Air pollution forecasting can provide reliable information about the future situation, which is useful for an efficient operation of air control and helps to plan prevention. Dynamics are usually reflected by various factors, such as temperature, humidity, wind direction, speed, snowfall, rainfall, so on, increase difficulty in understanding change pollutant concentration. In this paper, a short-term model based on deep learning proposed PM2.5 (particulate matter with aerodynamic diameter...

Abstract Niobium pentoxides (Nb 2 O 5 ) have attracted extensive interest for ultrafast lithium‐ion batteries due to their impressive rate/capacity performance and high safety as intercalation anodes. However, the intrinsic insulating properties unrevealed mechanisms of complex phases limit further applications. Here, a facile efficient method is developed construct three typical carbon‐confined Nb (TT‐Nb @C, T‐Nb H‐Nb @C) nanoparticles via mismatched coordination reaction during...

Abstract The access of large‐scale distributed generation (DG) easily leads to energy imbalance in distribution network. To deal with this issue, paper proposes an optimal schedule method for network considering the participation source‐load‐storage aggregation groups (SAGs). Firstly, system model consisting layer and SAGs is established, objectives constraints each are also given. Secondly, fluctuation on load side, a forecasting based Adaboost integrated convolutional neural networks...

The practice of ultra-short-term power load forecasting serves as a critical strategy for enabling rapid response and real-time dispatch in systems. By improving the accuracy forecasting, both safety systems efficiency electricity usage can be significantly enhanced. Addressing challenges posed by non-linear temporal characteristics grid data, this study introduces novel model, integrating Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory networks (BiLSTM), an...

In this paper, a flexible voltage control strategy, which takes good use of the distributed energy storage (DES) units, is proposed to enhance stability and robustness dc distribution network. The characteristics ac/dc interface in network are analyzed, virtual inertia capacitance given demonstrate interactive influence ac systems. strategy for DES located at microgrid or terminal bus designed based on characteristics, enabling respond both variation frequency change utility grid. A...

This letter proposes a two-layer active disturbance rejection control (ADRC) method with the compensation of estimated equivalent input disturbances (EID) for load frequency (LFC) multi-area interconnected power system. has capability rejecting effects random variations and parameter uncertainties, guaranteeing high dynamic performance. The state-space model LFC system is established. Then full-order generalized state observer (GSO) employed in ADRC to estimate EID. A stability condition...

A spacer is a challenging part of high voltage direct current (HVDC) gas-insulated substations and transmission lines. This paper proposes novel design HVDC spacers by self-adaptively controlling the surface charge. The briefly reviews dielectric modification methods that aim to limit charge accumulation. introduced based on concept adaptively charges using nonlinear materials, its control regulation abilities are investigated. accumulation decay properties compared downsized spacer. results...

In the islanded microgrid, distributed generators are controlled with virtual synchronous generator (VSG) strategy to simulate rotor inertia and droop characteristics of generators, in order enhance voltage frequency support capabilities. Since capacity location distribution each VSG is random, output impedance, line impedance its mismatched, resulting inaccurate sharing reactive power. Based on study power schemes without communication system parameters detection, aiming at contradiction...

This research proposes a hybrid model that combines the convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) to accurately estimate energy output of photovoltaic system. Firstly, Pearson correlation analysis is introduced screen out meteorological factors with high (PV) power output. Then, network-bidirectional (CNN-BiLSTM) combined algorithm used extract characteristics influencing by CNN, BiLSTM for timing prediction. Last but not least, using simulation...

Energy storage systems will play a key role in the power system of twenty first century considering large penetrations variable renewable energy, growth transport electrification and decentralisation heating loads. Therefore reliable real time methods to optimise energy storage, demand response generation are vital for operations. This paper presents concise review battery an example modelling applications second details adaptive approach solve this load levelling problem with storage. A...

In this paper, the hardware and software design implementation are proposed for wide-area damping control (WADC) strategies. First, hardware-in-the-loop (HIL) real-time (RT) simulation system based on RT-Lab is established as closed-loop test platform WADC applied in multi-machine interconnected power system. The hardware, which contains microcontroller, peripheral analog to digital converter, developed AVR 32-bit development board. Three algorithms, i.e., phase-compensation, delay-dependent...

The cascaded H-bridge (CHB) converter is a favorable candidate in the photovoltaic (PV) field due to its modular multilevel construction, which could pursue maximum energy yield by adopting power point tracking (MPPT) strategy. However, with increasing penetration of PV generation, challenges are arising for grid stability deficiency inertia and damping. To improve system besides efficiency, this article proposes control strategy CHB combination MPPT virtual synchronous generator (VSG)....

To enhance the accuracy of theft detection for electricity consumers, this paper introduces a novel strategy based on fusion dual-time feature and deep learning methods. Initially, considering electricity-consumption features at dual temporal scales, employs convolutional networks (TCN) with long short-term memory (LSTM) multi-level extraction module (LSTM-TCN) neural network (DCNN) to parallelly extract these scales. Subsequently, extracted are coupled input into fully connected (FC) layer...

Voltage stability draws more and attentions in dc systems due to the high penetration of constant power loads. This paper gives comprehensive analysis for voltage proposes a sliding-mode control method enhance margin systems. First, by impedance-based approach, three-phase rectifier serving as source converter is examined when interacted with tightly regulated load converter. The dynamic input characteristics side are investigated find relationship between negative incremental impedance...

Abstract Betweenness centrality is an indicator of a node’s in network. It equal to the number shortest paths from all vertices others that pass through node. Most real-world large networks display hierarchical community structure, and their betweenness computation possesses rather high complexity. Here we propose new decomposition approach speed up complex networks. The advantage this method its effective utilization local structural information community. presented can significantly...

Distributed power generators in islanded microgrid usually adopt droop control strategy or virtual synchronous generator (VSG) to simulate Q-U characteristic of for voltage support ability enhancement and reliability improvement. However, because the distributed generators' capacities locations are random, they cannot share reactive load proportional their rated capacity, resulting overload protection, even system instability. In order improve sharing precision, stability simultaneously, an...

With increasing penetration of renewable energy sources into modern power systems, multiparallel inverters are commonly employed in the interface to utility grid, giving rise potential resonance problems. In order better understand issue, is analyzed under an asynchronous carrier. First, inverter equivalent circuits modeled within a controlled bandwidth. investigate characteristics carrier, crossover filter function used separate output currents different frequencies for two-parallel analyze...

With the increasing contribution of power production by photovoltaic (PV) systems to electricity supply, PV forecasting becomes increasingly important. There are many factors influencing performance, such as air temperature, humidity, insolation, wind speed, direction and so on. This study proposes a Takagi–Sugeno (T–S) fuzzy model‐based short‐term approach. First, means correlation analysis, influential selected model inputs. Then, c‐mean clustering algorithm recursive least squares method...

Tourism development in ecologically vulnerable areas like the lake Baikal region Eastern Siberia is a challenging problem. To this end, dynamical models of AC/DC hybrid isolated power system consisting four grids with renewable generation units and energy storage systems are proposed using advanced methods based on deep reinforcement learning integral equations. First, wind solar irradiance potential several sites Baikal’s banks analyzed as well electric load function climatic conditions....

Automatic battery equalizers require no sensing circuits, which reduces their cost, size, and complexity. However, among existing methods, additional resistors or diodes are used in the balancing paths to ensure safe operation, degrades conversion efficiency speed. To overcome this problem, article proposes an equalizer architecture based on half-bridge LLC converter. The inherent current limitation characteristic of proposed improves High is achieved due soft-switching operations switches....