Leveraging both temporal and spatial correlations to predict wind speed remains one of the most challenging less studied areas prediction. In this paper, problem predicting speeds for multiple sites is investigated by using spatio-temporal correlation. We proposed a deep architecture termed predictive network (PSTN), which unified framework integrating convolutional neural (CNN) long short-term memory (LSTM). Initially, features are extracted from matrices CNN at bottom model. Then, LSTM...

Phasor measurement unit (PMU) data manipulation attacks (PDMAs) may blind the control centers to real-time operating conditions of power systems. Detecting these accurately is essential ensure normal operation system monitoring and control. Using long-term accumulated historical PMU measurements train a machine learning model detect PDMAs has shown promising results. In this paper, deepautoencoder-based anomaly measurers are deployed throughout build distributed PDMA detection framework. The...

A probabilistic load flow (PLF) method using Copula and improved Latin hypercube sampling is proposed. The stochastic dependence between input random variables considered. theory adopted to establish the probability distribution of correlated variables. Based on discrete data, an accuracy established by evaluated adopting power output wind farms located at New Jersey. performance proposed PLF investigated IEEE 14‐bus 118‐bus test systems.

The voltage instability phenomena may occur in distribution networks. decline of stability level will restrict the increase load served by companies. Distributed generation (DG) is increasingly drawing great attention people. development DGs bring new chances to traditional power systems. connected networks are potential improve system stability. In this paper, familiar interfaces between and introduced. operation modes control characteristics typical DGs, including asynchronous generators,...

This paper proposes a multiagent-based adaptive wide-area current differential protection system. A power grid is divided into primary and backup zones dynamically online with the help of developed expert system, relay agents these cooperate to perform relaying function. Predictive self-healing strategies have been proposed prevent misoperation agent, communication system has simulated. The performance based on computer simulation presented in this paper.

Accurate transient stability assessment (TSA) is a fundamental requirement for ensuring secure and stable operation of power systems. Tremendous efforts have been made to apply artificial intelligence approaches TSA with phasor measurement unit data. However, many previous may be failed provide favorable accuracy due the shallow architectures error-prone hand-crafting features. This paper proposed model TSA, which termed multi-branch stacked denoising autoencoder (MSDAE). unified framework...

A probabilistic load flow method based on polynomial normal transformation (PNT) and Latin hypercube sampling (LHS) is proposed. The correlation between input random variables has been taken into consideration. proposed uses the statistical moments matrix of instead their marginal distribution functions joint functions, which are very difficult to be obtained, establish probability models by PNT LHS. node voltage line calculated Monte Carlo simulation method. Performance investigated using...

Due to the lack of standard information model and description, configuration communication network is still out unified engineering process substation automation, which may result in incorrect long time. The IEC 61850-based comprehensive its description language (SCL) files are proposed for standard, correct, convenient network. established by means extraction with function decomposition, mapping management base (MIB). detailed achieved extending existing SCL 61850-6. case study a typical...

One of the significant impacts growing penetration intermittent renewable energy sources is upon frequency response power system. Compared with dispersive electric vehicle storage, battery swapping station (BSS), as an emerging form can provide a more reliable supplementary regulation service for control. This study has proposed new automatic generation control (AGC) strategy using controllable storage in BSSs, referred to station‐to‐grid (S2G). A Monte–Carlo stochastic simulation method...

Leveraging multiple heterogeneous measurements to predict wind power has long been a challenging task in the electrical community. In this paper, deep architecture incorporated with multitask learning and multimodal for prediction, termed predictive stacked autoencoder (PSAE), is presented. PSAE unified framework integrating autoencoders (SAEs), one feature fusion layer, prediction terminal which expands from two spatial dimensions, including depth width, compared conventional models....

With the increasing integration of wind farms, modification current tools for evaluating and managing power systems such as available transfer capability (ATC) becomes an important issue. This study presents a computationally accurate efficient method in ATC with large amount uncertainty based on Latin hypercube sampling (LHS) scenario clustering techniques. LHS is used Monte Carlo simulation to select system state high efficiency good precision. Cholesky decomposition combined into process...

This paper presents an evaluation method of strength measure for the static voltage stability analysis hybrid multi-infeed dc (HMIdc) system, which comprises both line-commutated converter-based high (LCC-HVdc) and source HVdc (VSC-HVdc) links. For this purpose, coupled single-port model proposed originally to assess ac power systems is first extended in HMIdc system. In particular, how LCC-HVdc VSC-HVdc links should be modeled further investigated. Then, a generalized effective...

Flexible intelligent electronic devices (IEDs) are highly desirable to support free allocation of function IED by means software reconfiguration without any change hardware. The application generic hardware platforms and component-based technology seems be a good solution. Due the advent IEC 61850, with standard communication interface can used implement different kinds functions high flexibility. remaining challenge is unified model that specifies various components appropriate granularity...

Taking transmission sections as the monitoring objects of power system security and stability level can largely improve efficiency analysis control in operation. However, existing approaches for identifying mainly depend on years experience, which is not suitable complicated variable systems with large scales. Thus, a novel method automatic identification using complex network theory proposed. The proposed presents fundamental conditions identifies them from three levels: lines, key links...

Searching for the Extreme Operating Conditions (EOCs) is one of core problems power system relay protection setting calculation. The current methods based on brute-force search, heuristic algorithms, and mathematical programming can hardly meet requirements today's systems in terms computation speed due to drastic changes operating conditions induced by renewables electronics. This paper proposes an EOC fast search method, named Graph Dueling Double Deep Q Network (Graph D3QN), which...

Laver fluffy is an indispensable link in the processing of laver products. After fluffing, acquires appealing color, which conducive to better marketability. During primary mechanical laver, a valve capable rapid opening and closing required ensure that laver’s surface becomes lustrous post-processing. However, products can meet specific requirements fluffing are scarce. This study proposes novel principle for high-speed switching control valve. quickly turn on or cut off high-pressure gas...