arXiv subjects

Mikhail Kanevski

Publications and source records attributed to Mikhail Kanevski.

At least 19 recordsLinked to original sources

arXiv2022

Spatio-temporal estimation of wind speed and wind power using machine learning: predictions, uncertainty and technical potential

The growth of wind generation capacities in the past decades has shown that wind energy can contribute to the energy transition in many parts of the world. Being highly variable and complex to model, the quantification of the spatio-temporal variation of wind power and the related uncertainty is highly relevant for energy planners. Machine Learning has become a popular tool to perform wind-speed and power predictions. However, the existing approaches have several limitations. These include (i) insufficient consideration of spatio-temporal correlations in wind-speed data, (ii) a lack of existing methodologies to quantify the uncertainty of wind speed prediction and its propagation to the wind-power estimation, and (iii) a focus on less than hourly frequencies. To overcome these limitations, we introduce a framework to reconstruct a spatio-temporal field on a regular grid from irregularly distributed wind-speed measurements. After decomposing data into temporally referenced basis functions and their corresponding spatially distributed coefficients, the latter are spatially modelled using Extreme Learning Machines. Estimates of both model and prediction uncertainties, and of their propagation after the transformation of wind speed into wind power, are then provided without any assumptions on distribution patterns of the data. The methodology is applied to the study of hourly wind power potential on a grid of 250 by 250 squared meters for turbines of 100 meters hub height in Switzerland, generating the first dataset of its type for the country. The potential wind power generation is combined with the available area for wind turbine installations to yield an estimate of the technical potential for wind power in Switzerland. The wind power estimate presented here represents an important input for planners to support the design of future energy systems with increased wind power generation.

eess.SP↗

arXiv2021

A survey of active learning algorithms for supervised remote sensing image classification

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

cs.CV↗

arXiv2021

Spatio-temporal evolution of global surface temperature distributions

Climate is known for being characterised by strong non-linearity and chaotic behaviour. Nevertheless, few studies in climate science adopt statistical methods specifically designed for non-stationary or non-linear systems. Here we show how the use of statistical methods from Information Theory can describe the non-stationary behaviour of climate fields, unveiling spatial and temporal patterns that may otherwise be difficult to recognize. We study the maximum temperature at two meters above ground using the NCEP CDAS1 daily reanalysis data, with a spatial resolution of 2.5 by 2.5 degree and covering the time period from 1 January 1948 to 30 November 2018. The spatial and temporal evolution of the temperature time series are retrieved using the Fisher Information Measure, which quantifies the information in a signal, and the Shannon Entropy Power, which is a measure of its uncertainty -- or unpredictability. The results describe the temporal behaviour of the analysed variable. Our findings suggest that tropical and temperate zones are now characterized by higher levels of entropy. Finally, Fisher-Shannon Complexity is introduced and applied to study the evolution of the daily maximum surface temperature distributions.

stat.AP↗

arXiv2021

Advanced analysis of temporal data using Fisher-Shannon information: theoretical development and application in geosciences

Complex non-linear time series are ubiquitous in geosciences. Quantifying complexity and non-stationarity of these data is a challenging task, and advanced complexity-based exploratory tool are required for understanding and visualizing such data. This paper discusses the Fisher-Shannon method, from which one can obtain a complexity measure and detect non-stationarity, as an efficient data exploration tool. The state-of-the-art studies related to the Fisher-Shannon measures are collected, and new analytical formulas for positive unimodal skewed distributions are proposed. Case studies on both synthetic and real data illustrate the usefulness of the Fisher-Shannon method, which can find application in different domains including time series discrimination and generation of times series features for clustering, modeling and forecasting. The paper is accompanied with Python and R libraries for the non-parametric estimation of the proposed measures.

stat.ME↗

arXiv2020

A Novel Framework for Spatio-Temporal Prediction of Environmental Data Using Deep Learning

As the role played by statistical and computational sciences in climate and environmental modelling and prediction becomes more important, Machine Learning researchers are becoming more aware of the relevance of their work to help tackle the climate crisis. Indeed, being universal nonlinear function approximation tools, Machine Learning algorithms are efficient in analysing and modelling spatially and temporally variable environmental data. While Deep Learning models have proved to be able to capture spatial, temporal, and spatio-temporal dependencies through their automatic feature representation learning, the problem of the interpolation of continuous spatio-temporal fields measured on a set of irregular points in space is still under-investigated. To fill this gap, we introduce here a framework for spatio-temporal prediction of climate and environmental data using deep learning. Specifically, we show how spatio-temporal processes can be decomposed in terms of a sum of products of temporally referenced basis functions, and of stochastic spatial coefficients which can be spatially modelled and mapped on a regular grid, allowing the reconstruction of the complete spatio-temporal signal. Applications on two case studies based on simulated and real-world data will show the effectiveness of the proposed framework in modelling coherent spatio-temporal fields.

