Kernel Density Estimation Method Research Articles

The Structural Causes and Trend Evolution of Imbalance and Insufficiency of Development of Digital Inclusive Finance in China

In this paper, to provide references for coordinating the development of digital inclusive finance between provinces and regions in China, we measure the digital financial inclusion index of 31 provinces in China from 2011 to 2020 based on three dimensions: coverage breadth, depth of use and digitalization degree. By means of the weighted Dagum Gini coefficient and quantile standardization, we explored the degree of imbalance and insufficiency of the development of digital inclusive finance in China and four major regions and its structural causes. Using the Kernel density estimation method and Markov chain analysis method, we further investigate the evolution trend of imbalance and insufficiency. The study finds that (1) the Digital Inclusive Financial Index in China and the four major regions rose significantly, with the COVID-19 epidemic reducing its growth rate. Of these, the eastern region has the highest development level. (2) The imbalance level of digital inclusive finance development obviously has reduced. The level of imbalance is highest within the eastern region, and the development gap between the eastern and western regions is the widest. The imbalance of overall development is mainly due to regional imbalance. The imbalance of coverage breadth and depth of use is the main structural cause of unbalanced development in the four major regions. There is a trend of bipolarization or multipolarization in China and the other three major regions, with the exception of the central region. (3) The western region is the least developed. The development shortcoming of digital inclusive finance in China and the four major regions is the breadth of coverage. There are the “Club Convergence” and the “Matthew Effect” in the eastern, central and western regions.

Spatiotemporal evolution pattern and simulation of the coupling of carbon productivity and land development in the Yangtze River Delta, China

It is important to explore the spatiotemporal characteristics of the coupling of carbon productivity (CP) and the land development degree (LD) in Chinese counties and to simulate the development trend of their collaborative evolution. In this study, first, the spatiotemporal evolution patterns of the regional CP and LD in the Yangtze River Delta from 2003 to 2017 were analyzed. Then, the coupling coordination model and kernel density estimation method were used to reveal their coupling characteristics. Finally, two types of Markov transition matrices were constructed to empirically analyze the dynamic change rule of their coordinated development, and their long-term evolution trends were simulated under different neighborhood characteristics. The results revealed that the CP initially decreased and then rapidly increased from 2003 to 2017, while the LD continuously decreased from 13.2 thousand people/km2 in 2003 to 10.9 thousand people/km2 in 2017. The coupling coordination degree (CCD) between the CP and LD exhibited a two-stage decline, followed by a rise over the past 15 years, with the CCD level shifting downward in 19.68% of the counties and upward in 19.05% of the counties. The transition of the CCD level had a significant spatial spillover effect. If adjacent to counties with a low CCD for a long period of time, the proportion of the counties with a low CCD would increase to 61.65%, while long-term neighboring counties with a high CCD would eventually stabilize the percentage of counties with a high CCD at 61.83%. Incorporating the spatial spillover effect into low-carbon land management at the county level can offer scientific direction for advancing low-carbon land development.

Efficient Non-parametric Neural Density Estimation and Its Application to Outlier and Anomaly Detection

The main goal of this thesis is to develop efficient non-parametric density estimation methods that can be integrated with deep learning architectures, for instance, convolutional neural networks and transformers. Density estimation methods can be applied to different problems in statistics and machine learning. They may be used to solve tasks such as anomaly detection, generative models, semi-supervised learning, compression, text-to-speech, among others. The present work will mainly focus on the application of the method in anomaly and outlier detection tasks such as medical anomaly detection, fraud detection, video surveillance, time series anomaly detection, industrial damage detection, among others. A recent approach to non-parametric density estimation is neural density estimation. One advantage of these methods is that they can be integrated with deep learning architectures and trained using gradient descent. Most of these methods are based on neural network implementations of normalizing flows which transform an original simpler distribution to a more complex one. The approach of this thesis is based on a different idea that combines random Fourier features with density matrices to estimate the underlying distribution function. The method can be seen as an approximation of the popular kernel density estimation method but without the inherent computational cost.

