Window Algorithm Research Articles

Analysis and precision of the Terrestrial Surface Temperature using Landsat 8 and Sentinel 3 images: Study applied to the city of Granada (Spain)

The use of Landsat 8 and Sentinel 3 satellite images to calculate the land surface temperature (LST) has become one of the most common methods in urban areas. Knowing its precision and seasonal variability will affect future studies of urban climate alteration, such as Urban Heat Island (UHI) that allow sustainable decisions that increase the resilience of cities. In this research, the LST of a medium-sized city, Granada (Spain), was determined using six split window (SW) algorithms: three for Landsat 8 and another three for Sentinel 3A and 3B. Data were validated by measurements in situ at different parts of the city, chosen to represent the different land covers established in the Corine Land Cover inventory (2020). This city has a unique geographical situation, substantial pollution and high daily temperature variations, making it very suitable for study. The data validation by means of statistical analysis gave R2 linear fitting coefficients, the mean square error (RMSE), the mean bias error (MBE) and standard deviation. Our results reveal that the methods based on Landsat 8 SW algorithms present higher mean values (8.16 K) than Sentinel 3 (5.07 K) and Sentinel 3B (2.21 K), the differences being even greater during dry periods. Still, the SWs analyzed can be considered effective and reliable for determining the city’s LST. The use of a Convolutional Neural Network (CNN) in improving the quality of satellite images has reported significant improvements in the LST results obtained.

A global grid model for the correction of the vertical zenith total delay based on a sliding window algorithm

Reanalysis products have been applied to calculate the tropospheric delay for Global Navigation Satellite System (GNSS) positioning purposes widely. It is necessary to obtain high-precision tropospheric delay information from GNSS users with a high-precision tropospheric vertical stratification model because the height of the grids of the atmospheric reanalysis data is inconsistent with that of GNSS users, especially in regions with high terrains. In addition, the variation of the tropospheric delay in the vertical direction is much higher than that in the horizontal direction. The zenith total delay (ZTD) vertical stratification model is also key to the development of real-time and high-precision ZTD models. A new approach, the sliding window algorithm, is proposed to develop a ZTD vertical stratification model. In this work, a ZTD vertical stratification model considering spatiotemporal factors is developed based on the second Modern-Era Retrospective analysis for Research and Applications (MERRA-2) data, which is named the GZTD-H model. Radiosonde and International GNSS Service (IGS) data are treated as reference values to evaluate the performance of the GZTD-H model, which is compared to the model GPT2w. The results show that the GZTD-H model realizes the highest performance in ZTD layered vertical interpolation against ZTD layered profiles obtained at radiosonde sites, which achieves an improvement of 10% over the model GPT2w. Compared to model GPT2w, the GZTD-H model attains a spatial interpolation improvement of 8% for the Global Geodetic Observing System (GGOS) Atmosphere gridded ZTD over the surface ZTD calculated from radiosonde profiles. Furthermore, compared to model GPT2w, the model GZTD-H also attains improvements of 11% over the precise ZTD products acquired at IGS sites. In terms of model parameters, the GZTD-H model is greatly reduced and optimized over model GPT2w. Hence, the applicability of this model is enhanced in terms of GNSS atmospheric sounding and precise GNSS positioning.

An exploratory analysis of urbanization effects on climatic variables: a study using Google Earth Engine

Rapid global economic expansion has resulted in a drastic increase of urbanization while impacting the Earth’s entire ecology. This study evaluates the impact of historical land-use/land-cover (LU/LC) change signatures on seasonal variation of climatic variables using a cloud platform-Google Earth Engine. Due to rapid urbanization and the noticeable spatio-temporal difference in the climate, administrative units of Dakshina Kannada district are taken for demonstration. The LU/LC of the district extracted from high-resolution images of Landsat using random forest classification, land surface temperature (LST) extracted from the thermal band of Landsat images using the mono window algorithm, evapotranspiration (ET) data extracted from MOD16A2.006 and precipitation data from CHIPRS was used. The data was extracted for the pre-monsoon and post-monsoon period 2001–2019. The district has seen a 13.67% reduction in the forest area with 18.81% increase in the built-up areas. The LST and ET has seen a progressive drift in the past two decades, with an increase of 4.07 °C in median temperature in forest areas and a decline of 2.19 mm in median ET value, which necessitates monitoring forest encroachment. The higher variation in maximum LST in built-up land ( $$0.36\,^{\circ }\hbox {C/year/sq.km}$$ ) (near the industrial area) indicates that LU/LC change signature is the predominant driving factor and is associated with the physical characteristics of the built-up area. The ET exhibited a decreasing rate of 0.62 mm/year/sq.km of the built-up land. This study highlights the power of Google Earth Engine and free availability of satellite data in environmental protection, land-use management and sustainable development in the region.

