Consideration Of Spatial Correlation Research Articles

DESAT: A Distance-Enhanced Strip Attention Transformer for Remote Sensing Image Super-Resolution

Transformer-based methods have demonstrated impressive performance in image super-resolution tasks. However, when applied to large-scale Earth observation images, the existing transformers encounter two significant challenges: (1) insufficient consideration of spatial correlation between adjacent ground objects; and (2) performance bottlenecks due to the underutilization of the upsample module. To address these issues, we propose a novel distance-enhanced strip attention transformer (DESAT). The DESAT integrates distance priors, easily obtainable from remote sensing images, into the strip window self-attention mechanism to capture spatial correlations more effectively. To further enhance the transfer of deep features into high-resolution outputs, we designed an attention-enhanced upsample block, which combines the pixel shuffle layer with an attention-based upsample branch implemented through the overlapping window self-attention mechanism. Additionally, to better simulate real-world scenarios, we constructed a new cross-sensor super-resolution dataset using Gaofen-6 satellite imagery. Extensive experiments on both simulated and real-world remote sensing datasets demonstrate that the DESAT outperforms state-of-the-art models by up to 1.17 dB along with superior qualitative results. Furthermore, the DESAT achieves more competitive performance in real-world tasks, effectively balancing spatial detail reconstruction and spectral transform, making it highly suitable for practical remote sensing super-resolution applications.

A deep learning-based combination method of spatio-temporal prediction for regional mining surface subsidence

In coal mining areas, surface subsidence poses significant risks to human life and property. Fortunately, surface subsidence caused by coal mining can be monitored and predicted by using various methods, e.g., probability integral method and deep learning (DL) methods. Although DL methods show promise in predicting subsidence, they often lack accuracy due to insufficient consideration of spatial correlation and temporal nonlinearity. Considering this issue, we propose a novel DL-based approach for predicting mining surface subsidence. Our method employs K-means clustering to partition spatial data, allowing the application of a gate recurrent unit (GRU) model to capture nonlinear relationships in subsidence time series within each partition. Optimization using snake optimization (SO) further enhances model accuracy globally. Validation shows our method outperforms traditional Long Short-Term Memory (LSTM) and GRU models, achieving 99.1% of sample pixels with less than 8 mm absolute error.

Distributed photovoltaic short‐term power forecasting using hybrid competitive particle swarm optimization support vector machines based on spatial correlation analysis

AbstractIn order to further improve the accuracy of distributed photovoltaic (DPV) power prediction, this paper proposes a support vector machine (SVM) model based on hybrid competitive particle swarm optimization (HCPSO) with consideration of spatial correlation (SC), for realizing short‐term PV power prediction tasks. Firstly, the spatial correlation analysis is conducted on the distributed PV stations. The k‐means clustering method based on morphological similarity distance improvement and mutual information function is used to select the best reference station and the best delay, which generates strongly correlated PV sequences. Then, a hybrid algorithm of particle swarm optimization (PSO) and sine cosine algorithm (SCA) in a competitive framework (HCPSO) is proposed, aiming to fuse the fast convergence capability of PSO algorithm with the global search capability of SCA algorithm, while enabling the algorithm to effectively handle high‐dimensional optimization problems based on a competitive mechanism. Finally, the HCPSO algorithm is combined with SVM algorithm, which expands the applicable scenarios of the SVM model and effectively improves the accuracy of PV short‐term prediction.

A fuel-efficient reliable path finding algorithm in stochastic networks under spatial correlation

Transport activities are regarded as a major source of fuel consumption and CO2 emission production. To reduce the negative impact of traffic-related CO2 emission, this paper proposes a reliable path-finding algorithm for improving fuel efficiency in stochastic networks under the uncertainty of travel time and fuel consumption with the consideration of spatial correlation. A reliable constrained path-finding model is developed and formulated to minimize the fuel consumption budget while guaranteeing the specified on-time arrival probability. A heuristic label setting algorithm is developed to precisely solve the formulated problem. The proposed algorithm overcomes the time-consuming drawbacks of traditional path enumeration algorithms. The applicability and efficiency of the proposed algorithm are verified on real-world traffic data acquired from the Beijing and Xi‘an networks in China. The experiments demonstrate that our proposed algorithm can significantly reduce fuel consumption compared to existing studies. The experiment in Beijing shows that using the proposed algorithm can reduce 0.9 kg of CO2 emissions on average per trip compared to existing studies.