An adaptive binned kernel density estimation method for analysis of wave energy converters

ABSTRACT This research proposes a new adaptive binned Kernel Density Estimation (KDE) method to fit significant wave height data in order to more accurately calculate the sea state parameter distribution tails. The brief details of the proposed new methodology are to utilize the fast Fourier transform to perform the convolution of the data with the kernel to find the kernel density estimates on the bin nodes. The Crámer-von Mises test statistic for the proposed adaptive KDE distribution is calculated to be 0.0926 based on a measured wave data set. The Crámer-von Mises test statistic value for the fitted 3-parameter Weibull distribution based on the identical measured data set is 251.2463, indicating a much poorer fit to the empirical distribution function. The advantages of using a more reliable and accurate contour line derived using the proposed new method for long term reliability analysis of wave energy converters have been finally substantiated.

Surface Wildfire Regime and Simulation-Based Wildfire Exposure in the Golestan National Park, NE Iran

This research analyzes the spatiotemporal patterns of wildfire regime attributes (e.g., seasonality, size, frequency, and burn rate) across the Golestan National Park (GNP), northeast Iran over the last two decades. We used a variety of data, including existing vegetation data, current vegetation survey, and historical wildfire data, and then data were processed through ArcMap. We also predicted fire exposure profiles (burn probability (BP), conditional flame length (CFL (m)), and fire size (FS (ha)) by the application of the minimum travel time (MTT) fire spread algorithm. The kernel density estimation (KDE) method was used to estimate wildfire likelihood, based on recent wildfires (2000–2020) that occurred in the GNP. Finally, we developed a logistic regression model to investigate how independent variables such as weather, fuel, and topographic data influence wildfires in the park. Wildfires in the landscape have not been constant in either space or time. Their extent, seasonality, frequency, and other wildfire regime characters varied considerably across the landscape. Our results highlighted that shrublands in the southern part of the park showed, in general, the highest values in terms of the wildfire regime attributes. Large fires (10–100 ha, 51%) and very large fires (>100 ha, 24%), fire intervals greater than 10 years (90%), and high burn rates (>1% y−1, 35%) are all characteristics that contribute to high wildfire activity in shrublands. Similarly, areas predicted to have high wildfire exposure levels (average BP = 0.004; average CFL = 1.60 m; average FS = 840 ha) are found in the fuel models of high-load grass and medium-load shrub. Finally, the regression model results revealed that weather and fuel were the most influential parameters (R2 ≥ 0.2), while topography had comparatively less influence in the study area. In light of these results, we suggest proactively incorporating this information into fire and fuel management which can help develop a fire prevention plan, predict fire ignition probability and frequency, and finally address altered fire regimes threatening the park.

Spatial Disequilibrium and Dynamic Evolution of Eco-Efficiency in China’s Tea Industry

Eco-efficiency is a significant target for evaluating the agricultural ecosystem and measuring sustainable agricultural development through quantitative analysis. It is also an essential part of constructing the ecological tea garden, which offers a directional function in realizing the green development of the tea industry. After measuring the eco-efficiency of China’s tea industry using the super-efficiency SBM model, this paper analyzes the spatial disequilibrium and dynamic evolution trend of the eco-efficiency in China’s tea industry through the method of Dagum Gini Coefficient and Kernel Density Estimation. The results show that the level of eco-efficiency in China’s tea industry was improved overall, and the spatial disequilibrium was significantly reduced. The differences within the tea region decreased as follows: tea regions in Southwest China, South China, south of the Yangtze River, and north of the Yangtze River; the overall difference in the eco-efficiency in the tea industry mainly comes from the contribution of the interregional difference in tea regions, and the second contribution comes from the intraregional difference in tea regions and the difference in super-variable density. The eco-efficiency of the tea industry has been improved both nationally and within the top four tea regions; the disequilibrium between areas and within the tea region has been largely alleviated, but there is still room to optimize the input–output structure and promote the eco-efficiency.

