Search

Echo state network (ESN) has widely attracted many researchers due to its training process without backpropagation. However, it is hard for single ESN to fit those complex and polytrophic situations. Under this case, a novel multiscale deep ESN (MDESN) is developed in this study, which integrates the deep learning, parallel structure and multiscale reservoir state matrix mode, respectively. The deep framework is responsible for collecting multivariate dimensional reservoir states; Upgradation of various reservoir states could efficiently dig hidden characteristics for decoding. The parallel structure could be trained simultaneously to decrease time consumption through the multiple thread method. According to the multivariate high dimensional mapping method, MDESN could obtain more robust generalization information compared to the state-of-the-art models. MDESN is evaluated in two chaos prediction benchmarks (Lorenz and Mackey-Glass) and real solar irradiance predictions. Various prediction horizons including one-step-ahead and multi-step-ahead prediction are designed and conducted respectively to verify the effectiveness and robustness of MDESN. Furthermore, RMSE, MAE, MAPE, and R are adopted to evaluate our proposed model. The statistical results demonstrate that MDESN has the best-performing adaptability and robustness compared to classical ESN, deep ESN (DESN), and other state-of-the-art models. [ABSTRACT FROM AUTHOR]

Boundary identification of field sources is an indispensable task for interpreting field data. Initially, people used the distribution characteristics of data to obtain boundary information of field sources, making it difficult to identify weak anomalies amidst strong background anomalies. To address this issue, automatic control filters based on a certain window size were employed to identify the distribution of field sources. However, this method's results heavily relied on the window size and were not well applicable to complex anomalies. In recent years, features reflecting the boundary information of field sources have mainly been derived from the derivatives of scalar field data. Then, the correspondence between imaging results and boundaries is utilized to identify the horizontal boundaries of field sources. Specifically, extreme values of magnetic anomaly horizontal derivatives and zero values of vertical derivatives correspond to geological body boundaries. Existing boundary identification methods mainly utilize a balanced boundary identification filter composed of the ratio of first-order horizontal and vertical derivatives to delineate the positions of geological bodies, but the method has lower resolution and generality. Therefore, this paper proposes combining boundary detection filters based on ratios of derivatives of different orders with multiscale unsupervised deep learning. This approach utilizes different orders of derivative ratios to obtain higher-resolution edge imaging results. Additionally, a combination of Deep Image Prior (DIP) and Generative Adversarial Network-None Local (GAN-NL) networks for multiscale unsupervised deep learning is established to determine the horizontal position of sources based on extreme values of edge imaging results. The multiscale DIP network is used to identify the source position, and a self-attention mechanism neural network is added to the DIP network to enhance its learning ability, which can remove noise without requiring a large amount of data label.

In this article we present the results obtained applying the multiple scale expansion up to the order $\varepsilon^6$ to a dispersive multilinear class of equations on a square lattice depending on 13 parameters. We show that the integrability conditions given by the multiple scale expansion give rise to 4 nonlinear equations, 3 of which seem to be new, depending at most on 2 parameters.

• A piezoelectric energy harvesting system model based on parametric excitation and low frequency is proposed. • Geometric and damping complex nonlinear factors are considered. • An accurate theoretical model of piezoelectric energy harvesting system excited by parametric excitation is established. • Load resistance effect on the parametric excited initial threshold is explored. • Integrated parametric excitation and effect of tip mass induce significant improvement to the performance. Capitalizing on the advantages of parametric excitation and real-life low frequency vibration, the steady-state response of piezoelectric energy harvesting system composing of a cantilever beam with a tip mass and horizontal excitation at the fixed end is investigated. By applying Hamilton's principle, the nonlinear partial differential equation of a cantilever bimorph piezoelectric energy harvesting system with an additional tip mass is derived and analyzed. The cantilever is modeled as an axially non-elongated Euler Bernoulli beam with geometric and damping nonlinearities. By using the Galerkin method, the nonlinear partial differential equation is reduced to an electromechanical coupling system that governs a cantilever piezoelectric energy harvesting system with a tip mass under parametric excitation. The first-order resonance response of this harvesting system is studied by using the method of multiple scales. The analytical response expressions and first-order amplitude functions for vertical displacement, output voltage and output power are derived and analyzed. The influence of different impedance and tip mass on the piezoelectric energy harvesting system performance is summarized and concluded. For the parametric excitation system, the load resistance shows significant influence on the initial threshold of the piezoelectric energy harvesters under parametric excitation. For a short circuit, the initial threshold increases with increasing load resistance. For an open circuit, the initial threshold decreases with increasing load resistance. For an open circuit resistance or a short circuit resistance under parametric excitation, an increase in the end block mass effectively reduces the resonance frequency of the piezoelectric energy harvesters. Consequently, the energy harvesting bandwidth is increased, and the harvesting effect of the piezoelectric energy harvesters is enhanced. In this paper, a more accurate and practical theoretical model is established to predict the electromechanical coupling behavior of cantilever piezoelectric energy harvester under parametric excitation. Using this approach, it is possible to complement parametric and direct excitation such that the advantages of parametric excitation can be achieved. It is also significant to improve the energy conversion efficiency of energy harvesting system. [Display omitted] [ABSTRACT FROM AUTHOR]

