An estimate of the net direction climate interactions in different geographical regions is made by constructing a directed network from regular latitude-longitude grid nodes, using directionality index (DI) based on conditional mutual information (CMI). Two datasets surface air temperature anomalies—one monthly averaged and another daily averaged—are analyzed compared. The links are interpreted terms known atmospheric tropical extra-tropical variability patterns. Specific relevant selected,...

Abstract. The fact that the climate on earth is a highly complex dynamical system well-known. In last few decades great deal of effort has been focused understanding how phenomena in one geographical region affects other regions. Complex networks are powerful framework for identifying interdependencies. To further exploit knowledge links uncovered via network analysis (for, e.g., improvements prediction), good physical mechanisms underlying these required. Here we focus role atmospheric...

This work aims to optimize the overall performance of a model oscillator, as an energy harvester Lévy-like mesoscopic fluctuations through piezoelectric conversion. As further step in description realistic harvesting device we consider monostable Woods-Saxon which can interpolate between square well and harmonic-like behaviors. We study interplay potential shape noise's spectrum statistics. The dependence power output on parameters determining those features indicates directions former be increased.

We introduce a software generator for class of colored (self-correlated) and non-Gaussian noise, whose statistics spectrum depend on two parameters, $q$ $\tau$. Inspired by Tsallis' nonextensive formulation statistical physics, the so-called $q$-distribution is handy source self-correlated noise large range applications. The $q$-noise -- which tends smoothly $q=1$ to Ornstein--Uhlenbeck with autocorrelation $\tau$ generated via stochastic differential equation, using Heun method (a second...

During the last decade techniques of complex network analysis have found application in climate research. The main idea consists embedding characteristics variables, e.g., temperature, pressure or rainfall, into topology networks by appropriate linear and nonlinear measures. Applying such measures on time series leads to defining links between their corresponding locations studied region, whereas are network's nodes. resulted networks, consequently, analysed using various tools present...

Aiming to optimize piezoelectric energy harvesting from strongly col-ored fat-tailed fluctuations, we have recently studied the performance ofa monostable inertial device under a noise whose statistics depends on aparameter q (bounded for < 1, Gaussian = > 1).We interplay between potential shape (interpolatingbetween square-well and harmonic-like behaviors) noise’s statis-tics spectrum, showed that its output power grows as increasesabove 1. We now report real experiment an electronic...

We introduce a software generator for class of colored (self-correlated) and non-Gaussian noise, whose statistics spectrum depend on two param- eters, q τ. Inspired by Tsallis’ nonextensive formulation statistical physics, the so-called q-distribution is handy source self-correlated noise large range applications. The q-noise—which tends smoothly = 1 to Ornstein–Uhlenbeck with autocorrelation τ—is generated via stochastic differential equation, using Heun method (a second order Runge–Kutta...

The forecasting of large ramps in wind power output known as ramp events is crucial for the incorporation volumes energy into national electricity grids. Large variations supply must be compensated by ancillary sources which can include use fossil fuels. Improved prediction will help to reduce dependency on supplemental along with their associated costs and emissions. In this paper, we discuss limitations current predictive practices explore Machine Learning methods enhance event...

Energy landscapes are a useful aid for the understanding of dynamical systems, and valuable tool their analysis. For broad class rate models neural networks, we derive global Lyapunov function which provides an energy landscape without any symmetry constraint. This newly obtained `nonequilibrium potential' (NEP) predicts with high accuracy outcomes dynamics in globally stable cases studied here. Common features this bistability --with implications working memory slow oscillations --and...

A brief review is made of the birth and evolution "nonequilibrium potential" (NEP) concept.As if providing a landscape for qualitative reasoning were not helpful enough, NEP adds aquantitative dimension to theory differential equations provides globalLyapunov function deterministic dynamics. Here we illustrate usefulness todraw results on stochastic thermodynamics: Jarzynski equality in Wilson–Cowan model (apopulation-competition neocortex) "thermodynamic uncertainty relation" (TUR)in KPZ...

We introduce the first automated models for classifying natural language descriptions provided in cost documents called "Bills of Quantities" (BoQs) popular infrastructure construction industry, into International Construction Measurement Standard (ICMS). The we deployed and systematically evaluated multi-class text classification are learnt from a dataset more than 50 thousand items retrieved 24 large projects across United Kingdom. describe our approach to representation subsequent...