Making sense of dislocation correlations

Since crystal plasticity is the result of moving and interacting dislocations, it seems self-evident that continuum plasticity should in principle be derivable as a statistical continuum theory of dislocations, though in practice we are still far from doing so. One key to any statistical continuum theory of interacting particles is the consideration of spatial correlations. However, because dislocations are extended one-dimensional defects, the classical definition of correlations for point particles is not readily applicable to dislocation systems: the line-like nature of dislocations entails that a scalar pair correlation function does not suffice for characterizing spatial correlations and a hierarchy of two-point tensors is required in general. The extended nature of dislocations as closed curves leads to strong self-correlations along the dislocation line. In the current contribution, we thoroughly introduce the concept of pair correlations for general averaged dislocation systems and illustrate self-correlations as well as the content of low order correlation tensors using a simple model system. We furthermore detail how pair correlation information may be obtained from three-dimensional discrete dislocation simulations and provide a first analysis of correlations from such simulations. We briefly discuss how the pair correlation information may be employed to improve existing continuum dislocation theories and why we think it is important for analyzing discrete dislocation data.

DPCrowd: Privacy-Preserving and Communication-Efficient Decentralized Statistical Estimation for Real-Time Crowdsourced Data

In Internet of Things (IoT) driven smart-world systems, real-time crowd-sourced databases from multiple distributed servers can be aggregated to extract dynamic statistics from a larger population, thus providing more reliable knowledge for our society. Particularly, multiple distributed servers in a decentralized network can realize real-time collaborative statistical estimation by disseminating statistics from their separate databases. Despite no raw data sharing, the real-time statistics could still expose the data privacy of crowd-sourcing participants. For mitigating the privacy concern, while traditional differential privacy (DP) mechanism can be simply implemented to perturb the statistics in each timestamp and independently for each dimension, this may suffer a great utility loss from the real-time and multi-dimensional crowd-sourced data. Also, the real-time broadcasting would bring significant overheads in the whole network. To tackle the issues, we propose a novel privacy-preserving and communication-efficient decentralized statistical estimation algorithm (DPCrowd), which only requires intermittently sharing the DP protected parameters with one-hop neighbors by exploiting the temporal correlations in real-time crowd-sourced data. Then, with further consideration of spatial correlations, we develop an enhanced algorithm, DPCrowd+, to deal with multi-dimensional infinite crowd-data streams. Extensive experiments on several datasets demonstrate that our proposed schemes DPCrowd and DPCrowd+ can significantly outperform existing schemes in providing accurate and consensus estimation with rigorous privacy protection and great communication efficiency.

Random field-aided tracking of autonomous kinetically passive wireless agents

Continuous miniaturization of circuitry has open the door for various novel application scenarios of millimeter-sized wireless agents such as for the exploration of difficult-to-access fluid environments. In this context, agents are envisioned to be employed, e.g., for pipeline inspection or groundwater analysis. In either case, the demand for miniature sensors is incompatible with propulsion capabilities. Consequently, the agents are condemned to be kinetically passive and are, thus, subject to the fluid dynamics present in the environment. In these situations, the localization is complicated by the fact that unknown external forces (e.g., from the fluid) govern the motion of the agents. In this work, a comprehensive framework is presented that targets the simultaneous estimation of the external forces stemming from the fluid and the agents’ positions which are traversing the environment. More precisely, a Bayesian hierarchical model is proposed that models’ relevant characteristics of the fluid via a spatial random field and incorporates this as control input into the motion model. The random field model facilitates the consideration of spatial correlation among the agents’ trajectories and, thereby, improves the localization significantly. Additionally, this is combined with multiple particle filtering to account for the fact that within such underground fluid environments, only a localization based on distance and/or bearing measurements is feasible. In the results provided in this work, which are based on realistic computational fluid dynamics simulations, it is shown that—via the proposed spatial model—significant improvements in terms of localization accuracy can be achieved.