To address the issue of the insufficient ability of existing models in the extraction of crucial features in person re-identification (ReID) caused by occlusions and pose variations, a person re-identification algorithm driven by a global attention mechanism and relation network is proposed. Firstly, in the backbone network, a global attention module is embedded into the ResNei-50 network to capture weight information in both spatial and channel dimensions. Secondly, in the relation network, multiple local features of different scales are obtained through horizontal partitioning. Two modules are designed to extract the global contrast features and relation features among local features. The local relation features are fused with global contrast information and fed into the classification network. Finally, in the loss optimization stage, a joint loss consisting of Circle Loss, triplet loss, and cross-entropy loss is employed for network training. Experiments are conducted on five commonly used datasets, and the results are analyzed. The proposed algorithm achieves a rank-1 accuracy of 63.2% and 95.4%, as well as mAP scores of 53.8% and 88.2% on the Occluded-DukeMTMC and Market 1501 datasets, respectively, demonstrating its advancement. [ABSTRACT FROM AUTHOR]

Leaf spot (LS) caused by Cercosporidium personatum is one of the most harmful peanut diseases in the late growth stage and severely affects the yield of peanuts. Hyperspectral disease detection technology is efficient, objective, and accurate and is suitable for large-scale crop management practices. To establish a multi-scale spectral index (SI) with high accuracy and stability for the detection of peanut LS disease, the spectral reflectance of different disease severity levels at leaf, plant, and field scales was collected, and the difference in wavelength caused by disease severity was analyzed using the mean, variance, and dispersion matrix of hyperspectral reflectance. Meanwhile, the feature weights at different scales were obtained using Relief-F, and the average feature weights identified 540, 660, and 770 nm as multi-scale sensitive wavelengths. Three new SIs were constructed by combining single, ratiometric, and normalized wavelengths. The new SIs were compared and analyzed with 35 commonly used SIs by correlation analysis and M-statistic values, and 6 SIs were significantly correlated with disease severity levels and had good separability. Finally, k-nearest neighbor (KNN) and multinomial logistic regression (MLR) were used to evaluate the ability of the above SIs to detect LS severity. The results showed that the leaf spot multi-scale spectral index (LS-MSSI) constructed in this study was superior to the other SIs and obtained high accuracy at different scales simultaneously. At the leaf and plant scales, the MLR obtained high accuracy, with the overall accuracy (OA) reaching 93.77% and 92.50% and Kappa reaching 91.59% and 89.97%, respectively. At the field scale, the KNN obtained high accuracy, with the OA and Kappa reaching 90.29% and 87.04%, respectively. The LS-MSSI proposed in this study has high accuracy, stability, and robustness in the detection of LS severity at multiple scales, providing a technical basis and scientific guidance for the detection and precise management of peanuts. [ABSTRACT FROM AUTHOR]