Fault detection for NOx emission process in thermal power plants using SIP-PCA

With the era of big data, data-driven models are increasingly vital to just-in-time decision support in pollution emission management and planning. This article aims to evaluate the usability of the proposed data-driven model to monitor NOx emission from a coal-fired boiler process using easily measured process variables. As the emission process is highly complex, process variables interact with each other, and they cannot guarantee that all the variables in the actual operation obey the Gaussian distributions. As conventional principal component analysis (PCA) can only extract variance information, a novel data-driven model is proposed, called survival information potential-based PCA (SIP-PCA) model, is proposed in this work. First, an improved PCA model is established based on the SIP performance index. SIP-PCA can extract more information in the latent space from the process variables following the non-Gaussian distributions. Then, the control limits for fault detection are determined based on the kernel density estimation method. Finally, the proposed algorithm is successfully applied to a real NOx emission process. By monitoring the operation of process variables, possible failures can be detected as soon as possible. Fault isolation and system reconstruction can be implemented in time, preventing NOx emissions from exceeding its standard.

A Semi-Automated Two-Step Building Stock Monitoring Methodology for Supporting Immediate Solutions in Urban Issues

The Sustainable Development Goals (SDGs) have addressed environmental and social issues in cities, such as insecure land tenure, climate change, and vulnerability to natural disasters. SDGs have motivated authorities to adopt urban land policies that support the quality and safety of urban life. Reliable, accurate, and up-to-date building information should be provided to develop effective land policies to solve the challenges of urbanization. Creating comprehensive and effective systems for land management in urban areas requires a significant long-term effort. However, some procedures should be undertaken immediately to mitigate the potential negative impacts of urban problems on human life. In developing countries, public records may not reflect the current status of buildings. Thus, implementing an automated and rapid building monitoring system using the potential of high-spatial-resolution satellite images and street views may be ideal for urban areas. This study proposed a two-step automated building stock monitoring mechanism. Our proposed method can identify critical building features, such as the building footprint and the number of floors. In the first step, buildings were automatically detected by using the object-based image analysis (OBIA) method on high-resolution spatial satellite images. In the second step, vertical images of the buildings were collected. Then, the number of the building floors was determined automatically using Google Street View Images (GSVI) via the YOLOv5 algorithm and the kernel density estimation method. The first step of the experiment was applied to the high-resolution images of the Pleiades satellite, which covers three different urban areas in Istanbul. The average accuracy metrics of the OBIA experiment for Area 1, Area 2, and Area 3 were 92.74%, 92.23%, and 92.92%, respectively. The second step of the experiment was applied to the image dataset containing the GSVIs of several buildings in different Istanbul streets. The perspective effect, the presence of more than one building in the photograph, some obstacles around the buildings, and different window sizes caused errors in the floor estimations. For this reason, the operator’s manual interpretation when obtaining SVIs increases the floor estimation accuracy. The proposed algorithm estimates the number of floors at a rate of 79.2% accuracy for the SVIs collected by operator interpretation. Consequently, our methodology can easily be used to monitor and document the critical features of the existing buildings. This approach can support an immediate emergency action plan to reduce the possible losses caused by urban problems. In addition, this method can be utilized to analyze the previous conditions after damage or losses occur.

Wind Power Interval Prediction Based on Robust Kernel Density Estimation

Wind power output has a high degree of randomness, so it is difficult to describe it accurately with some typical probability distribution. The extreme values influence the sample’s general non-parametric kernel density estimation method, and the estimated results are relatively conservative. A wind power interval prediction method based on robust kernel density estimation is proposed to improve the compactness and accuracy of interval prediction. In probability estimation, this method will assign a small weight to the extreme sample data to reduce its influence on the probability density estimation accuracy and has better robustness than the general kernel density estimation method. In addition, the bandwidth of robust kernel density estimation is a dynamic parameter, which can be adjusted with the sample data to improve the prediction accuracy and avoid over-conservative prediction.

