.e., MXNet, Caffe, Keras, and Tensorflow), optimizes them, generates codes, and provides performance estimates. The compiler combines quantization information from the native environment and optimizations to feed the runtime with code as efficient as any hardware expert could write. We present tools partitioning a CNN into subgraphs for the division of work to CPU cores and FPGAs. Notice that the software will not change when or if the FPGA design becomes an ASIC, making our work vertical and not just a proof-of-concept FPGA project. We show experimental results for accuracy, latency, and power for several networks: In summary, we can achieve up to 4 times higher throughput, 3 times better power efficiency than the GPUs, and up to 20 times higher throughput than the latest CPUs. To our knowledge, we provide solutions faster than any previous FPGA-based solutions and comparable to any other top-of-the-shelves solutions.

The Impacts of Market Segmentation on Thermal Power Generation Efficiency

China’s power industry is in a critical transformation period. The new round of power system reform in 2015 will have a profound impact on China’s power industry. Therefore, it’s necessary to analyze the influencing factors of thermal power generation efficiency. Based on the thermal power generation industry related data in China’s 30 provinces from 2005 to 2017, this paper studies the impacts of market segmentation on thermal power generation efficiency in China. And the empirical result shows that the market segmentation exhibit significant negative effects on the thermal power generation efficiency, that is, the thermal power generation efficiency significantly decrease 1.6799 for each unit increase of market segmentation index of thermal power industry. Besides, by decomposing the dynamic thermal power efficiency index, we find that the “innovation effect” is the primary channel for the market segmentation to make effects on the thermal power generation efficiency. Furthermore, our findings are still robust after considering endogenous problems and eliminating the relevant data. Finally, research conclusions of our study paper provide empirical supports for the efficient development of China’s power market.

Power Reduction in Punch-Through Current-Based Electro-Thermal Annealing in Gate-All-Around FETs

Device guidelines for reducing power with punch-through current annealing in gate-all-around (GAA) FETs were investigated based on three-dimensional (3D) simulations. We studied and compared how different geometric dimensions and materials of GAA FETs impact heat management when down-scaling. In order to maximize power efficiency during electro-thermal annealing (ETA), applying gate module engineering was more suitable than engineering the isolation or source drain modules.

Analysis and Detection of Erosion in Wind Turbine Blades

This paper studies erosion at the tip of wind turbine blades by considering aerodynamic analysis, modal analysis and predictive machine learning modeling. Erosion can be caused by several factors and can affect different parts of the blade, reducing its dynamic performance and useful life. The ability to detect and quantify erosion on a blade is an important predictive maintenance task for wind turbines that can have broad repercussions in terms of avoiding serious damage, improving power efficiency and reducing downtimes. This study considers both sides of the leading edge of the blade (top and bottom), evaluating the mechanical imbalance caused by the material loss that induces variations of the power coefficient resulting in a loss in efficiency. The QBlade software is used in our analysis and load calculations are preformed by using blade element momentum theory. Numerical results show the performance of a blade based on the relationship between mechanical damage and aerodynamic behavior, which are then validated on a physical model. Moreover, two machine learning (ML) problems are posed to automatically detect the location of erosion (top of the edge, bottom or both) and to determine erosion levels (from 8% to 18%) present in the blade. The first problem is solved using classification models, while the second is solved using ML regression, achieving accurate results. ML pipelines are automatically designed by using an AutoML system with little human intervention, achieving highly accurate results. This work makes several contributions by developing ML models to both detect the presence and location of erosion on a blade, estimating its level and applying AutoML for the first time in this domain.

Ferroelectric-based synapses and neurons for neuromorphic computing

The shift towards a distributed computing paradigm, where multiple systems acquire and elaborate data in real-time, leads to challenges that must be met. In particular, it is becoming increasingly essential to compute on the edge of the network, close to the sensor collecting data. The requirements of a system operating on the edge are very tight: power efficiency, low area occupation, fast response times, and on-line learning. Brain-inspired architectures such as Spiking Neural Networks (SNNs) use artificial neurons and synapses that simultaneously perform low-latency computation and internal-state storage with very low power consumption. Still, they mainly rely on standard complementary metal-oxide-semiconductor (CMOS) technologies, making SNNs unfit to meet the aforementioned constraints. Recently, emerging technologies such as memristive devices have been investigated to flank CMOS technology and overcome edge computing systems' power and memory constraints. In this review, we will focus on ferroelectric technology. Thanks to its CMOS-compatible fabrication process and extreme energy efficiency, ferroelectric devices are rapidly affirming themselves as one of the most promising technology for neuromorphic computing. Therefore, we will discuss their role in emulating neural and synaptic behaviors in an area and power-efficient way.