In 2019, COVID-19 quickly spread across the world, infecting billions of people and disrupting the normal lives of citizens in every country. Governments, organizations, and research institutions all over the world are dedicating vast resources to research effective strategies to fight this rapidly propagating virus. With virus testing, most countries publish the number of confirmed cases, dead cases, recovered cases, and locations routinely through various channels and forms. This important data source has enabled researchers worldwide to perform different COVID-19 scientific studies, such as modeling this virus’s spreading patterns, developing prevention strategies, and studying the impact of COVID-19 on other aspects of society. However, one major challenge is that there is no standardized, updated, and high-quality data product that covers COVID-19 cases data internationally. This is because different countries may publish their data in unique channels, formats, and time intervals, which hinders researchers from fetching necessary COVID-19 datasets effectively, especially for fine-scale studies. Although existing solutions such as John’s Hopkins COVID-19 Dashboard and 1point3acres COVID-19 tracker are widely used, it is difficult for users to access their original dataset and customize those data to meet specific requirements in categories, data structure, and data source selection. To address this challenge, we developed a toolset using cloud-based web scraping to extract, refine, unify, and store COVID-19 cases data at multiple scales for all available countries around the world automatically. The toolset then publishes the data for public access in an effective manner, which could offer users a real time COVID-19 dynamic dataset with a global view. Two case studies are presented about how to utilize the datasets. This toolset can also be easily extended to fulfill other purposes with its open-source nature.

Copyright of Journal of Engineering Geology / Gongcheng Dizhi Xuebao is the property of Journal of Engineering Geology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Human-wildlife conflict impacts native wildlife populations and people in socio-economically poor areas. In the Shigar Valley, Karakoram range, Pakistan, subsistence mixed farming is the predominant land use and red foxes (Vulpes vulpes montana) are widely persecuted. Knowledge of the effects of human activity on fox behavior and populations in the region is lacking.We carried out 38 transect surveys within 87 sampling units of grid size 3 km × 3 km and recorded 42 resting and 24 active natal dens from 2015 to 2017. We evaluated 17 model subsets for describing probabilistic Resource Selection Functions (RSFs) predicting den occurrence at landscape, patch, and micro-habitat scales in terms of anthropogenic factors, including distance to agricultural land and roads. We also tested and controlled for associations between den occurrence and biotic and abiotic variables, including prey occurrence. We found that: 1) distance to roads negatively influenced fox den occurrence; 2) Indian pika (Ochotona roylei) burrow occurrence positively influenced fox den occurrence; (3) Natal dens had larger and more numerous openings than resting dens; 4) den occurrence was greatest at moderate elevations; and 5) den occurrence was negatively related to forest cover and positively correlated with shrub cover. The findings suggest that foxes do not strongly avoid human activity, and that den types are variable. Further studies are required on the function of foxes in this landscape focusing on ecosystem services such as seed dispersal, carrion removal and regulation of prey populations.

The nonlinear vibration behavior and dynamic instability of Euler–Bernoulli nanobeams under thermo-magneto-mechanical loads is the main objective of the present paper. Firstly, a short Euler–Bernoulli nanobeam is modeled and exposed to an external parametric excitation. Based on the nonlocal continuum theory and nonlinear von Karman beam theory, the nonlinear governing differential equation of motion is derived. Secondly, to transport the partial differential equation to the ordinary differential equation, Galerkin method is applied. Then, multiple scales method, as an analytical approach, is used to solve the equation. At the end, modulation equation of Euler–Bernoulli nanobeams is obtained. Then, to evaluate the dynamic instability of the system, trivial and nontrivial steady-state solutions are discussed. Emphasizing the effect of parametric excitation, for considering the instability regions, bifurcation points are studied and investigated. As a result, it can be observed that the damping coefficient plays an effective role as well as parametric excitation in stability and frequency response of the system. [ABSTRACT FROM AUTHOR]