Fault Detection of Wastewater Treatment Plants Based on an Improved Kernel Extreme Learning Machine Method

In order to ensure the stable operation, improve efficiency, and enhance sustainability of wastewater treatment systems, this paper investigates the fault detection problem in wastewater treatment process based on an improved kernel extreme learning machine method. Firstly, a kernel extreme learning machine (KELM) model optimized by an improved mutation bald eagle search (IMBES) optimizer is proposed to generate point predictions of effluent quality parameters. Then, based on the point prediction results, the confidence interval of effluent quality parameters is calculated using kernel density estimation (KDE) method. This interval represents the bounds of system uncertainty and unknown disturbance at normal conditions and can be treated as the threshold for fault diagnosis. Finally, the effectiveness of the proposed method is illustrated by two datasets obtained from the BSM1 wastewater simulation platform and an actual water platform. Experimental results show that compared with other methods such as CNN, LSTM, and IBES-LSSVM, this method has a significant improvement in prediction accuracy, and at the same confidence level, it ensures fault detection rate while generating smaller confidence intervals.

Extraction of Building Information Based on Multi-Source Spatiotemporal Data for Earthquake Insurance in Urban Areas

Establishing a database of building exposures is an important basic work in earthquake insurance research. How to efficiently, accurately, and scientifically construct the risk exposure database of buildings has become a hot topic these days. Based on multi-source data, a system for extracting seismic information from urban buildings was constructed in the Tangshan urban area, and a perfect earthquake insurance risk database was established in this study. In the extraction system, the U-net identification method, spatial overlay and kernel density estimation method, Kriging interpolation method, statistical analysis, and multi-temporal land cover data analysis were used, respectively, to extract the information of footprint areas, use function, story number, structure type, and construction age of the urban buildings. The extraction results are stratified and randomly sampled, and the confusion matrix is introduced to verify the extraction effect. The results show that the building covers an area of about 50 million square meters in the urban area of Tangshan City. With the training and validation of the U-net model, the global accuracy of the building footprint areas recognition model is 71%. By comparing the results of manually determined real data with the extraction results of this study for a sample of 660 buildings, it was found that the overall accuracy rates of the extraction results of building function, story number, structure type, and construction age were 88.62%, 86.65%, 86.49%, and 85.58%, respectively, and kappa coefficients were all over 0.8. These indicate that the information on buildings extracted by the method of this study is accurate and reliable. This study can provide data and methods for the establishment of the exposure database of earthquake insurance and provide strong data support for pre-earthquake disaster prevention, post-earthquake emergency rescue, and disaster loss assessment.

Determination of the impacts of urban-planning of the urban land area using GIS hotspot analysis

Public investments are commonly linked with urban growth and change in points of interest (POI). This study aims to detect the changes in 5 POI categories within 250, 500, and 1000 m catchment areas around major public investments and transportation lines between 2008 and 2019 in Eskisehir city, Turkey. Due to the high number of POI data in the area, hot spot maps were produced for the target years with Kernel Density Estimation (KDE) method and subtracted with raster operation to create the POI differencing maps. The results revealed that establishing the tram line (ESTRAM) affected the emergence of POIs from 2008 to 2019, while no significant influence of other public investments on the POIs was detected in the city. Still, some investments, including Atatürk Culture and Convention Center, Şelalepark, the Zoo, and the City Hospital, had limited impacts on POIs. This research is expected to contribute to the prediction of the effects of public investments on the spatial and functional development processes and provide an analytical framework.

Trivariate Kernel Density Estimation of Spatiotemporal Crime Events with Case Study for Lithuania

The paper presents the results of the investigation of the applicability of spatiotemporal kernel density estimation (KDE) methods for density mapping of violent crime in Lithuania. Spatiotemporal crime research helps to understand and control specific types of crime, thereby contributing to Sustainable Development Goals. The target dataset contained 135,989 records of the events registered by the police of Lithuania from 2015–2018 that were classified as violent. The research focused on choosing appropriate KDE functions and their parameters for modeling the spatiotemporal point pattern of this particular type of crime. The aim was to estimate density, mass, and intensity function(s) so that they can be used in further confirmatory spatial modeling. The application-driven objective was to obtain reliable and practically interpretable KDE surfaces of crime events. Several options for improving and extending the investigated KDE methods are demonstrated.