A Statistical-Based Light-Weight Anomaly Detection Framework for Wireless Body Area Networks

Abstract In healthcare scenario, the major challenge in anomaly detection for remote patient monitoring is to classify true medical conditions and false alarms. This paper proposes a light-weight anomaly detection (LWAD) framework for detecting anomalies in remote patient monitoring based on wireless body area networks. The proposed framework uses distance correlation for finding correlated (both linear and non-linear) physiological parameters. It also uses a statistical-based improvised dynamic sliding window algorithm for efficient short-range prediction of physiological parameters. Finally, the proposed LWAD framework detects anomalies using anomaly detection framework based on robust statistical techniques. The validation of LWAD framework is performed using three real time datasets with various statistical measures. The proposed LWAD framework outperforms existing methods.

Estimation and Validation of Land Surface Temperature by using Remote Sensing & GIS for Chittoor District, Andhra Pradesh

Land Surface Temperature (LST) quantification is needed in various applications like temporal analysis, identification of global warming, land use or land cover, water management, soil moisture estimation and natural disasters. The objective of this study is estimation as well as validation of temperature data at 14 Automatic Weather Stations (AWS) in Chittoor District of Andhra Pradesh with LST extracted by using remote sensing as well as Geographic Information System (GIS). Satellite data considered for estimation purpose is LANDSAT 8. Sensor data used for assessment of LST are OLI (Operational Land Imager) and TIR (Thermal Infrared). Thermal band contains spectral bands of 10 and 11 were considered for evaluating LST independently by using algorithm called Mono Window Algorithm (MWA). Land Surface Emissivity (LSE) is the vital parameter for calculating LST. The LSE estimation requires NDVI (Normalized Difference Vegetation Index) which is computed by using Band 4 (visible Red band) and band 5 (Near-Infra Red band) spectral radiance bands. Thermal band images having wavelength 11.2 µm and 12.5 µm of 30th May, 2015 and 21st October, 2015 were processed for the analysis of LST. Later on validation of estimated LST through in-suite temperature data obtained from 14 AWS stations in Chittoor district was carried out. The end results showed that, the LST retrieved by using proposed method achieved 5 per cent greater correlation coefficient (r) compared to LST retrieved by using existing method which is based on band 10.

Control Strategy of RGV Operation Blockage and Deadlock in Plane Mobile Stereo Garage

Aiming at the safety requirements of multi-RGV and multi-lift planar mobile stereo garage access vehicles, a time window-based system RGV operation blocking and deadlock control strategy is proposed. Combining the improved banker algorithm with the time window principle, the banker algorithm is used to evaluate the security of the RGV running path. If an unsafe state is detected, the RGV operation blocking and deadlock control strategy based on the time window algorithm is given. The verification is detailed through specific cases. The results show that the RGV operation blocking and deadlock control strategy proposed in this paper can improve the RGV operation efficiency while increasing the system security.

A two-stage temporal proposal network for precise action localization in untrimmed video

In this paper, we propose a two-stage temporal proposal algorithm for the action detection task of long untrimmed videos. In the first stage, we propose a novel prior-minor watershed algorithm for action proposals with precise prior watershed proposal algorithm and minor supplementary sliding window algorithm. Here, we propose the correctness discriminator to fill the proposals that watershed proposal algorithm may omit with the sliding window proposals. In the second stage, an extended context pooling (ECP) is firstly proposed with two modules (internal and context). The context information module of ECP can structure the proposals and enhance the extended features of action proposals. Different level of ECP is introduced to model the action proposal region and make its extended context region more targeted and precise. Then, we propose a temporal context regression network, which adopts a multi-task loss to realize the training of the temporal coordinate regression and the action/background classification simultaneously, and outputs the precise temporal boundaries of the proposals. Here, we also propose prior-minor ranking to balance the effect of the prior watershed proposals and the minor supplementary proposals. On three large scale benchmarks THUMOS14, ActivityNet (v1.2 and v1.3), and Charades, our approach achieves superior performances compared with other state-of-the-art methods and runs over 1020 frames per second (fps) on a single NVIDIA Titan-X Pascal GPU, indicating that our method can efficiently improve the precision of action localization task.