• Positive position feedback (PPF) controllers added to a composite plate with mixed excitations. • A new six degrees of freedom model simulating nonlinear vibrations of the plate are proposed. • Perturbation technique is applied to have the solutions of the main system. • Numerical outcomes show that PPF methods have big effects on reducing vibrations of the amplitudes. • Numerical solution be an excellent harmony with steady-state amplitude using frequency response curves. The idea of this manuscript is to convert nonlinear positive position feedback (PPF) controllers with symmetric cross-ply composite piezoelectric laminated plates among mixed excitations. A new six degrees of freedom model simulating nonlinear vibrations (NVs) of a composite plate has proposed. The perturbation technique has used to study the transient and steady-state response of nonlinear dynamic equations that presented the main system. The stability of the system via frequency response curves (FRCs) and force response curves (FRCs) have discussed and calculated. The effect on every coefficient's system has considered numerically. A comparison has been made between PPF controller and another control Nonlinear Saturation Control (NSC) to show the amount of vibration reduction produced by the system using PPF control is better than that NSC. Numerical outcomes show that PPF methods have big effects on reducing vibrations of the amplitudes. Then, the predictions from numerical simulations are in excellent harmony with frequency response curves. Lastly, the comparison with newly available papers has been reported. [ABSTRACT FROM AUTHOR]

Multi-scale feature fusion has been proven effective in substantial person re-identification (ReID) works. However, the existing multi-scale feature fusion is based on features of different semantic levels. We propose a novel multi-scale and multi-branch feature representation for person ReID, namely Ms-Mb. It merges the features of the same semantic level and integrates attention modules to learn robust and representative feature representations. Through the heterogeneous losses supervision, the final feature representation of the image is more discriminative for person ReID. Sufficient ablation study has proven that the multi-scale feature fusion, the attention module and heterogeneous losses training strategy contribute to the performance boost of Ms-Mb. We have conducted experiments on four mainstream benchmarks including Market1501, DukeMTMC-reID, CUHK03 and MSMT17. Extensive experimental results show that our approach Ms-Mb achieves state-of-the-art performances on Market1501 (Rank-1 = 95.8 % , mAP = 89.9 %), DukeMTMC-reID (Rank-1 = 90.8 % , mAP = 82.2 %) and MSMT17 (Rank-1 = 81.9 % , mAP = 59.3 %) without using additional external data or re-ranking. Our approach yields competitive results compared to the state-of-the-art method on CUHK03 (Rank-1 = 75.4 % , mAP = 72.9 %) and surpasses the other methods by a large margin. [ABSTRACT FROM AUTHOR]

Analyzing the spatial patterns of net primary productivity (NPP) and its driving forces in transnational areas provides a solid basis for understanding regional ecological processes and ecosystem services. However, the spatial patterns of NPP and its driving forces have been poorly understood on multiple scales in transnational areas. In this study, the spatial patterns of NPP in the transnational area of the Tumen River (TATR) in 2016 were simulated using the Carnegie Ames Stanford Approach (CASA) model, and its driving forces were analyzed using a stepwise multiple linear regression model. We found that the total amount of NPP in the TATR in 2016 was approximately 14.53 TgC. The amount of NPP on the Chinese side (6.23 TgC) was larger than those on the other two sides, accounting for 42.88% of the total volume of the entire region. Among different land-use and land-cover (LULC) types, the amount of NPP of the broadleaf forest was the largest (11.22 TgC), while the amount of NPP of the bare land was the smallest. The NPP per unit area was about 603.21 gC/(m2·yr) across the entire region, while the NPP per unit area on the Chinese side was the largest, followed by the Russian side and the DPRK's side. The spatial patterns of NPP were influenced by climate, topography, soil texture, and human activities. In addition, the driving forces of the spatial patterns of NPP in the TATR had an obvious scaling effect, which was mainly caused by the spatial heterogeneity of climate, topography, soil texture, and human activities. We suggest that effective land management policies with cooperation among China, the DPRK, and Russia are needed to maintain NPP and improve environmental sustainability in the TATR. [ABSTRACT FROM AUTHOR]