Stochastic semi-analytical modeling of reinforced filaments for additive manufacturing

This study presents a novel stochastic modeling approach that addresses two challenges encountered in micromechanical modeling of short-fiber composites. Firstly, the challenge of the time-consuming pre-processing required for extracting fibers from micro CT scans is tackled by introducing a new stochastic generation technique based on the kernel density estimation (KDE) method. This enables the generation of artificial fibers for micromechanical models, thus saving considerable time and effort. Secondly, the challenge of presenting a modeling approach that considers multiple fibers while reducing the computational effort associated with the simulation is addressed through a novel semi-analytical approach. To demonstrate the effectiveness of the stochastic modeling approach, it is applied to filaments of recycled poly(ethylene terephthalate) reinforced with recycled short carbon fibers that are used for additive manufacturing of composite parts. The results obtained from the stochastic modeling approach are compared with those from a direct modeling approach that considers 1050 fibers extracted from a micro CT scan. The novel approach is shown to provide similar predictions of elastic properties as the direct modeling approach while using only 40–50 fibers. Furthermore, the results are in close agreement with experimental data, highlighting the effectiveness of the proposed approach.

Division and Analysis of Accident-Prone Areas near Highway Ramps Based on Spatial Autocorrelation

This study focuses on identifying accident-prone areas and analyzing the factors contributing to the distribution of traffic accidents near highway ramps. A combined method of kernel density estimation, spatial autocorrelation analysis, and multivariate logistic regression analysis helped to identify accident hotspots. Through data collection and analysis, the clustering characteristics of traffic accidents in the diversion and merging areas were identified. Four levels of accident-prone areas were divided according to the accident rates. The factors influencing the spatial distribution of accidents were analyzed. The results showed that traffic accidents in the diversion area were concentrated near the exit, but the accidents in merging areas had a wider range of distribution. The analysis of this phenomenon was conducted using the multinomial logit model results. The important factors of different accident-prone areas were clarified. The temperature, the accident lane, weather conditions, and the time of day had significant impacts on the spatial distribution of traffic accidents. The study’s findings provide an important decision-making basis for highway accident prevention management.

The In-Situ Spatial-Temporal Evolution of the Settlement Space along the Grand Canal Tianjin Section from the Perspective of Cultural Heritage

The Grand Canal of China, as a traffic artery in ancient China, has exerted profound influence on the development of the cities, towns and rural settlements along it. Now, with China’s rapid urbanization, numerous settlements along the canal are confronted with problems like slow cognition of the cultural heritage, loss of the local culture and damage of historical relics. Hence, research on the protection and updating of the settlement space along the canal is in urgent need. This paper, targeting the settlement space along the Grand Canal Tianjin Section, adopts such research methods as the Historical GIS analytical method, the kernel density estimation method and the average nearest-neighbor method to quantitatively analyze the evolution stage and distribution characteristics of the settlement space along the Grand Canal Tianjin Section from the perspective of cultural relics to reveal the in-situ rules of evolution of the settlement space. The research result indicates that the spatial-temporal evolution of the settlement space can be divided into five such periods of the settlements along the canal as the pre-canal period, the rudimentary period, the development period, the mature period and the transformation period. Each period features different cultural relics and settlement distribution patterns. The research shows that the natural geographical environment, the grain transportation projects and management systems, the wasteland opening, grain growing and military defense policies, the agricultural, commercial and economic growth and the migration and folk culture are the influencing factors of the spatial evolution of the settlement space along the Grand Canal Tianjin Section.