Human activity classification using Decision Tree and Naïve Bayes classifiers

With rapid development in wireless sensor networks and continuous improvements in developing artificial intelligence-based scientific solutions, the concept of ambient assisted living has been encouraged and adopted. This is due to its widespread applications in smart homes and healthcare. In this regard, the concept of human activity recognition (HAR) & classification has drawn numerous researchers’ attention as it improves the quality of life. However, before using this concept in real-time scenarios, it is required to analyse its performance following activities of daily living using benchmarked data set. In this continuation, this work has adopted the activity classification algorithms to improve their accuracy further. These algorithms can be used as a benchmark to analyse others’ performance. Initially, the raw 3-axis accelerometer data is first preprocessed to remove noise and make it feasible for training and classification. For this purpose, the sliding window algorithm, linear and Gaussian filters have been applied to raw data. Then Naive Bayes (NB) and Decision Tree (DT) classification algorithms are used to classify human activities such as: sitting, standing, walking, sitting down and standing up. From results, it can be seen that maximum 89.5% and 99.9% accuracies are achieved using NB and DT classifiers with Gaussian filter. Furthermore, we have also compared the obtained results with its counterpart algorithms in order to prove its effectiveness.

A genetic fuzzy contention window optimization approach for IEEE 802.11 WLANs

The role of IEEE 802.11 wireless local area networks (WLANs) become vital due to its low cost deployment and the aspiration to improve its performance has become the need of the day. Among existing challenges the most prominent are success ratio, packet loss ratio, collision rate, fairness index, and energy consumption, advances in wireless technologies stress to surmount these challenges. To attain these, performance enhancement, genetic fuzzy-contention window optimization (GF-CWO) approach, is proposed which combined the fuzzy logic controller (FLC) and genetic algorithm (GA), through this way GA optimally tuned the FLC. For this purpose three alogrithms are proposed, namely Algorithm-1: GF-CWO for WLANs, Algorithm-2: GA for GF-CWO, and Algorithm-3: FLC for GF-CWO. Proposed GF-CWO approach is tested for binary exponential back-off (BEB), selected being a de facto standard algorithm, and for Channel Status based Sliding Contention Window (CS-SCW) algorithm, selected being a fuzzy logic based algorithm, implemented in MATLAB. Recorded the resultant values for 5, 10, 15, 20, 25, 30, 35, 40, 45, 50 nodes in different files for BEB, CS-SCW, and GF-CWO, respectively. Later on these files were used to evaluate success ratio, packet loss ratio, collision rate, fairness index and energy consumption and also generated the results graphically. The results generated through simulated test confirmed that the GF-CWO has effectively enhanced the performance.

RETRACTED ARTICLE: Trust model simulation of cross border e-commerce based on machine learning and Bayesian network

In recent years, many scholars at home and abroad have devoted themselves to the research of e-commerce trust, which has even become a hot topic in scientific research. When we explore the factors that affect consumers' choice of e-commerce transactions, e-commerce trust is undoubtedly an important breakthrough in this issue. E-commerce trust can have a profound impact on the popularity of e-commerce among customers and the future development. But in the current environment, consumers are still unfamiliar or skeptical about e-commerce. Based on these problems, this paper proposes a cross-border e-commerce trust model based on the existing trust evaluation technology, machine learning theory and the cloud model in Bayesian network technology to prevent fraud. With the support of cloud model, trust evaluation method has been improved. Trust has changed from qualitative to quantitative, and the relevant data can reflect the nature of trust more intuitively and objectively. In addition, the model can also start from the enterprise, according to the e-commerce enterprise for consumer privacy protection, information trading and other services to conduct a comprehensive evaluation of the enterprise, giving consumers a clearer choice. In addition, the evaluation quantity and adaptive time window algorithm are also introduced in the evaluation of e-commerce enterprises. All kinds of measures are to prevent illegal businesses from misleading consumers by credit fraud and speculation. Through the above methods, researchers can understand the trust degree of users for cross-border e-commerce, and put forward relevant methods to improve user credit. At the same time, through a variety of theoretical methods such as machine learning and Bayesian network, we can provide enterprises with more market and user demand information, make them aware of the shortcomings and constantly improve their business in the future, so as to improve the trust of users to enterprises, and attract a wider user group, so as to provide more ideas and guidance for the development of cross-border e-commerce.