Impact of Asymmetric Weight Update on Neural Network Training With Tiki-Taka Algorithm

Recent progress in novel non-volatile memory-based synaptic device technologies and their feasibility for matrix-vector multiplication (MVM) has ignited active research on implementing analog neural network training accelerators with resistive crosspoint arrays. While significant performance boost as well as area- and power-efficiency is theoretically predicted, the realization of such analog accelerators is largely limited by non-ideal switching characteristics of crosspoint elements. One of the most performance-limiting non-idealities is the conductance update asymmetry which is known to distort the actual weight change values away from the calculation by error back-propagation and, therefore, significantly deteriorates the neural network training performance. To address this issue by an algorithmic remedy, Tiki-Taka algorithm was proposed and shown to be effective for neural network training with asymmetric devices. However, a systematic analysis to reveal the required asymmetry specification to guarantee the neural network performance has been unexplored. Here, we quantitatively analyze the impact of update asymmetry on the neural network training performance when trained with Tiki-Taka algorithm by exploring the space of asymmetry and hyper-parameters and measuring the classification accuracy. We discover that the update asymmetry level of the auxiliary array affects the way the optimizer takes the importance of previous gradients, whereas that of main array affects the frequency of accepting those gradients. We propose a novel calibration method to find the optimal operating point in terms of device and network parameters. By searching over the hyper-parameter space of Tiki-Taka algorithm using interpolation and Gaussian filtering, we find the optimal hyper-parameters efficiently and reveal the optimal range of asymmetry, namely the asymmetry specification. Finally, we show that the analysis and calibration method be applicable to spiking neural networks.

Evaluating multifunctional efficiency of a structural battery composite via thermo-electro-chemical modelling

The emerging research field of structural batteries aims to combine the functions of load bearing and energy storage to improve system-level energy storage in battery-powered vehicles and consumer products. Structural batteries, when implemented in electric vehicles, will be exposed to greater temperature fluctuations than conventional batteries in EVs. However, there is a lack of published data regarding how these thermal boundary conditions impact power capabilities of the structural batteries. To fill this gap, the present work simulates transient temperature-dependent specific power capabilities of high aspect ratio structural battery composite by solving one-dimensional heat transfer equation with heat source and convective boundary conditions. Equivalent circuit modeling of resistivity-induced losses is used with a second-order finite difference method to examine battery performance. More than 60 different run configurations are evaluated, examining how thermal boundary conditions and internal heat influence power capabilities and multifunctional efficiency of the structural battery. The simulated structural battery composite is shown to have good specific Young’s modulus (79.5 to 80.3% of aluminum), a specific energy of 158 Wh/kg, and specific power of 41.2 to 55.2 W/kg, providing a multifunctional efficiency of 1.15 to 1.17 depending on configuration and thermal loading conditions and demonstrating the potential of load-bearing structural batteries to achieve mass savings. This work emphasizes the dependency of power efficiency on cell design and external environmental conditions. Insulating material is shown to improve multifunctional efficiency, particularly for low ambient temperatures. It is demonstrated that as cell temperature increases due to high ambient temperature or heat generation in the battery, the specific power efficiency increases exponentially due to a favorable nonlinear relation between ionic conductivity and cell temperature. The simulations also demonstrate a thermal feedback loop where resistivity-induced power losses can lead to self-regulation of cell temperature. This effect reduces run-averaged losses, particularly at low ambient temperature.

Total Factor Energy Efficiency of China’s Thermal Power Industry

Coal-based thermal power generation has long been the main source of power generation in the mainland of China. The efficiency of power generation is an important factor that determines the energy conservation and emission reduction as well as the sustainable development of the power industry in China. By comparing the regional differences of 30 provinces in the mainland from 2013 to 2017, this study uses the Super-DDF model and the TFEE to comprehensively evaluate the energy efficiency of thermal power generation. Empirical results: Overall efficiency: eastern efficiency (1.181) is the highest, followed by western (0.956), central (0.951) and northeastern (0.926). Total factor energy efficiency: eastern efficiency (0.923) is the highest, followed by western (0.754), central (0.742) and northeastern (0.710). The government and power industry managers should fully consider the regional differences in the field of thermal power generation when formulating policies so as to improve the power efficiency and promote the green development of power industry in China. Based on the analysis results, although the coal-fired power industry is more mature than other alternative energy industries, the expansion of thermal power generation cannot be considered if CO2 emissions are to be reduced. Additionally, the market share and competitiveness of the local power industry can be increased based on the different conditions of the resource endowments of each region.