A Review on Clustering Methods for Climatology Analysis and Its Application over South America

South America’s climatic diversity is a product of its vast geographical expanse, encompassing tropical to subtropical latitudes. The variations in precipitation and temperature across the region stem from the influence of distinct atmospheric systems. While some studies have characterized the prevailing systems over South America, they often lacked the utilization of statistical techniques for homogenization. On the other hand, other research has employed multivariate statistical methods to identify homogeneous regions regarding temperature and precipitation, but their focus has been limited to specific areas, such as the south, southeast, and northeast. Surprisingly, there is a lack of work that compares various multivariate statistical techniques to determine homogeneous regions across the entirety of South America concerning temperature and precipitation. This paper aims to address this gap by comparing three such techniques: Cluster Analysis (K-means and Ward) and Self Organizing Maps, using data from different sources for temperature (ERA5, ERA5-Land, and CRU) and precipitation (ERA5, ERA5-Land, and CPC). Spatial patterns and time series were generated for each region over the period 1981-2010. The results from this analysis of spatially homogeneous regions concerning temperature and precipitation have the potential to significantly benefit climate analysis and forecasts. Moreover, they can offer valuable insights for various climatological studies, guiding decision-making processes in diverse fields that rely on climate information, such as agriculture, disaster management, and water resources planning.

An unsupervised clustering method for nuclear magnetic resonance transverse relaxation spectrums based on the Gaussian mixture model and its application

To make the quantitative results of nuclear magnetic resonance(NMR) transverse relaxation(T;) spectrums reflect the type and pore structure of reservoir more directly, an unsupervised clustering method was developed to obtain the quantitative pore structure information from the NMR T;spectrums based on the Gaussian mixture model(GMM). Firstly, We conducted the principal component analysis on T;spectrums in order to reduce the dimension data and the dependence of the original variables. Secondly, the dimension-reduced data was fitted using the GMM probability density function, and the model parameters and optimal clustering numbers were obtained according to the expectation-maximization algorithm and the change of the Akaike information criterion. Finally, the T;spectrum features and pore structure types of different clustering groups were analyzed and compared with T;geometric mean and T;arithmetic mean. The effectiveness of the algorithm has been verified by numerical simulation and field NMR logging data. The research shows that the clustering results based on GMM method have good correlations with the shape and distribution of the T;spectrum, pore structure, and petroleum productivity, providing a new means for quantitative identification of pore structure, reservoir grading, and oil and gas productivity evaluation.

Application of Systematic Clustering Method in the Classification and Changes of Atmospheric Ozone in Hunan Province

The cluster analysis method needs continuous improvement and perfection in the research and application of the spatial differentiation and change of pollutants.In this paper,the date of monthly highest concentration of ozone(O_(3))and the concentration value of that day were selected as the similarity coefficient between classes.Single-factor cluster analysis was performed on O_(3)during 2016-2019 and the COVID-19 outbreak of 2020 in Hunan Province using the Ward method.The clustering results showed that the spatial distribution of atmospheric O_(3)in the 14 regions of Hunan Province was most suitable to be classified according to class III clustering areas.That is,the Changsha-Zhuzhou-Xiangtan urban agglomeration was the center,and the high-value area was in northern Hunan.The transition area was in central and southern Hunan,while the low-value area was centered in western Hunan.The partition results were in good agreement with the homogeneous subset of one-way ANOVA and the distribution of monitoring values during the same period.The comparison showed that the inter-class plates in the two periods corresponded well,and the intra-class area showed a continuous geographical distribution,and there were dynamic changes in the spatial differentiation of the O_(3)plates in different periods.In 2020,the center of the O_(3)high-value area plate in Hunan Province moved eastward and extended southward,focusing on the middle and lower reaches of the Xiangjiang River basin,and extending to the upstream area;the regional plate in the transition area expanded significantly;the low-value area plate shrank to the two cities in western Hunan.The abnormal emissions and abnormal climate during the COVID-19 epidemic had an impact on the spatial differentiation of O_(3)in Hunan Province.