stat.ML↗

arXiv2020

Uncertainty Quantification in Extreme Learning Machine: Analytical Developments, Variance Estimates and Confidence Intervals

Uncertainty quantification is crucial to assess prediction quality of a machine learning model. In the case of Extreme Learning Machines (ELM), most methods proposed in the literature make strong assumptions on the data, ignore the randomness of input weights or neglect the bias contribution in confidence interval estimations. This paper presents novel estimations that overcome these constraints and improve the understanding of ELM variability. Analytical derivations are provided under general assumptions, supporting the identification and the interpretation of the contribution of different variability sources. Under both homoskedasticity and heteroskedasticity, several variance estimates are proposed, investigated, and numerically tested, showing their effectiveness in replicating the expected variance behaviours. Finally, the feasibility of confidence intervals estimation is discussed by adopting a critical approach, hence raising the awareness of ELM users concerning some of their pitfalls. The paper is accompanied with a scikit-learn compatible Python library enabling efficient computation of all estimates discussed herein.

stat.ML↗

arXiv2020

On Feature Selection Using Anisotropic General Regression Neural Network

The presence of irrelevant features in the input dataset tends to reduce the interpretability and predictive quality of machine learning models. Therefore, the development of feature selection methods to recognize irrelevant features is a crucial topic in machine learning. Here we show how the General Regression Neural Network used with an anisotropic Gaussian Kernel can be used to perform feature selection. A number of numerical experiments are conducted using simulated data to study the robustness of the proposed methodology and its sensitivity to sample size. Finally, a comparison with four other feature selection methods is performed on several real world datasets.

stat.ML↗

arXiv2019

Analysis of air pollution time series using complexity-invariant distance and information measures

Air pollution is known to be a major threat for human and ecosystem health. A proper understanding of the factors generating pollution and of the behavior of air pollution in time is crucial to support the development of effective policies aiming at the reduction of pollutant concentration. This paper considers the hourly time series of three pollutants, namely NO$_2$, O$_3$ and PM$_{2.5}$, collected on sixteen measurement stations in Switzerland. The air pollution patterns due to the location of measurement stations and their relationship with anthropogenic activities, and specifically land use, are studied using two approaches: Fisher-Shannon information plane and complexity-invariant distance between time series. A clustering analysis is used to recognize within the measurements of a same pollutant group of stations behaving in a similar way. The results clearly demonstrate the relationship between the air pollution probability densities and land use activities.

stat.AP↗

arXiv2019

Investigating the time dynamics of wind speed in complex terrains by using the Fisher-Shannon method

In this paper, the time dynamics of the daily means of wind speed measured in complex mountainous regions are investigated. For 293 measuring stations distributed over all Switzerland, the Fisher information measure and the Shannon entropy power are calculated. The results reveal a clear relationship between the computed measures and both the elevation of the wind stations and the slope of the measuring sites. In particular, the Shannon entropy power and the Fisher information measure have their highest (respectively lowest) values in the Alps, where the time dynamics of wind speed follows a more disordered pattern. The spatial mapping of the calculated quantities allows the identification of two regions, which is in agreement with the topography of the Swiss territory. The present study could contribute to a better characterization of the temporal dynamics of wind speed in complex mountainous terrain.

stat.AP↗

arXiv2018

Fisher-Shannon complexity analysis of high-frequency urban wind speed time series

1Hz wind time series recorded at different levels (from 1.5 to 25.5 meters) in an urban area are investigated by using the Fisher-Shannon (FS) analysis. FS analysis is a well known method to get insight of the complex behavior of nonlinear systems, by quantifying the order/disorder properties of time series. Our findings reveal that the FS complexity, defined as the product between the Fisher Information Measure and the Shannon entropy power, decreases with the height of the anemometer from the ground, suggesting a height-dependent variability in the order/disorder features of the high frequency wind speed measured in urban layouts. Furthermore, the correlation between the FS complexity of wind speed and the daily variance of the ambient temperature shows similar decrease with the height of the wind sensor. Such correlation is larger for the lower anemometers, indicating that ambient temperature is an important forcing of the wind speed variability in the vicinity of the ground.

stat.AP↗

arXiv2018

Wavelet variance scale-dependence as a dynamics discriminating tool in high-frequency urban wind speed time series

High frequency wind time series measured at different heights from the ground (from 1.5 to 25.5 meters) in an urban area were investigated by using the variance of the coefficients of their wavelet transform. Two ranges of scales were identified, sensitive to two different dynamical behavior of the wind speed: the lower anemometers show higher wavelet variance at smaller scales, while the higher ones are characterized by higher wavelet variance at larger scales. Due to the relationship between wavelet scale and frequency, the results suggest the existence of two frequency ranges, where the wind speed variability change according to the position of the anemometer from the ground. This study contributes to better understanding of the high frequency wind speed in urban areas and to a better knowledge of the underlying mechanism governing the wind fluctuations at different heights from the ground in particular in urban area.