Based on comprehensive analysis of core, well logging, seismic and production data, the multi-scale reservoir space, reservoir types, spatial shape and distribution of fractures and caves, and the configuration relationship with production wells in fracture-cavity carbonate reservoirs were studied systematically, the influence of them on the distribution of residual oil was analyzed, and the main controlling factors mode of residual oil distribution after water flooding was established. Enhanced oil recovery methods were studied considering the development practice of Tahe oilfield. Research shows that the main controlling factors of residual oil distribution after water flooding in fracture-cavity carbonate reservoirs can be classified into four categories: local high point, insufficient well control, flow channel shielding and weak hydrodynamic. It is a systematic project to improve oil recovery in fracture-cavity carbonate reservoirs. In the stage of natural depletion, production should be well regulated to prevent bottom water channeling. In the early stage of waterflooding, injection-production relationship should be constructed according to reservoir type, connectivity and spatial location to enhance control and producing degree of waterflooding and minimize remaining oil. In the middle and late stage, according to the main controlling factors and distribution characteristics of remaining oil after water flooding, remaining oil should be tapped precisely by making use of gravity differentiation and capillary force imbibition, enhancing well control, disturbing the flow field and so on. Meanwhile, backup technologies of reservoir stimulation, new injection media, intelligent optimization etc. should be developed, smooth shift from water injection to gas injection should be ensured to maximize oil recovery. [ABSTRACT FROM AUTHOR]

The use of multibeam echosounder systems (MBES) for detailed seafloor mapping is increasing at a fast pace. Due to their design, enabling continuous high-density measurements and the coregistration of seafloor’s depth and reflectivity, MBES has become a fundamental instrument in the advancing field of acoustic seafloor classification (ASC). With these data becoming available, recent seafloor mapping research focuses on the interpretation of the hydroacoustic data and automated predictive modeling of seafloor composition. While a methodological consensus on which seafloor sediment classification algorithm and routine does not exist in the scientific community, it is expected that progress will occur through the refinement of each stage of the ASC pipeline: ranging from the data acquisition to the modeling phase. This research focuses on the stage of the feature extraction; the stage wherein the spatial variables used for the classification are, in this case, derived from the MBES backscatter data. This contribution explored the sediment classification potential of a textural feature based on the recently introduced Weyl transform of 300 kHz MBES backscatter imagery acquired over a nearshore study site in Belgian Waters. The goodness of the Weyl transform textural feature for seafloor sediment classification was assessed in terms of cluster separation of Folk’s sedimentological categories (4-class scheme). Class separation potential was quantified at multiple spatial scales by cluster silhouette coefficients. Weyl features derived from MBES backscatter data were found to exhibit superior thematic class separation compared to other well-established textural features, namely: (1) First-order Statistics, (2) Gray Level Co-occurrence Matrices (GLCM), (3) Wavelet Transform and (4) Local Binary Pattern (LBP). Finally, by employing a Random Forest (RF) categorical classifier, the value of the proposed textural feature for seafloor sediment mapping was confirmed in terms of global and by-class classification accuracies, highest for models based on the backscatter Weyl features. Further tests on different backscatter datasets and sediment classification schemes are required to further elucidate the use of the Weyl transform of MBES backscatter imagery in the context of seafloor mapping.

The study of ecological vulnerability is of great significance to regional ecological environment protection and sustainable development. Bailongjiang Watershed (BLJW) of Gansu Province is a vital area for water conservation and environmental fragile in the upper Yangtze River. Therefore, it is necessary to understand the ecological vulnerability of BLJW for local sustainable development. Based on the sensitivity-recovery-pressure (SRP) model, landscape pattern index and the typical ecosystem type, we constructed the assessment system of ecological vulnerability by taking township as the spatial unit to analyze the spatiotemporal pattern of ecological vulnerability from the township, county, watershed scales in BLJW of Gansu Province in 2002 and 2014 via the auxiliary regression, entropy weight, synthetic index, and spatial autocorrelation methods. The results showed that from the view of the township scale, the townships of severe or extreme vulnerability were concentrated in the northwestern Tanchang and most of Wudu in 2002 and 2014. The ecological vulnerability in the watershed had changed from moderate vulnerability to light vulnerability at township scale. At the county scale, the change of ecological vulnerability in Zhouqu was relatively stable, while that of Wudu, Wenxian, Tanchang and Diebu were relatively violent from 2002 to 2014. The ecological vulnerability of Wudu, Wenxian and Diebu improved to a certain extent, while that of Tanchang was deteriorating gradually. On watershed scale, the average value of ecological vulnerability index in the watershed were 0.2976 and 0.2904 in 2002 and 2014, respectively, showing a small downward trend, indicating that the ecological vulnerability in the watershed decreased in the past 12 years. The coefficients of variation were 0.3905 and 0.4358, respectively, with an increasing trend, suggesting that the spatial pattern of ecological vulnerability in the watershed developed in an unbalanced way. In 2002, the Moran I of ecological vulnerability index in the watershed was 0.5460, implying that the ecological vulnerability in the watershed was spatially agglomerated. Specifically, the areas of high-high were mainly distributed in the northern and central Wudu, and northwestern Tanchang. In 2014, the Moran I of ecological vulnerability index in the watershed increased to 0.5635, which indicated that the spatial agglomeration was more significant, and the areas of high-high were concentrated in central Wudu and northwestern Tanchang. [ABSTRACT FROM AUTHOR]