Spatial and Multi-Temporal Analysis of Land Surface Temperature through Landsat 8 Images: Comparison of Algorithms in a Highly Polluted City (Granada)

Over the past decade, satellite imaging has become a habitual way to determine the land surface temperature (LST). One means entails the use of Landsat 8 images, for which mono window (MW), single channel (SC) and split window (SW) algorithms are needed. Knowing the precision and seasonal variability of the LST can improve urban climate alteration studies, which ultimately help make sustainable decisions in terms of the greater resilience of cities. In this study we determine the LST of a mid-sized city, Granada (Spain), applying six Landsat 8 algorithms that are validated using ambient temperatures. In addition to having a unique geographical location, this city has high pollution and high daily temperature variations, so that it is a very appropriate site for study. Altogether, 11 images with very low cloudiness were taken into account, distributed between November 2019 and October 2020. After data validation by means of R2 statistical analysis, the root mean square error (RMSE), mean bias error (MBE) and standard deviation (SD) were determined to obtain the coefficients of correlation. Panel data analysis is presented as a novel element with respect to the methods usually used. Results reveal that the SC algorithms prove more effective and reliable in determining the LST of the city studied here.

Development of a set of oligonucleotides for the identification of the Sakhalin sturgeon Acipenser mikadoi Hilgendorf, 1892 by PCR

Sturgeons (family Acipenseridae) are valuable commercial fish and aquaculture resources. The Acipenser genus includes about 20 species, 12 of which are recorded for the Russian territory. One of the rarest is the Sakhalin sturgeon A. mikadoi Hilgendorf, 1892. At present, its population size is significantly decreased and the species is close to extinction. Natural populations of the species have survived in the Tumnin River in Khabarovsk Region and the Viakhtu River in Sakhalin Region. Vihtu River in Sakhalin Region. Due to the small numbers of rare and endangered fish species, noninvasive approaches, including the use of DNA from the environment, appear to be the most suitable for monitoring their species and genetic diversity. The method is well established in this area and has been successfully tested for the monitoring of several species of rare and endangered sturgeons. This paper presents the results of the development of primers specific for the mitochondrial DNA of the species A. mikadoi. A matrix of 15 sequences of complete mitochondrial genomes from 3 sturgeon species whose ranges may overlap - A. mikadoi (KX276658), A. dauricus (KJ402277) and A. schrenckii (MH973728- MH973734, KX276660, KX276659, KF150287, KC905169, KC820796) - was generated. A. medirostris (NC_028405), which is genetically closest to the Sakhalin sturgeon, was also included in the analysis. A sliding window algorithm was applied to the sequence matrix. A graph showing the distribution of divergence values along the mitochondrial genome of sturgeon was plotted based on the results of the analysis Fig. 1. Based on matrices of individual genome fragments (16S rRNA, COI, ND1, and D-loop), a search for species-specific primers was performed in the DECIPHER program. A Primer-BLAST algorithm was run based on each pair of primers. Taxa Actinopterygii and Acipenseridae were used as references separately in different runs. In addition, primers were checked using local reference sequences of the complete mitochondrial genome of representatives of the genus Acipenser. The results of primer evaluation are summarized in the table Table 1. As a result, a combination of primer pairs developed on the basis of D-loop and COI fragments seems to be the most successful for error-free identification of Sakhalin sturgeon.

Improved Artificial Potential Field and Dynamic Window Method for Amphibious Robot Fish Path Planning

Aiming at the problems of “local minimum” and “unreachable target” existing in the traditional artificial potential field method in path planning, an improved artificial potential field method was proposed after analyzing the fundamental causes of the above problems. The method solved the problem of local minimum by modifying the direction and influence range of the gravitational field, increasing the virtual target and evaluation function, and the problem of unreachable targets is solved by increasing gravity. In view of the change of motion state of robot fish in amphibious environments, the improved artificial potential field method was fused with a dynamic window algorithm, and a dynamic window evaluation function of the optimal path was designed on the basis of establishing the dynamic equations of land and underwater. Then, the simulation experiment was designed under the environment of Matlab2019a. Firstly, the improved and traditional artificial potential field methods were compared. The results showed that the improved artificial potential field method could solve the above two problems well, shorten the operation time and path length, and have high efficiency. Secondly, the influence of different motion modes on path planning is verified, and the result also reflects that the amphibious robot can avoid obstacles flexibly and reach the target point accurately according to its own motion ability. This paper provides a new way of path planning for the amphibious robot.