stat.AP↗

arXiv2018

Community detection analysis in wind speed-monitoring systems using mutual information-based complex network

A mutual information-based weighted network representation of a wide wind speed monitoring system in Switzerland was analysed in order to detect communities. Two communities have been revealed, corresponding to two clusters of sensors situated respectively on the Alps and on the Jura-Plateau that define the two major climatic zones of Switzerland. The silhouette measure is used to evaluate the obtained communities and confirm the membership of each sensor to its cluster.

physics.soc-ph↗

arXiv2018

Analysis of temporal properties of wind extremes

The 10-minute average wind speed series recorded at 132 stations distributed rather homogeneously in the territory of Switzerland are investigated. Wind extremes are defined on the base of run theory: fixing a percentile-based threshold of the wind speed distribution, a wind extreme is defined as a sequence of consecutive wind values (or duration of the extreme) above the threshold. This definition allows to analyse the sequence of extremes as a temporal point process marked by the duration of the extremes. The average probability density function of the duration of the extremes of the wind speed measured in Switzerland does not depend on the percentile-based threshold and decrease with the increase of the extreme duration. The time-clustering behaviour of the sequences of the wind extremes was analysed by using the global and local coefficient of variation and the Allan Factor. The wind extremes are globally time-clustered, although they tend to behave as a Poisson process with the increase of the minimum extreme duration. Locally, the wind extremes tend to be clustered for any percentile-based threshold for stations located above about 2,000 m a.s.l. By using the Allan Factor, it was revealed that wind extremes tend to be clustered even at lower timescales especially for the higher stations.

stat.AP↗

arXiv2018

Linearity versus non-linearity in high frequency multilevel wind time series measured in urban areas

In this paper, high frequency wind time series measured at different heights from the ground (from 5.5 to 25.5 meters) in an urban area were investigated. The spectrum of each series is characterized by a power-law behaviour at low frequency range, with a mean spectral exponent of about 1.5, which is rather consistent with the Kolmogorov spectrum of atmospheric turbulence. The detrended fluctuation analysis was applied on the magnitude and sign series of the increments of wind speed, in order to get information about the linear and nonlinear dynamics of the time series. Both the sign series and magnitude series are characterized by two timescale ranges; in particular the scaling exponent of the magnitude series in the high timescale range seems to be related with the height of the sensor. This study aims to understand better high frequency wind speed in urban areas and to disclose the underlying mechanism governing the wind fluctuations at different heights.

stat.AP↗

arXiv2018

Long-range fluctuations and multifractality in connectivity density time series of a wind speed monitoring network

This paper studies the daily connectivity time series of a wind speed-monitoring network using multifractal detrended fluctuation analysis. It investigates the long-range fluctuation and multifractality in the residuals of the connectivity time series. Our findings reveal that the daily connectivity of the correlation-based network is persistent for any correlation threshold. Further, the multifractality degree is higher for larger absolute values of the correlation threshold

physics.data-an↗

arXiv2017

Periodic fluctuations in correlation-based connectivity density time series: application to wind speed-monitoring network in Switzerland

In this paper, we study the periodic fluctuations of connectivity density time series of a wind speed-monitoring network in Switzerland. By using the correlogram-based robust periodogram annual periodic oscillations were found in the correlation-based network. The intensity of such annual periodic oscillations is larger for lower correlation thresholds and smaller for higher. The annual periodicity in the connectivity density seems reasonably consistent with the seasonal meteo-climatic cycle.

stat.AP↗

arXiv2017

Multifractal analysis of the time series of daily means of wind speed in complex regions

In this paper, we applied the multifractal detrended fluctuation analysis to the daily means of wind speed measured by 119 weather stations distributed over the territory of Switzerland. The analysis was focused on the inner time fluctuations of wind speed, which could be more linked with the local conditions of the highly varying topography of Switzerland. Our findings point out to a persistent behaviour of all the measured wind speed series (indicated by a Hurst exponent significantly larger than 0.5), and to a high multifractality degree indicating a relative dominance of the large fluctuations in the dynamics of wind speed, especially in the Swiss plateau, which is comprised between the Jura and Alp mountain ranges. The study represents a contribution to the understanding of the dynamical mechanisms of wind speed variability in mountainous regions.

stat.AP↗

arXiv2017

Unsupervised Feature Selection Based on Space Filling Concept

The paper deals with the adaptation of a new measure for the unsupervised feature selection problems. The proposed measure is based on space filling concept and is called the coverage measure. This measure was used for judging the quality of an experimental space filling design. In the present work, the coverage measure is adapted for selecting the smallest informative subset of variables by reducing redundancy in data. This paper proposes a simple analogy to apply this measure. It is implemented in a filter algorithm for unsupervised feature selection problems. The proposed filter algorithm is robust with high dimensional data and can be implemented without extra parameters. Further, it is tested with simulated data and real world case studies including environmental data and hyperspectral image. Finally, the results are evaluated by using random forest algorithm.

stat.ML↗