This paper presents a critical analysis of the two-scale command shaping (TSCS) feedforward control technique as applied to nonlinear Duffing oscillators. TSCS is an approach for tailoring a flexible system’s applied control input to reduce undesirable residual vibrations using multiple problem scales, command shaping of a linear subproblem, and cancelation of a remaining nonlinear subproblem. As shown herein, TSCS proves to be an effective method for feedforward control of nonlinear systems. The strategy outperforms conventional and nonlinearly informed command shaping strategies in traditional and non-traditional Duffing systems (e.g., Duffing systems with quadratic nonlinearity and Coulomb damping). The TSCS approach is also extended herein to nonlinear systems with uncertain parameters through the implementation of robust command shaping strategies and the extended Kalman filtering parameter estimation technique. [ABSTRACT FROM AUTHOR]

The relative importance of environmental variables for Spartina alterniflora distribution was investigated across different spatial scales using maximum entropy modelling (MaxEnt), a species distribution modelling technique. The results showed that elevation was the most important predictor for species presence at each scale. Mean diurnal temperature range and isothermality were the second most important predictors at national and regional scales respectively. Soil drainage class, pH and organic carbon were important on the northern Chinese coast. The importance of climatic variable type was highest at global and national scales and declined as the scale decreased. The importance of soil variable type was lower at coarser scales, but varied greatly at finer scales. The relationships between environmental variables and species presence changed as the variables' ranges changed across different scales. Climatic and soil variables were substantially affected by interactions among variables, which changed their relationships with species presence and relative importance. The modelled suitable area on the Chinese coast decreased from 54.16 to 12.64% limited by elevation from the global to national scale, and decreased to 8.04% limited by soil drainage, pH and organic carbon from the national to regional scale. The findings of the present study emphasise the importance of spatial scale for understanding relationships between environmental variables and the presence of S. alterniflora. [ABSTRACT FROM AUTHOR]

A reduced-order model that preserves physical meaning is important for generating insight in large-scale power system studies. The conventional model-order reduction for a multiple-timescale system is based on discarding states with fast (short timescale) dynamics. It has been successfully applied to synchronous machines, but is inaccurate when applied to power converters because the timescales of fast and slow states are not sufficiently separated. In the method proposed here, several fast states are at first discarded but a representation of their interaction with the slow states is added back. Recognizing that the fast states of many converters are linear allows well-developed linear system theories to be used to implement this concept. All the information of the original system relevant to system-wide dynamics, including nonlinearity, is preserved, which facilitates judgments on system stability and insight into control design. The method is tested on a converter-supplied mini power system and the comparison of analytical and experiment results confirms high preciseness in a broad range of conditions. [ABSTRACT FROM PUBLISHER]

