Accurate prediction of CB2 receptor ligand activity is pivotal for advancing drug discovery targeting this receptor, which implicated in inflammation, pain management, and neurodegenerative conditions. Although conventional machine learning deep techniques have shown promise, their limited interpretability remains a significant barrier to rational design. In work, we introduce CB2former, framework that combines Graph Convolutional Network with Transformer architecture predict activity. By...

Wind power output prediction is crucial to the balanced dispatch and market operation of system, accurate wind can guarantee security stability grid operation. In order realize highly support low-carbon transformation a method integrating convolutional neural network (CNN), bi-directional long short-term memory (BiLSTM) Attention mechanism proposed. To address nonlinear temporal characteristics data, CNN layer extracts local features, BiLSTM captures bidirectional dependencies, reduces...

The well-developed unified power flow controller (UPFC) has demonstrated its capability in providing voltage support and improving system stability. objective of this paper is to demonstrate the UPFC mitigating oscillations a wind farm integrated by employing eigenvalue analysis dynamic time-domain simulation approaches. For purpose, oscillation damping (PODC) designed for caused disturbances given interconnected system, including change tie-line power, changes outputs, others. Simulations...

Extreme events have brought horrible damage to power distribution systems and transportation systems. Generally speaking, post-disaster recovery relies on emergency resources, such as repair crews (RCs). However, due geographical interdependency, the dispatch of RCs is severely restricted by traffic conditions in To this end, article proposes a mixed-integer linear programming model synergistically Geographical interdependency modeled guarantee feasible practical routing schedules. A novel...

In order to solve the coordinated optimization problem of distributed generation (DG), on-load tap changer (OLTC) and capacitor banks, a dynamic reactive power (DRPO) method based on fuzzy time clustering is proposed. method, algorithm used cluster static optimal switching sequence OLTC banks. The decoupling control variables realized it meets constraints maximum times. Next, influencing factors DG limit are analysed. Considering ability OLTC, banks in voltage optimization, Since adjustment...

In the restructured electricity market, retailers are intermediaries between wholesale market and consumers. Considering uncertainty of price, should consider risks their profit caused by uncertain price when participating in retail competition. Indeed, retailers’ risk preferences will impact bidding strategies. To examine effects on strategies equilibrium outcomes an model for price-making is proposed employing mean–variance utility theory to retailers. The share function used characterize...

The output of renewable energy is strongly uncertain and random, the distribution voltage reactive power in regional grids changed with access to large-scale energy.In order quantitatively evaluate influence on operation, a novel combinational evaluation method containing proposed.Firstly, actual operation data load demand are clustered based K-means algorithm, several typical scenarios divided.Then, entropy weight (EWM) analytic hierarchy process (AHP) combined qualified rate, fluctuation,...

This paper proposed an entropy weight optimum seeking method (EWOSM) based on the typical scenarios partitioning and load distribution matching, to solve reactive power optimization problem in network under background of big data. Firstly, mathematic model is provided analyze relationship between data source schemes network, which illustrate feasibility using large amount historical optimization. Then, matching are presented, can select out some loads that have same or similar distributions...

Abstract To optimally harness the adjustable capabilities of reactive power sources for voltage control, a dynamic partitioning method that uses flow tracking branch cutting through Binary Particle Swarm Optimisation (BPSO) is proposed Active Distribution Networks (ADNs). Initially, limitations existing Voltage/Var Sensitivity (VVS) calculation methods are analysed, leading to proposition novel VVS capable capturing variations in source‐load timing characteristics. Subsequently,...

Abstract: As the current calculation method of voltage/var sensitivity (VVS) could not reflect temporal variations power sources and load, paper proposes a novel VVS to partition distribution network containing renewable energy. Firstly, defects existing methods are analyzed, is proposed by adding reactive output perturbation source. Secondly, matrix fuzzified obtain membership between each node. Based on relationship source nodes, nodes with strong coupling merged determine initial number...

A novel method for improving the stability of AVR (automatic voltage regulator) constant transformer triport UPS (uninterruptible power supply) systems is introduced. An active filter with an operational amplifier as feedback element controlling shown to reduce system instability; this device more compact and economical than alternative magnetic amplifier. Experimental simulation results demonstrate effectiveness a higher order suppressing instability.< <ETX...

Background: A novel fault location method of HVDC transmission line based on a concentric relaxation principle is proposed in this paper. Methods: Due to the different position fault, instantaneous energy measured from rectifier and inverter are different, ratio k between them relationship d. Through analysis amplitude-frequency characteristics, we found that wave attenuation characteristic low frequency traveling stable, amplitude larger, so get by using low-frequency data. By wavelet...

Abstract Wind-solar hybrid generation system can significantly improve the utilization efficiency of intermittent renewable energy as it make full use complementarity wind-solar. Meanwhile, will affect low frequency characteristics power system. Effect different wind-solar accommodation conditions on oscillation are analyzed based IEEE 2-area 4-unit Eigenvalue analysis and dynamic time-domain simulation adopted to figure out influence, including factors such separate integration,...

Background: Focusing on the stability problems brought by integrated wind power and fuel cell, objective of this paper is to analyze small-signal improvement a hybrid renewable energy system connected with doubly-fed induction generators (DFIGs) solid oxide cells (SOFCs) static series synchronous compensator (SSSC) power-oscillation damper (POD). Methods: For purpose, SSSC-POD controller designed state-space representation such for modal analysis developed. Then an approach coordination...

Background: With the institutional reform of power market and need for demandside response, requirements load forecasting accuracy are getting higher higher. In order to deeply explore influencing factors improve ultra-short-term forecasting, a method analyzing considering time series characteristics is proposed in this paper. Firstly, based on analysis four different types characteristics, eight-time characteristic parameters that can characterize curve may be related prediction model...

One of the major ways delivering renewable energy generation from large-scale bases in northeast, northwest, and northern China, is to bundle power wind-photovoltaic-thermal units then transmit it by alternating current/direct current (AC/DC) systems. The interactions among wind units, photovoltaic (PV) high voltage direct (HVDC) systems make stability interconnected system more complicated. In this paper, a framework model wind-PV-thermal-bundled transmitted AC/DC constructed based on...