A Clustering Method Based on Brain Storm Optimization Algorithm

In the field of data mining and machine learning,clustering is a typical issue which has been widely studied by many researchers,and lots of effective algorithms have been proposed,including K-means,fuzzy c-means(FCM)and DBSCAN.However,the traditional clustering methods are easily trapped into local optimum.Thus,many evolutionary-based clustering methods have been investigated.Considering the effectiveness of brain storm optimization(BSO)in increasing the diversity while the diversity optimization is performed,in this paper,we propose a new clustering model based on BSO to use the global ability of BSO.In our experiment,we apply the novel binary model to solve the problem.During the period of processing data,BSO was mainly utilized for iteration.Also,in the process of K-means,we set the more appropriate parameters selected to match it greatly.Four datasets were used in our experiment.In our model,BSO was first introduced in solving the clustering problem.With the algorithm running on each dataset repeatedly,our experimental results have obtained good convergence and diversity.In addition,by comparing the results with other clustering models,the BSO clustering model also guarantees high accuracy.Therefore,from many aspects,the simulation results show that the model of this paper has good performance.

A Research on Competitiveness of Guangxi City——Based on System Clustering Method and Principal Component Analysis Method

A total of 10 indices of regional economic development in Guangxi are selected.According to the relevant economic data,regional economic development in Guangxi is analyzed by using System Clustering Method and Principal Component Analysis Method.Result shows that System Clustering Method and Principal Component Analysis Method have revealed similar results analysis of economic development level.Overall economic strength of Guangxi is weak and Nanning has relatively high scores of factors due to its advantage of the political,economic and cultural center.Comprehensive scores of other regions are all lower than 1,which has big gap with the development of Nanning.Overall development strategy points out that Guangxi should accelerate the construction of the Ring Northern Bay Economic Zone,create a strong logistics system having strategic significance to national development,use the unique location advantage and rely on the modern transportation system to establish a logistics center and business center connecting the hinterland and the Asean Market.Based on the problems of unbalanced regional economic development in Guangxi,we should speed up the development of service industry in Nanning,construct the circular economy system of industrial city,and accelerate the industrialization process of tourism city in order to realize balanced development of regional economy in Guangxi,China.

Impacts of climate change on seasonal extreme waves in the Northwest Atlantic using a Spatial Neural Gas clustering method

Having estimates of wave climate parameters and extreme values play important roles for a variety of different societal activities,such as coastal management,design of inshore and offshore structures,marine transport,coastal recreational activities,fisheries,etc.This study investigates the efficiency of a state-of-the-art spatial neutral gas clustering method in the classification of wind/wave data and the evaluation of extreme values of significant wave heights(Hs),mean wave direction(MWD)and mean wave periods(T0)for two 39-year time periods;from 1979 to 2017 for the present climate,and from 2060 to 2098,for a future climate change scenario in the Northwest Atlantic.These data were constructed by application of a numerical model,WAVEWATCHIII TM(hereafter,WW3),to simulate the wave climate for the study area for both present and future climates.Data from the model was extracted for the wave climate,in terms of the wave parameters,specifically Hs,MWD and T0,which were analyzed and compared for winter and summer seasons,for present and future climates.In order to estimate extreme values in the study area,a Natural Gas(hereafter,NG)clustering method was applied,separate clusters were identified,and corresponding centroid points were determined.To analyze data at each centroid point,time series of wave parameters were extracted,and using standard stochastic models,such as Gumbel,exponential and Weibull distribution functions,the extreme values for 50 and 100-year return periods were estimated.Thus,the impacts of climate change on wave regimes and extreme values can be specified.