[Display omitted] • A piezoelectric energy harvesting system model is proposed based on parametric excitation and direct excitation. • An accurate theoretical model is established considering the influence of complex nonlinear factors of piezoelectric energy harvesting system excited by parametric excitation and direct excitation. • The energy harvesting bandwidth of the distributed parameter model is larger than that of the lumped parameter model. • Combining parametric excitation with direct excitation for compensation and complementarity purposes can be an ideal solution for vibration energy harvesting. In this paper, a cantilever beam piezoelectric energy harvester model of nonlinear partial differential equation based on Hamilton's Principle is established. A piezoelectric device using a bimorph cantilever beam with a fixed is subjected to horizontal and vertical excitation. Based on the Euler-Bernoulli thin beam model assumptions and inextensible deformation, the analysis further includes the effects of geometric nonlinearity and damping nonlinearity. The reduced-order nonlinear partial differential equations of motion with electro-mechanical coupling distributed parameter are derived by using Galerkin method. By using the method of multiple scales and solving the equations of motion, the energy harvester in its fundamental first-order resonance response is analyzed and the amplitudes of fundamental transverse deflection, output voltage and output power are determined. The major influence aspects of excitation amplitude, linear damping coefficient, nonlinear damping coefficient, impedance, nonlinear electro-mechanical coupling coefficient on transverse deflection, voltage and power are analyzed. The result shows that the linear damping coefficient and resistance affect the initial threshold of parametric excitation. The combination of parametric excitation and direct excitation gives full play to the advantages of parametric excitation and improve the energy conversion efficiency of the energy harvesting system. [ABSTRACT FROM AUTHOR]

The drastic variation of motion in spatial and temporal dimensions makes the video prediction task extremely challenging. Existing RNN models obtain higher performance by deepening or widening the model. They obtain the multi-scale features of the video only by stacking layers, which is inefficient and brings unbearable training costs (such as memory, FLOPs, and training time). Different from them, this paper proposes a spatiotemporal multi-scale model called MS-LSTM wholly from a multi-scale perspective. On the basis of stacked layers, MS-LSTM incorporates two additional efficient multi-scale designs to fully capture spatiotemporal context information. Concretely, we employ LSTMs with mirrored pyramid structures to construct spatial multi-scale representations and LSTMs with different convolution kernels to construct temporal multi-scale representations. We theoretically analyze the training cost and performance of MS-LSTM and its components. Detailed comparison experiments with twelve baseline models on four video datasets show that MS-LSTM has better performance but lower training costs. • We propose a new spatiotemporal multiscale model named MS-LSTM, which incorporates three orthogonal multiscale designs. • MS-LSTM employs stacked LSTMs with multiple convolution kernels (MK-LSTM) to form a mirrored pyramid structure, which can fully acquire the spatiotemporal context representation and generates video from coarse to fine. • We theoretically analyze the training cost and performance of MS-LSTM. Experiments on four datasets with twelve competing models have proved that MS-LSTM has lower training occupation but higher performance. [ABSTRACT FROM AUTHOR]

Atmospheric blocking is a long standing structure stalled in the mid-troposphere which is often associated with extreme weather events such as droughts, heatwaves, flood and cold air outbreak. A striking atmospheric blocking is identified to persist over the US during 13-17 August 2007, exacerbating the existing drought over the Southeastern US. This pronounced blocking event not only intensified the concurrent drought conditions, but also led to a record-breaking heatwave over the Southeast of the US. The excessive heat observed during this heatwave is attributable to the subsidence-associated adiabatic warming as well as the dry-and-warm air advection over Alabama and the neighboring states. At the local scale, we choose Birmingham, AL, as the study area for exploring the blocking influence on urban heat island. Based on the remote sensing data, the surface (skin) urban heat island is found to be 8 °C in this area on the block-onset day. This provides partial evidences that the surface urban heat island intensity is likely amplified by the blocking-induced heat waves. The present work provides a unique case study in which blocking, drought, heatwave and urban heat island all occur concurrently, and interplay across a spectrum of spatial scales. We conclude that atmospheric blocking is capable of reinforcing droughts, initiating heatwaves, and probably amplifying the urban heat island intensity during the concurrent period. [ABSTRACT FROM AUTHOR]