Comparison of the accuracy of daytime land surface temperature retrieval methods using Landsat 8 images in arid regions

Land surface temperature (LST) is a significant parameter in surface and atmospheric systems which can be used to describe the physical processes of energy and water exchange. There are several ways to retrieve land surface temperature using satellite imagery. The purpose of the current study is to compare the accuracy of different LST retrieval algorithms, including 11 types of split-window algorithms, two mono-window algorithms, and three single-channel algorithms. The validation of LST retrieval algorithms was done using temperature-based and cross-validation methods. In this study, 20 images from Landsat 8 satellite related to 2017 and 2018 were used to retrieve LST in different ways. The temperatures obtained from different methods were compared in one day, and the average of LST of the different methods was about 5 ℃ different, which indicates the importance of the existing validation methods. Using Sentinel-2 image classification, 20 points in homogeneous areas with pure pixels were selected to measure the LST using a thermometer for land validation. Results showed that in a split window algorithm which was applied on the water vapor image, the RMSE error has been reduced by 2 ℃. In cross-validation, MODIS LST images were used, and RMSE temporal image was prepared, and classified into five groups including less than 2, 2–3, 3–4 and more than 5 ℃ .Comparing the area of each of these groups in different methods showed that MW2 and SW1 methods have less accuracy than the other methods. The SW9 algorithm can retrieve temperature with an error of less than 2 ℃ in 80% of the area. The results of cross-validation showed that the SC2 algorithm had the highest accuracy among the algorithms that use a thermal band for LST retrieval with a RMSE spatial of 5.3 °C and the SW10 and SW11 have shown the highest accuracy among the algorithms that use two thermal bands with RMSE less than 5 ℃. In general, the algorithms that use atmospheric transmittance coefficients and water vapor as inputs are more accurate than the others.

Validation of non-linear split window algorithm for land surface temperature estimation using Sentinel-3 satellite imagery: Case study; Tehran Province, Iran

In recent years, land surface temperature (LST) has become critical in environmental studies and earth science. Remote sensing technology enables spatiotemporal monitoring of this parameter on large scales. This parameter can be estimated by satellite images with at least one thermal band. Sentinel-3 SLSTR data provide LST products with a spatial resolution of 1 km. In this research, direct and indirect validation procedures were employed to evaluate the Sentinel-3 SLSTR LST products over the study area in different seasons from 2018 to 2019. The validation method was based on the absolute (direct) evaluation of this product with field data and comparison (indirect) evaluation with the MODIS LST product and the estimated LST using the non-linear split-window (NSW) algorithm. Also, two emissivity estimation methods, (1) NDVI thresholding method (NDVI-THM) and (2) classification-based emissivity method (CBEM), were used to estimate the LST using the NSW method according to the two thermal bands of Sentinel-3 images. Then, the accuracy of these methods in estimating LST was evaluated using field data and temporal changes of vegetation, which the NDVI-THM method generated better results. For indirect evaluation between the Sentinel-3 LST product, MODIS LST product, and LST estimated using NSW, four filters based on spatial and temporal separates between pairs of pixels and pixel quality were used to ensure the accuracy and consistency of the compared pairs of a pixel. In general, the accuracy results of the LST products of MODIS and Sentinel-3, and LST estimated using NSW showed a similar trend for LST changes during the seasons. With respect to the two absolute and comparative validations for the Sentinel-3 LST products, summer with the highest values of bias (−1.24 K), standard deviation (StDv = 2.66 K), and RMSE (2.43 K), and winter with the lowest ones (bias of 0.14 K, StDv of 1.13 K, and RMSE of 1.12 K) provided the worst and best results for the seasons in the period of 2018–2019, respectively. According to both absolute and comparative evaluation results, the Sentinel-3 SLSTR LST products provided reliable results for all seasons on a large temporal and spatial scale over our studied area.