Statistical charts for determining sample size at various levels of accuracy and confidence in geotechnical site investigation

Site investigation is indispensable in geotechnical practice, but only a small portion of the soil affected by the intended geotechnical construction is examined and/or tested in the laboratory or in situ, leading to significant uncertainties in site investigation results. The uncertainty in the site investigation results depends greatly on the number of specimens tested (i.e. sample size) during site investigation. In many other areas that involve sampling, such as polls, surveys and quality control, sample sizes are often determined using statistical charts, based on the target levels of accuracy and confidence. However, conventional statistical charts, which only deal with independent data, are seldom used in geotechnical site investigation, because geotechnical properties are usually spatially correlated. No statistical chart with consideration of spatial correlation is available for determining sample sizes in geotechnical site investigation. Existing geotechnical design codes around the world (e.g. Eurocode 7) only provide conceptual principles on sample size determination. This study develops statistical charts for determination of sample size at different levels of both accuracy and confidence for geotechnical site investigation. Real cone penetration test data and real laboratory test data are used to illustrate application of the proposed statistical charts, and the charts are shown to perform well.

Cross Wind Effects on Vehicle–Track Interactions: A Methodology for Dynamic Model Construction

An efficient and accurate model for the dynamic assessment of vehicle-track behavior subjected to cross wind actions is developed in this paper, where the wind–vehicle–track interaction is regarded as a coupled vibration system. First, a vehicle–track interaction model is proposed by taking the hypotheses of wheel/rail rigid contact and displacement complementarity. Unlike explicit force-based methods, the vehicle-track systems are wholly coupled by interaction matrices and load vectors, which are computationally more efficient than most of the existing methods and fairly accurate in low frequency vibrations. Then, the fluctuating cross winds are simulated by the fast Fourier transform technique from spectral representations with the consideration of spatial correlation of multipoint wind time histories and vehicle movement. The unsteady cross wind forces are obtained by introducing weighting function. Finally, a modeling framework, with the coupled interactions between cross winds, vehicle, and the tracks included, is built effectively. Through the validated dynamic model, the cross wind effects on vehicle-track dynamic performance can be fully revealed. Besides, it is concluded that the dynamic performance of vehicle-track systems differs significantly in various excitation modes, i.e., average cross wind, fluctuating cross wind, and track irregularities.

An integrated approach for landslide susceptibility mapping by considering spatial correlation and fractal distribution of clustered landslide data

Natural disasters often show highly heterogeneous character due to complex geo-environmental settings. The spatial distribution of landslides is generally clustered at different scales. In this paper, we proposed a methodology for landslide susceptibility mapping (LSM) with consideration of spatial correlation and distribution of clustered landslide data. To quantify the spatial correlation of landslides, a normalized spatial-correlated scale index (NSCI) was introduced. Based on the definition of landslide frequency ratio, calibrated landslide potential index (CLPI) was proposed to account for the effect of landslide clustering. Considering the fractal distribution of landslides, the variable fractal dimension model (VFDM) was introduced to measure the spatial association between clustered landslides and conditional factors. Based on the definition of fractal dimension (D), the weights of the factors were obtained from fractal perspective. We proposed a weighted calibrated landslide potential model (WCLPM), obtained by the combination of CLPI values and weights of the factors. The proposed method is illustrated by example in Xinjiang, NW China, where landslide points are clustered at regional scale. In the example, the landslides were randomly split into two groups: one for building landslide model (training dataset) and the other for validating the model (validating dataset). Five landslide conditional factors (lithology, tectonic faults, elevation, slope, aspect) were selected, processed, and analyzed in a geographic information system (GIS) environment. Predictive accuracy of the WCLPM was evaluated and compared based on the calculation of area under the prediction-rate curve (AUPRC). The example shows that the proposed WCLPM provides good prediction for the study area (AUPRC = 0.8700). This study provided a novel and practical method for LSM.