Global Optimization Method Using SLE and Adaptive RBF Based on Fuzzy Clustering

High fidelity analysis models,which are beneficial to improving the design quality,have been more and more widely utilized in the modern engineering design optimization problems.However,the high fidelity analysis models are so computationally expensive that the time required in design optimization is usually unacceptable.In order to improve the efficiency of optimization involving high fidelity analysis models,the optimization efficiency can be upgraded through applying surrogates to approximate the computationally expensive models,which can greately reduce the computation time.An efficient heuristic global optimization method using adaptive radial basis function（RBF） based on fuzzy clustering（ARFC） is proposed.In this method,a novel algorithm of maximin Latin hypercube design using successive local enumeration（SLE） is employed to obtain sample points with good performance in both space-filling and projective uniformity properties,which does a great deal of good to metamodels accuracy.RBF method is adopted for constructing the metamodels,and with the increasing the number of sample points the approximation accuracy of RBF is gradually enhanced.The fuzzy c-means clustering method is applied to identify the reduced attractive regions in the original design space.The numerical benchmark examples are used for validating the performance of ARFC.The results demonstrates that for most application examples the global optima are effectively obtained and comparison with adaptive response surface method（ARSM） proves that the proposed method can intuitively capture promising design regions and can efficiently identify the global or near-global design optimum.This method improves the efficiency and global convergence of the optimization problems,and gives a new optimization strategy for engineering design optimization problems involving computationally expensive models.

Track-Pattern-Based Characteristics of Extratropical Transitioning Tropical Cyclones in the Western North Pacific

Based on the Regional Specialized Meteorological Center(RSMC)Tokyo-Typhoon Center best-track data and the NCEP-NCAR reanalysis dataset,extratropical transitioning(ET)tropical cyclones(ETCs)over the western North Pacific(WNP)during 1951–2021 are classified into six clusters using the fuzzy c-means clustering method(FCM)according to their track patterns.The characteristics of the six hard-clustered ETCs with the highest membership coefficient are shown.Most tropical cyclones(TCs)that were assigned to clusters C2,C5,and C6 made landfall over eastern Asian countries,which severely threatened these regions.Among landfalling TCs,93.2%completed their ET after landfall,whereas 39.8%of ETCs completed their transition within one day.The frequency of ETCs over the WNP has decreased in the past four decades,wherein cluster C5 demonstrated a significant decrease on both interannual and interdecadal timescales with the expansion and intensification of the western Pacific subtropical high(WPSH).This large-scale circulation pattern is favorable for C2 and causes it to become the dominant track pattern,owning to it containing the largest number of intensifying ETCs among the six clusters,a number that has increased insignificantly over the past four decades.The surface roughness variation and three-dimensional background circulation led to C5 containing the maximum number of landfalling TCs and a minimum number of intensifying ETCs.Our results will facilitate a better understanding of the spatiotemporal distributions of ET events and associated environment background fields,which will benefit the effective monitoring of these events over the WNP.

Grey Clustering Evaluation on Regional Eco-environmental Quality Based on Normalized Index Value

[Objective] The aim was to assess regional eco-environmental quality by means of grey clustering method based on normalized index value. [Method] Eco-environmental quality in Chaohu basin was assessed by using grey clustering method based on normalized index value, and the evaluation results were compared with those of unascertained measure method to verify the feasibility of grey clustering method used to evaluate regional eco-environmental quality. [Result] In the grey clustering assessment method based on normalized index value, indices whose standard normalized values in the same grade were close to each other were classified into one class and had the same whitening function, which reduced the number of whitening functions. Grey clustering method based on normalized index value was used to assess eco-environmental quality in Chaohu basin, and the evaluation results were basically in accordance with those of unascertained measure method, namely eco-environmental quality in Hefei, Chaohu and Lu’an belonged to the third (pass), fourth (worse) and fifth grade (bad), except for one grade difference in overall basin, and the results showed that the method had practicality and could be applied to assess regional eco-environmental quality. [Conclusion] The study could provide theoretical foundation for the establishment of comprehensive management countermeasures of regional ecological environment.

Improvements to the fuzzy mathematics comprehensive quantitative method for evaluating fault sealing

Fuzzy mathematics is an important means to quantitatively evaluate the properties of fault sealing in petroleum reservoirs.To accurately study fault sealing,the comprehensive quantitative evaluation method of fuzzy mathematics is improved based on a previous study.First,the single-factor membership degree is determined using the dynamic clustering method,then a single-factor evaluation matrix is constructed using a continuous grading function,and finally,the probability distribution of the evaluation grade in a fuzzy evaluation matrix is analyzed.In this study,taking the F1 fault located in the northeastern Chepaizi Bulge as an example,the sealing properties of faults in different strata are quantitatively evaluated using both an improved and an un-improved comprehensive fuzzy mathematics quantitative evaluation method.Based on current oil and gas distribution,it is found that our evaluation results before and after improvement are significantly different.For faults in＂best＂and＂poorest＂intervals,our evaluation results are consistent with oil and gas distribution.However,for the faults in＂good＂or＂poor＂intervals,our evaluation is not completelyconsistent with oil and gas distribution.The improved evaluation results reflect the overall and local sealing properties of target zones and embody the nonuniformity of fault sealing,indicating the improved method is more suitable for evaluating fault sealing under complicated conditions.

