Sea surface temperature (SST) forecasting is the task of predicting future values a given sequence using historical SST data, which beneficial for observing and studying hydroclimatic variability. Most previous studies ignore spatial information in prediction models have limitations to process large-scale data. A novel model integrated Deep Gated Recurrent Unit Convolutional Neural Network (DGCnetwork) proposed this paper. The DGCnetwork has compact structure focuses on learning deep...

Accurately identifying protein functions is essential to understand life mechanisms and thus advance drug discovery. Although biochemical experiments are the gold standard for determining functions, they often time-consuming labor-intensive. Here, we proposed a novel composite deep-learning method, MKFGO, infer Gene Ontology (GO) attributes through integrating five complementary pipelines built on multi-source biological data. MKFGO was rigorously benchmarked 1522 non-redundant proteins,...

We propose an F-band phase shifter based on the nematic liquid crystals (NLCs). The proposed is formed by a voltage-controlled cavity through introducing NLC layer between dipole structure array and metal floor. Under action of electric field, orientation molecules will be deflected. adjust resonant frequency reflected electromagnetic (EM) wave tuning permittivity. transmission characteristics LC parameters are calculated analyzed for EM waves within range from 85 to 115 GHz. LC-based device...

We propose a new GPU based technique to render realistic rain in real-time. It consists of two parts: off-line image analysis videos, and real-time particle-based synthesis rain. Videos real are analyzed extract the mattes; random samples these stroke mattes then used for online synthesis. incorporate pre-computed radiance transfer (PRT) our particle system take into account scene radiance. The function raindrop can be highly efficiently evaluated with closed-form solution. Our approach...

Many factors affect the predicted value of power load to varying degrees. The introduction market mechanisms in reform has brought more uncertainty forecasting. forecast under new situation electricity needs consider impact on paper improves SVR algorithm and proposes a mid-long-term forecasting method reform. simulation shows that will develop rapidly

With the dense access of Large-scale high voltage direct current (HVDC), problem AC and HVDC interaction is prominent. To ensure safe stable operation AC/DC hybrid power grid, it necessary to accurately simulate transient steady state according actual control program. Firstly, main circuit transmission system its parameters are introduced. Secondly, principle structure model introduced in detail. Finally, PSCAD component library E-TRAN combined. The which can DC response established by...

In this paper, we consider the problem of state-estimation in presence Denial-of-Service (DoS) attack. We formulate as an state estimation for a plant with switching measured outputs. absence attack, state-estimator has access to all outputs, however, only subset measurements are made available state-estimator. seek find that results minimum error worst-case attack strategy. First, parameterize set state-estimators result stable error. Then, will show any can be written generalized...

Abstract Photovoltaic output has strong randomness and volatility. Research on the scenario reduction method of PV is great significance for planning, design, operation power grids. This paper proposes Nonlinear Brain Storm Optimizer (NBSO) to improve k-means algorithm, uses Standard Deviation Based Index (STDI) evaluate effectiveness clustering. The simulation results show that clustering proposed in this can effectively reduce scenes, generated typical scene set fit volatility diversity...

Abstract As a flexible unit that can respond quickly, CSP stations be included in the planned power source of supply planning, which effectively improve consumption volatile new energy and resource utilization. This paper mainly studies optimal planning grids containing fluctuating sources proposes joint based on bi-level programming model. The results calculation examples show proposed model considering is better terms economy, efficiency, practicality great importance to improvement...

This paper considers the security of distributed Cyber-Physical Systems (CPS) under malicious attacks, where CPS components interact over a communication network. We consider deception attacks both at individual component level as well network level. At level, false-data-injection corrupts sensor measurements and/or actuator signals. While between may be altered by adversarial false-data. develop robust model-based detection and identification algorithm for class discrete Linear Time...

To overcome the dimensionality disaster and feature redundancy problem, caused Polarization dimensions increasing, in deep learning network training for landcover classification using polarimetric SAR image, a novel image architecture DeepLabV3 after twice reduction with wavelet fusion PCA is proposed. The three-layer decomposition processing are implemented before to reduce dimension. full-polarization data of Gaofen-3 Baihu Farm area was used verification. Compared traditional method,...

The title compound, C12H12O4, was synthesized by the reaction of 1,3-benzodioxole-5-carbaldehyde with diethoxy­phospho­rylacetic acid ethyl ester in presence sodium hydride tetra­hydro­furan. 1.3-benzodioxolane system is planar and subtends an angle 0.82 (8)° acrylate unit.