Spatial joint analysis for zonal daytime and nighttime crash frequencies using a Bayesian bivariate conditional autoregressive model

This study presents a joint analysis of daytime and nighttime crash frequencies at the zone level with consideration of spatial correlations. Crash data from 131 traffic analysis zones in Hong Kong in 2011 are investigated. A Bayesian bivariate conditional autoregressive model is proposed to establish links between crash frequencies and traffic attributes, road network characteristics, and land use patterns. The proposed model allows not only for the distinct heterogeneous and spatial effects of each dependent variable, but also for the correlations between them.The parameter estimates indicate that more daytime and nighttime crashes are associated with more vehicle hours traveled and with networks that have greater global integration. Average speed alone has a significant negative effect on daytime crashes. The crash risk in commercial and other areas is lower than that in residential areas, but the crash risk in areas of mixed residential and commercial use is higher. Meanwhile, significant spatial autocorrelation emerges across zones and explains 46.7% and 48.2% extra-Poisson variations for daytime and nighttime crash frequencies, respectively. High positive correlations are found in both heterogeneous and spatial effects. These findings, together with its better performance on model fit than the univariate counterparts, demonstrate the strength of the proposed model.

Study on the calculation method for dynamic wind deflection of transmission line in strong wind area

In order to improve the design and early-warning accuracy for wind deflection of transmission line in strong wind area, it is necessary to calculate the dynamic wind deflection of transmission line in consideration of fluctuating wind effect. In this paper, the generation method for fluctuating wind velocity in consideration of spatial correlation is discussed, then the wind load on every element node of conductor is determined on the basis of the calculation formula for wind load, and finally the dynamic wind deflection angle of insulator and the dynamic wind deflection displacement of conductor is calculated by using the nonlinear dynamic finite element method.

A Spatiotemporal Multi-View-Based Learning Method for Short-Term Traffic Forecasting

Short-term traffic forecasting plays an important part in intelligent transportation systems. Spatiotemporal k-nearest neighbor models (ST-KNNs) have been widely adopted for short-term traffic forecasting in which spatiotemporal matrices are constructed to describe traffic conditions. The performance of the models is closely related to the spatial dependencies, the temporal dependencies, and the interaction of spatiotemporal dependencies. However, these models use distance functions and correlation coefficients to identify spatial neighbors and measure the temporal interaction by only considering the temporal closeness of traffic, which result in existing ST-KNNs that cannot fully reflect the essential features of road traffic. This study proposes an improved spatiotemporal k-nearest neighbor model for short-term traffic forecasting by utilizing a multi-view learning algorithm named MVL-STKNN that fully considers the spatiotemporal dependencies of traffic data. First, the spatial neighbors for each road segment are automatically determined using cross-correlation under different temporal dependencies. Three spatiotemporal views are built on the constructed spatiotemporal closeness, periodic, and trend matrices to represent spatially heterogeneous traffic states. Second, a spatiotemporal weighting matrix is introduced into the ST-KNN model to recognize similar traffic patterns in the three spatiotemporal views. Finally, the results of traffic pattern recognition under these three spatiotemporal views are aggregated by using a neural network algorithm to describe the interaction of spatiotemporal dependencies. Extensive experiments were conducted using real vehicular-speed datasets collected on city roads and expressways. In comparison with baseline methods, the results show that the MVL-STKNN model greatly improves short-term traffic forecasting by lowering the mean absolute percentage error between 28.24% and 46.86% for the city road dataset and, between 53.80% and 90.29%, for the expressway dataset. The results suggest that multi-view learning merits further attention for traffic-related data mining under such a dynamic and data-intensive environment, which owes to its comprehensive consideration of spatial correlation and heterogeneity as well as temporal fluctuation and regularity in road traffic.

