Electric load forecasting has always been a key component of power grids. Many countries have opened up electricity markets and facilitated the participation multiple agents, which create competitive environment reduce costs to consumers. In market, multi-step short-term becomes increasingly significant for market bidding spot price calculation, but performances traditional algorithms are not robust unacceptable enough. recent years, rise deep learning gives us opportunity improve accuracy...

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

In many applications, phase shift modulation (PSM) has been proven to be a promising supplement of pulse frequency (PFM) in LLC resonant converter. However, steady-state analysis PSM converter is not thoroughly studied yet. This paper presents generalized model which can provide accurate prediction on voltage and current behavior, dc gain all the other information. Then, operation modes, mode boundaries distribution are discussed with three cases according relationship between switching...

Residential load forecasting is important for many entities in the electricity market, but profile of single residence shows more volatilities and uncertainties. Due to difficulty producing reliable point forecasts, probabilistic becomes popular as a result catching volatility uncertainty by intervals, density, or quantiles. In this paper, we propose unified quantile regression deep neural network with time-cognition tackling challenging issue. At first, convolutional multiscale convolution...

Maxillary sinus segmentation plays an important role in the choice of therapeutic strategies for nasal disease and treatment monitoring. Difficulties traditional approaches deal with extremely heterogeneous intensity caused by lesions, abnormal anatomy structures, blurring boundaries cavity. 2D 3D deep convolutional neural networks have grown popular medical image due to utilization large labeled datasets learn discriminative features. However, images, are not competent extracting more...

Improving energy efficiency management has become an important task for current electricity market participating entities, and monitoring consumption of pivotal appliances plays role in many applications. This paper focuses on detecting whether a residence hold certain type appliance based their consumption. We propose data-driven deep learning approach with attention mechanism to detect residential from low-resolution aggregate data. Firstly, the historical profile each user is divided into...

Household load forecasting provides great challenges as a result of high uncertainty in individual consumption profile. Traditional models based on machine learning tried to explore depending clustering, spectral analysis, and sparse coding with hand craft features. Recently, deep skills like recurrent neural network attempt learn the one-hot encoding which is too simple not efficient. In this paper, for first time, we proposed multitask convolutional household forecasting. The baseline one...

Abstract In this work, flame retarding phosphonate‐functionalized imidazolium ionic liquids (PFILs) were synthesized and employed as electrolyte additives for Li/LiFePO 4 half‐cells. Physical electrochemical properties, including conductivity, oxidation limit, impedance spectroscopy (EIS) lithium‐ion transference number (T Li + ), investigated. The results demonstrate that the addition of PFIL can improve performance batteries (LIBs). Hybrid with 10 wt % bidentate liquid (b‐PFIL) exhibits...

The convolutional neural network (CNN) has made great success in many fields, and is gradually being applied edge-computing systems. Taking the limited budget of resources systems into consideration, implementation CNNs on embedded devices preferred. However, accompanying increasingly complex huge cost memory, which constrains its devices. In this paper, we propose an efficient, pipelined convolution module based a Brain Floating-Point (BF16) to solve problem, composed quantization unit,...

Based on the principle and mathematical model of Active Power Filter(APF), this paper proposes a improved method which is aiming to solve selection for dc capacitor parameter. On basis instantaneous reactive power theory energy flow relationships ac sides in APF inverters different compensation purposes are analyzed. The voltage fluctuation caused by inverter losses active component. Taking three-phase non-controllable load as an example, calculation parameter given. less than 0.55 per cent....

Application of low-voltage insulated gate bipolar transistors (IGBTs) in high voltage power electronics system is attractive for low cost, operating frequency, weight and volume. The paper presents a multi-module-cascade High-Voltage Switch based on modular multilevel converter (MMC). With the cascade modules, it realized high-voltage composite switches with working voltages several kilovolts. switch proposed can absorb spikes caused by parasitic parameters back energy stored to supply....

Accurate and comprehensive reconstruction of in-cylinder combustion process is essential for timely monitoring engine state. This article developed a method based on the zero-dimensional (0-D) physical model integrated with big data. The traditional 0-D prediction cumulative fuel mass improved, factor temperature introduced to adjust heat release rate, which solves problem difficulty in calibrating rate. Then, convolutional neural network-gated recurrent unit (CNN-GRU), as deep network,...

Abstract A measurement method and system based on conventional radars has been proposed to address the problem of radar cross-section (RCS) ship targets in open seas. Calibration errors for RCS ground plane range (GPR) are analysed, error elimination accurate calculation formula given. An is designed by improving radars. These systems measure two Luneburg lens reflectors different sizes lake. The average big reflector obtained, calculated be 2.73% compared with its actual value, which...

Abstract According to the characteristics of multi-demand integrated energy customers, multi-faceted projects and complex technology categories, a decision-making software for system planning is proposed. By constructing investment income model in park, realizes accurate capacity evaluation system, which provides auxiliary decision support end users, service providers so on. The intelligent optimization scheme through algorithm, optimal park. enriches means design, improves efficiency planning.

Sign language recognition and translation bridges the gap between hearing-impaired people ordinary people. Compared with sign (SLR), continuous (CSLT) is closer to people's speaking habits, has greater practicality. However, there are problems such as insufficient information short pauses in sentences that difficult separate. To solve these problems, this paper gets coarse-grained arm movement, fine-grained finger movement hand rotation through MYO armband; an encoder-decoder model attention...

To address challenges in manual detection of electronic component defects facility greenhouses, this paper presents an defect method using the Improved YOLOv5 recognition algorithm. By introducing Convolutional Block Attention Module into backbone network, model's identification and classification types are enhanced, network's receptive field is improved, accuracy increased. The proposed also utilizes α-CIoU loss function to expedite network regression. effectiveness model was evaluated on a...