A new method for long-term source apportionment with time-dependent factor profiles and uncertainty assessment using SoFi Pro: application to 1 year of organic aerosol data

Abstract. A new methodology for performing long-term source apportionment (SA) using positive matrix factorization (PMF) is presented. The method is implemented within the SoFi Pro software package and uses the multilinear engine (ME-2) as a PMF solver. The technique is applied to a 1-year aerosol chemical speciation monitor (ACSM) dataset from downtown Zurich, Switzerland. The measured organic aerosol mass spectra were analyzed by PMF using a small (14 d) and rolling PMF window to account for the temporal evolution of the sources. The rotational ambiguity is explored and the uncertainties of the PMF solutions were estimated. Factor–tracer correlations for averaged seasonal results from the rolling window analysis are higher than those retrieved from conventional PMF analyses of individual seasons, highlighting the improved performance of the rolling window algorithm for long-term data. In this study four to five factors were tested for every PMF window. Factor profiles for primary organic aerosol from traffic (HOA), cooking (COA) and biomass burning (BBOA) were constrained. Secondary organic aerosol was represented by either the combination of semi-volatile and low-volatility organic aerosol (SV-OOA and LV-OOA, respectively) or by a single OOA when this separation was not robust. This scheme led to roughly 40 000 PMF runs. Full visual inspection of all these PMF runs is unrealistic and is replaced by predefined user-selected criteria, which allow factor sorting and PMF run acceptance/rejection. The selected criteria for traffic (HOA) and BBOA were the correlation with equivalent black carbon from traffic (eBCtr) and the explained variation of m/z 60, respectively. COA was assessed by the prominence of a lunchtime concentration peak within the diurnal cycle. SV-OOA and LV-OOA were evaluated based on the fractions of m/z 43 and 44 in their respective factor profiles. Seasonal pre-tests revealed a non-continuous separation of OOA into SV-OOA and LV-OOA, in particular during the warm seasons. Therefore, a differentiation between four-factor solutions (HOA, COA, BBOA and OOA) and five-factor solutions (HOA, COA, BBOA, SV-OOA and LV-OOA) was also conducted based on the criterion for SV-OOA. HOA and COA contribute between 0.4–0.7 µg m−3 (7.8 %–9.0 %) and 0.7–1.2 µg m−3 (12.2 %–15.7 %) on average throughout the year, respectively. BBOA shows a strong yearly cycle with the lowest mean concentrations in summer (0.6 µg m−3, 12.0 %), slightly higher mean concentrations during spring and fall (1.0 and 1.5 µg m−3, or 15.6 % and 18.6 %, respectively), and the highest mean concentrations during winter (1.9 µg m−3, 25.0 %). In summer, OOA is separated into SV-OOA and LV-OOA, with mean concentrations of 1.4 µg m−3 (26.5 %) and 2.2 µg m−3 (40.3 %), respectively. For the remaining seasons the seasonal concentrations of SV-OOA, LV-OOA and OOA range from 0.3 to 1.1 µg m−3 (3.4 %–15.9 %), from 0.6 to 2.2 µg m−3 (7.7 %–33.7 %) and from 0.9 to 3.1 µg m−3 (13.7 %–39.9 %), respectively. The relative PMF errors modeled for this study for HOA, COA, BBOA, LV-OOA, SV-OOA and OOA are on average ±34 %, ±27 %, ±30 %, ±11 %, ±25 % and ±12 %, respectively.

A mine digital integrated protector with harmonic detection and analysis function

Aiming at the protection problems and harmonic detection problems of motor circuit, lighting signal circuit in coal mine, a mine digital comprehensive protector based on STM32 is designed. The protector can realize the protection and harmonic detection of motor circuit, lighting circuit and signal circuit in coal mine, and use windowed interpolation fast Fourier algorithm to calculate and analyze harmonic. The protector has the functions of under voltage, over-voltage, overload, short circuit, three-phase unbalance, open phase, leakage locking, insulation danger indication, over temperature, under temperature, wind power and gas locking, harmonic analysis, local storage, remote control and other functions. The experimental results show that the system design scheme is feasible and effective.