Experimental and statistical study on the irregularity of carbonation depth of cement mortar under supercritical condition

The heterogeneity of a cement-based material results in a random spatial distribution of carbonation depth. Currently, there is a lack of both experimental and numerical investigations aiming at a statistical understanding of this important phenomenon. This paper presents both experimental and numerical supercritical carbonation test results of cement mortar blocks. The carbonation depths are measured along the carbonation boundary by the proposed rapid image processing technique, which are then statistically studied by calculating, e.g., their probability density and power spectral density (PSD). The results indicate that the distribution of the carbonation depth can be approximately represented by a lognormal distribution function and the PSD has quantitative correlation with some of the statistic parameters used in the simulations. In particular, the effects of the autocorrelation lengths and the coefficient of variation of porosity, which are used to define the random porosity field, on the irregularity of carbonation depth are analyzed numerically in details and validated by experimental results. The study has shown that using a random field of porosity with due consideration of spatial correlation and variance, the irregularity of carbonation depth can be realistically captured by the numerical model. The numerical results confirm that lognormal distributions represent the random nature of carbonation depth well and the average and variance of the irregular carbonation depth increase with the increase of carbonation time, autocorrelation length and coefficient of variation of porosity. The study also offers a potential method to numerically calibrate some of the statistic parameters required by a numerical carbonation model through comparing the PSD with that from experimental tests. Overall the methodology adopted in the paper can provide a foundation for future investigations on probability analysis of carbonation depth and other similar work based on multi-scale and -physics modelling.

The effect of random porosity field on supercritical carbonation of cement-based materials

In this paper, the supercritical carbonation process of cement-based materials is modelled by introducing a random porosity field to simulate the heterogeneous geometry of the carbonation profile. The suitability of two different random fields of porosity, based on the probability density function (PDF) and the ellipsoidal autocorrelation function (EAF) methods, are investigated, respectively, in simulating the distribution of porosity in cement mortar. After incorporating the above random fields into an established supercritical carbonation model, it is found that with some modifications, the EAF method with consideration of spatial correlation produces better simulation of the irregularities of the carbonation zones that have been observed from experimental results. It is also found that for given average porosity and coefficient of variation, the predicted average and maximum carbonation depths have much smaller coefficients of variation. The validated EAF supercritical carbonation model is then used in parametric studies that are conducted to assess the effect of various factors on the carbonation depth of the chemical process.

Trapped in the web of water: Groundwater-fed springs are island-like ecosystems for the meiofauna.

We investigated whether the equilibrium theory of island biogeography (ETIB) can be applied to the meiofauna of groundwater‐fed springs. We tested whether copepod species richness was related with spring area, discharge, and elevation. Additionally, five hypotheses are tested based on species distribution patterns, dispersal ability, and life‐history characteristics of several guilds (stygobiotic, nonstygobiotic, cold stenotherm, and noncold stenotherm species). Thirty springs in the central Apennines (Italy) were considered. A multimodel selection procedure was applied to select best‐fit models using both ordinary least‐squares regressions and autoregressive models. Mantel tests were used to investigate the impact of spatial autocorrelation in determining interspring similarity (ßsor), pure turnover (ßsim), intersite nestedness (ßnest = ßsor − ßsim), and matrix nestedness (measured using NODF and other metrics). Explicit consideration of spatial correlations reduced the importance of predictors of overall species richness, noncold stenotherm species (both negatively affected by elevation), cold stenotherm species, and nonstygobiotic species, but increased the importance of area for the stygobiotic species. We detected nested patterns in all cases, except for the stygobites. Interspring distances were positively correlated with ßsor and ßnest (but not with ßsim) for the entire data set and for nonstygobiotic, cold stenotherm, and noncold stenotherm species. In the case of stygobites, interspring geographical distances were marginally correlated with ßsor and no correlation was found for ßsim and ßnest. We found support for ETIB predictions about species richness, which was positively influenced by area and negatively by elevation (which expresses the size of source of immigrants). Low turnover and high nestedness are consistent with an equilibrium scenario mainly regulated by immigration and extinction. Stygobites, which include many distributional and evolutionary relicts, have a low capability to disperse through the aquifers and tend to be mainly confined to the springs where they drifted out and were trapped by springbed sediments.