Composite power system risk evaluation considering the accurate model of renewable power output

The fluctuation of renewable energy will threaten the reliable operation of the composite transmission and generation power system, so the risk level of the composite system with renewable energy attached must be studied. However, the accuracy of the renewable energy model has typically been overlooked in current risk assessment approaches. Therefore, this study provides an accurate hybrid wind/solar power output model that takes the wind farm’s seasonal and spatial characteristics, as well as the correlation between wind and solar PV power into account. For the first, build the seasonal Weibull model of the wind speed and correct the speed captured by each wind turbine considering different spatial characteristics of the wind farm. Secondly, build the probability distribution function of the solar power based on the kernel density estimation method. In the next, construct the joint probability distribution function considering the correlation between wind and solar PV power based on copula theory. Then, using the suggested accurate renewable energy model, the sequential Cross-entropy (SCE) technique is utilized to evaluate the risk indices of the composite system with renewable energy. To validate the models and technique, the modified IEEE-RTS79 is studied. The findings indicate that the accuracy of the renewable energy model influences the risk assessment outcomes.

A Novel Operating State Evaluation Method for Photovoltaic Strings Based on TOPSIS and Its Application

PV strings are essential for energy conversion in large-scale photovoltaic (PV) power plants. The operating state of PV strings directly affects the power generation efficiency and economic benefits of PV power plants. In the process of evaluating PV arrays, a reference array needs to be identified. By comparing PV arrays with the reference array, the operational status of the PV arrays can be evaluated. However, in the actual operation of PV power stations, it is difficult to directly determine the reference state of a PV array due to random fluctuations in the PV power output. In order to solve the problems mentioned above, this paper proposes a method to select the reference state and perform a grading evaluation of PV strings. Additionally, the proposed method is based on the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) algorithm, which is used to rank the performance of PV arrays to determine their status. In order to solve the problem of random fluctuations in PV power generation, a probability distribution model of the PV string conversion efficiency was built by using the kernel density estimation method. Then, the characteristic indicator of the PV string’s operating state was described by the output power of the PV string and its probability distribution model. Then, based on the operating characteristic indicator, the reference state of the PV string was determined using the TOPSIS method, and the grading evaluation of the operating state of the PV string was realized. Finally, the effectiveness of the proposed method was verified using the actual data of a PV power station.

Research on Spatial Pattern of Henan Pension Institutions Based on Accessibility

With the deepening of aging, pension institutions have gradually become the focus of research, and the reasonable distributions and locations of pension institutions have great impact on their own functions. Taking Henan Province as the study area, the spatial distribution characteristics of pension institutions are analyzed by the nearest neighbor index method and the kernel density estimation method, the accessibility of pension institutions with a single service radius and its influencing factors are analyzed by improved potential model and two-step floating catchment area method. The results are that: (1) the pension institutions in Henan Province generally show the characteristics of “central cohesion-denseness in the north and sparseness in the south”, and have the trend of development in marginal areas. (2) The spatial accessibility of pension institutions decreases from northeast to southwest, positively affected by economy, public service, per capita income of residents and negatively impacted by the elderly population. (3) The improved potential model is more suitable to evaluate the spatial accessibility of pension institutions in Henan Province.

Adaptive fusion affinity graph with noise-free online low-rank representation for natural image segmentation

Affinity graph-based segmentation methods have become a major trend in computer vision. The performance of these methods rely on the constructed affinity graph, with particular emphasis on the neighborhood topology and pairwise affinities among superpixels. However, these graph-based methods ignore the noisy data from images, that influence the accuracy of pairwise similarities. Multiscale combinatorial grouping and graph fusion also generate a higher computational complexity. In this paper, we propose an adaptive fusion affinity graph with noise-free low-rank representation in an online manner for natural image segmentation. An input image is first over-segmented into superpixels at different scales and then filtered by an improved kernel density estimation method. Moreover, we select global nodes of these superpixels on the basis of their subspace-preserving presentation, which reveals the feature distribution of superpixels exactly. To reduce time complexity while improving performance, a sparse representation of global nodes based on noise-free online low-rank representation is used to obtain a global graph at each scale. Experimental results on BSD300, BSD500, MSRC, SBD, and PASCAL VOC show the effectiveness of our method in comparison with the state-of-the-art approaches. The code is available at https://github.com/Yangzhangcst/AFA-graph.