Spatial variation and threshold analysis of the urban wetland cooling effect-a case study of Changchun city

Due to the complexity of the surrounding environment, the cold and wet effects of urban wetlands and the quantitative descriptions of the cold and wet effects of the wetlands are not clear. Therefore, the focus of current research is gradually shifting towards the impact of the microspatial morphology and quantity differentiation of urban wetlands. This paper adopts the July 4, 2016 Landsat 8 data of cloud cover that is less than 0.2, the 10th band as the data source using single window algorithm surface temperature inversion, and further analysis of urban spatial distribution characteristics of land surface temperature. Secondly, based on the numerical analysis of the difference in surface temperature between urban wetland types, shapes, and independent wetland space sizes, the cold temperature effect and spatial distribution of urban wetlands were discussed. Temperature inversion results show that the lowest temperature is 17.94°C, which appears in the Songhuajiang wetland, and the highest temperature is 32.52°C, which appears in the FAW Industrial Concentration Zone in the main urban area of Changchun City. From the perspective of the spatial difference of the temperature distribution, the areas with high temperatures are generally distributed in the industrial concentration area in the southwest of Changchun, followed by other areas with high temperatures in the western part of Changchun. Low-temperature areas are mainly distributed in areas with relatively high distribution densities of rivers and lakes in the northeast and central parts of Changchun. This shows that the temperature is directly related to the urban building density. The temperature is higher in places with high building density. Wetland water bodies generally belong to low temperature areas and are generally 1-2°C lower than the surrounding environment. For the main urban area, the wetland water body is, on average, 2-4°C lower than the surrounding environment, while in areas with high building density, the wetland water temperature is 6-8°C lower than the surrounding area. And further urban wetlands with micro difference threshold temperature distribution characteristics are as follows: (1) The cold and wet effects of the wetland are related to its shape; and, the cooling and influence range of circular wetlands are larger than that of striped wetlands. (2) The cooling effect of a wetland in a circular wetland typified by a lake wetland is not related to the size of the water area when the area exceeds 92 hm2; and, the cooling range of a strip-shaped wetland has nothing to do with the width of the water surface. However, impact range is related to the width of the water surface, the larger the width of the river surface, the larger the range of influence. (3) The cooling effect of wetlands are related to the surrounding environment. When the surrounding area is hard ground, the cooling rate is large, and when the surrounding area is green land, the cooling rate is small.

A Nonintrusive Load Monitoring Method for Microgrid EMS Using Bi-LSTM Algorithm

Nonintrusive load monitoring in smart microgrids aims to obtain the energy consumption of individual appliances from the aggregated energy data, which is generally confronted with the error identification of the load type for energy disaggregation in microgrid energy management system (EMS). This paper proposes a classification strategy for the nonintrusive load identification scheme based on the bilateral long-term and short-term memory network (Bi-LSTM) algorithm. The sliding window algorithm is used to extract the detected load event features and obtain the load features of data samples. In order to accurately identify these load features, the steady state information is combined as the input of the Bi-LSTM model during training. Comprising long-term and short-term memory (LSTM) network and recurrent neural network (RNN), Bi-LSTM has the advantages of stronger recognition ability. Finally, precision (P), recall (R), accuracy (A), and F1 values are used as the evaluation method for nonintrusive load identification. The experimental results show the accuracy of the Bi-LSTM identification method for load start and stop state feature matching; moreover, the method can identify relatively low-power and multistate appliances.

Anomalies Detection Using Isolation in Concept-Drifting Data Streams

Detecting anomalies in streaming data is an important issue for many application domains, such as cybersecurity, natural disasters, or bank frauds. Different approaches have been designed in order to detect anomalies: statistics-based, isolation-based, clustering-based, etc. In this paper, we present a structured survey of the existing anomaly detection methods for data streams with a deep view on Isolation Forest (iForest). We first provide an implementation of Isolation Forest Anomalies detection in Stream Data (IForestASD), a variant of iForest for data streams. This implementation is built on top of scikit-multiflow (River), which is an open source machine learning framework for data streams containing a single anomaly detection algorithm in data streams, called Streaming half-space trees. We performed experiments on different real and well known data sets in order to compare the performance of our implementation of IForestASD and half-space trees. Moreover, we extended the IForestASD algorithm to handle drifting data by proposing three algorithms that involve two main well known drift detection methods: ADWIN and KSWIN. ADWIN is an adaptive sliding window algorithm for detecting change in a data stream. KSWIN is a more recent method and it refers to the Kolmogorov–Smirnov Windowing method for concept drift detection. More precisely, we extended KSWIN to be able to deal with n-dimensional data streams. We validated and compared all of the proposed methods on both real and synthetic data sets. In particular, we evaluated the F1-score, the execution time, and the memory consumption. The experiments show that our extensions have lower resource consumption than the original version of IForestASD with a similar or better detection efficiency.