Transport in two-dimensional modulation-doped semiconductor structures

We develop a theory for the maximum achievable mobility in modulation-doped 2D GaAs-AlGaAs semiconductor structures by considering the momentum scattering of the 2D carriers by the remote ionized dopants, which must invariably be present in order to create the 2D electron gas at the GaAs-AlGaAs interface. The minimal model, assuming first-order Born scattering by random quenched remote dopant ions as the only scattering mechanism, gives a mobility much lower (by a factor of 3 or more) than that observed experimentally in many ultrahigh-mobility modulation-doped 2D systems, establishing convincingly that the model of uncorrelated scattering by independent random remote quenched dopant ions is often unable to describe the physical system quantitively. We theoretically establish that the consideration of spatial correlations in the remote dopant distribution can enhance the mobility by (up to) several orders of magnitudes in experimental samples. The precise calculation of the carrier mobility in ultrapure modulation-doped 2D semiconductor structures thus depends crucially on the unknown spatial correlations among the dopant ions in the doping layer which may manifest sample to sample variations even for nominally identical sample parameters (i.e., density, well width, etc.), depending on the details of the modulation-doping growth conditions.

A Probabilistic Approach to Small Signal Stability Analysis of Power Systems with Correlated Wind Sources

This paper presents a probabilistic methodology for small signal stability analysis of power system with correlated wind sources. The approach considers not only the stochastic characteristics of wind speeds which are treated as random variables with Weibull distributions, while also the wind speed spatial correlations which are characterized by a correlation matrix. The approach based on the 2m+1 point estimate method and Cornish Fisher expansion, the orthogonal transformation technique is used to deal with the correlation of wind farms. A case study is carried out on IEEE New England system and the probabilistic indexes for eigenvalue analysis are computed from the statistical processing of the obtained results. The accuracy and efficiency of the proposed method are confirmed by comparing with the results of Monte Carlo simulation. The numerical results indicate that the proposed method can actually capture the probabilistic characteristics of mode properties of the power systems with correlated wind sources and the consideration of spatial correlation has influence on the probability of system small signal stability.

A predictive model for Arias intensity at multiple sites and consideration of spatial correlations

SUMMARYArias intensity, Ia, has been identified as an efficient intensity measure for the estimation of earthquake‐induced losses. In this paper, a new model for the prediction of Arias intensity, which incorporates nonlinear site response through the use of the average shear‐wave velocity and a heteroskedastic variance structure, is proposed. In order to estimate the effects of ground motions on spatially‐distributed systems, it is important to take into account the spatial correlation of the intensity measure. However, existing loss‐estimation models, which use Ia as an input, do not take this aspect of the ground motion into account. Therefore, the potential to model the spatial correlation of Arias intensity is also investigated. The empirical predictive model is developed using recordings from the Pacific Earthquake Engineering Research Center Next Generation of Attenuation database whereas the model for spatial correlation makes use of the well‐recorded events from this database, that is the Northridge and Chi‐Chi earthquakes. Copyright © 2011 John Wiley & Sons, Ltd.

Effect of spatial correlation on regional trends in rain events over India

The regional trends are evaluated in the frequency of various rain events using the daily gridded (1° × 1°) rainfall dataset for the time period 1901–2004, prepared by the India Meteorological Department (IMD). In terms of intensity, the events are classified as low, moderate, heavy and extreme heavy, while short and long spells are classified on the basis of duration of rainfall. The analytical (parametric) and the empirical (bootstrap) techniques were used to incorporate the impact of spatial correlation in regional trends. It is observed that, consideration of spatial correlation reduces the significance level of the trends and the effective number of grid points falling under each category. Especially, the noticeable cross-correlation have reduced the significance of the trends in moderate and long spell rain events to a large extent, while the significance of trends in the extreme heavy and short spell events is not highly affected because of small cross-correlation.