Aquatic oligochaetes are very common in streams, and are used as biological assessment indicators as well as in the biological management of organic-enriched systems. In this study, we analyzed the effects of environmental factors influencing the distribution of aquatic oligochaetes Limnodrilus hoffmeisteri in streams. We used 13 environmental factors in three categories (i.e., geography, hydrology, and physicochemistry). Data on the distribution of oligochaetes and environmental factors were obtained from 1159 sampling sites throughout Korea on a nationwide scale. Hierarchical cluster analysis (HCA) and nonmetric multidimensional scaling (NMDS) were performed to analyze the relationships between the occurrence of aquatic oligochaetes and environmental factors. A random forest model was used to evaluate the relative importance of the environmental factors affecting the distribution of oligochaetes. HCA classified sampling sites into four groups according to differences in environmental factors, and NMDS ordination reflected the differences of environmental factors, in particular, water depth, velocity, and altitude, among the four groups defined in the HCA. Furthermore, using a random forest model, turbidity and water velocity were evaluated as highly important factors influencing the distribution of L. hoffmeisteri. [ABSTRACT FROM AUTHOR]

We present an adaptive Multiscale Mixed Finite Element Method (MsMFEM) for modeling multiphase flow in highly heterogeneous reservoir. In this framework, fractional flow model is used to approximate pressure and velocity solution on coarse scale, while resolution in fine scale is honored by the basis functions, which are calculated by local problems on fine scale. The adaptive computation in the flow problem is permitted by two different basis functions. In the numerical examples, our adaptive MsMFEM is applied to solve two dimensional and three dimensional reservoir simulation problems with highly heterogeneous porosity and permeability fields. The preliminary numerical simulation results presented show a significant speedup in comparison with the reference numerical method, which encourage and is beneficial for further investigation of the proposed method for reservoir numerical simulation. [ABSTRACT FROM AUTHOR]

Chilika, the largest coastal lagoon of Asia, represents a complex transitional zone between terrestrial and marine ecosystems. In this study benthic macro-invertebrate community structure was elucidated based on multiple spatial scales approach (station-wise, sector-wise and region-wise) across Chilika from June 2013 to February 2015 (temporal scale), encompassing 23 pre-selected stations in link with measurement of abiotic parameters. Sixty two macro-invertebrate species representing seven macrofaunal groups were encountered, of which gastropods were the most abundant in terms of density (~96%) and species composition (~43%), and crustaceans were the least. The structure of the macrofaunal community (in terms of abundance) showed significant differences across spatial (station-wise, sector-wise and region-wise), and temporal scales (month-wise). While macro-invertebrate species composition revealed heterogeneity both station-wise and month-wise but it was found to be homogenous within the sector-wise and region-wise scales. BIO-ENV analysis indicated that salinity together with sediment nature were major abiotic variables associated with observed macro-invertebrate patterns. The results together demonstrated that macro-invertebrate communities are adapted to prevailing environmental conditions of Chilika lagoon; homogeneous and heterogeneous distribution and diversity patterns across multiple spatial scales revealed assemblages distinct to this ecosystem. The multiple scales approach could prove to be useful for ecological monitoring of lagoons. [ABSTRACT FROM AUTHOR]

• A UMS-FRCNN method for automatic detection and classification. • Improving Classification Accuracy with Convolutional Region Proposal Networks. • Detect small textured regions using a region proposal component. • The network is trained end-to-end through a fully joint training method. In traditional industrial production, the classification of cork mainly adopts manual operation. Due to round cork blocks with complex textures, manual sorting is time-consuming and often leads to inefficiency. It will lead to inefficiency and a high error rate. Therefore, an automatic cork sorting system is urgently required to replace manual operation. In this paper, an advanced deep-learning method based on machine vision is adopted to detect the surface of corks and classify them automatically. To robustly detect the texture of cork profile, a Multiple Scale Faster R-CNN based on the upper and lower layers (UMS-FRCNN) approach is proposed. The network consists of a region proposal component and the region-of-interest (RoI) detection component. There are two main contributions in our proposed network, which play a significant role in achieving state-of-the-art performance in cork classification. Firstly, the multi-scale information is grouped in two aspects: region suggestion and RoI detection, to detect on small cork texture regions. Secondly, the convolution Region Proposal Network (RPN) is used to generate the candidate regions of the UMS-FRCNN model to improve the classification accuracy. The experimental results show that our method not only can accurately classify the corks but also has better performance than the Faster RCNN. Compared with other methods, our method has the advantages of high precision, less time-consuming, and no dependence labeling. Our work has been successfully utilized in a cork production plant. [ABSTRACT FROM AUTHOR]