Optimal operation of Internet Data Center with PV and energy storage type of UPS clusters

With the development of green data centers,a large number of Uninterruptible Power Supply(UPS)resources in Internet Data Center(IDC)are becoming idle assets owing to their low utilization rate.The revitalization of these idle UPS resources is an urgent problem that must be addressed.Based on the energy storage type of the UPS(EUPS)and using renewable sources,a solution for IDCs is proposed in this study.Subsequently,an EUPS cluster classification method based on the concept of shared mechanism niche(CSMN)was proposed to effectively solve the EUPS control problem.Accordingly,the classified EUPS aggregation unit was used to determine the optimal operation of the IDC.An IDC cost minimization optimization model was established,and the Quantum Particle Swarm Optimization(QPSO)algorithm was adopted.Finally,the economy and effectiveness of the three-tier optimization framework and model were verified through three case studies.

Clustering residential electricity load curve via community detection in network

Performing analytics on the load curve(LC)of customers is the foundation for demand response which requires a better understanding of customers'consumption pattern(CP)by analyzing the load curve.However,the performances of previous widely-used LC clustering methods are poor in two folds:larger number of clusters,huge variances within a cluster(a CP is extracted from a cluster),bringing huge difficulty to understand the electricity consumption pattern of customers.In this paper,to improve the performance of LC clustering,a clustering framework incorporated with community detection is proposed.The framework includes three parts:network construction,community detection,and CP extraction.According to the cluster validity index(CVI),the integrated approach outperforms the previous state-of-the-art method with the same amount of clusters.And the approach needs fewer clusters to achieve the same performance measured by CVI.

Space decomposition based parallelization solutions for the combined finiteediscrete element method in 2D

The combined finiteediscrete element method （FDEM） belongs to a family of methods of computationalmechanics of discontinua. The method is suitable for problems of discontinua, where particles aredeformable and can fracture or fragment. The applications of FDEM have spread over a number of disciplinesincluding rock mechanics, where problems like mining, mineral processing or rock blasting canbe solved by employing FDEM. In this work, a novel approach for the parallelization of two-dimensional（2D） FDEM aiming at clusters and desktop computers is developed. Dynamic domain decompositionbased parallelization solvers covering all aspects of FDEM have been developed. These have beenimplemented into the open source Y2D software package and have been tested on a PC cluster. Theoverall performance and scalability of the parallel code have been studied using numerical examples. Theresults obtained confirm the suitability of the parallel implementation for solving large scale problems. 2014 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. All rights reserved.

Numerical simulation of hydraulic fracturing and associated microseismicity using finite-discrete element method

Hydraulic fracturing （HF） technique has been extensively used for the exploitation of unconventional oiland gas reservoirs. HF enhances the connectivity of less permeable oil and gas-bearing rock formationsby fluid injection, which creates an interconnected fracture network and increases the hydrocarbonproduction. Meanwhile, microseismic （MS） monitoring is one of the most effective approaches to evaluatesuch stimulation process. In this paper, the combined finite-discrete element method （FDEM） isadopted to numerically simulate HF and associated MS. Several post-processing tools, includingfrequency-magnitude distribution （b-value）, fractal dimension （D-value）, and seismic events clustering,are utilized to interpret numerical results. A non-parametric clustering algorithm designed specificallyfor FDEM is used to reduce the mesh dependency and extract more realistic seismic information.Simulation results indicated that at the local scale, the HF process tends to propagate following the rockmass discontinuities; while at the reservoir scale, it tends to develop in the direction parallel to themaximum in-situ stress. 2014 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. All rights reserved.

Characteristics of earthquake clustering in and around Japan revealed by single-link cluster analysis

In this study,we apply the single-link cluster(SLC)method to analyze the characteristics of earth-quake clustering in Japan.Among the clustering algorithms,the SLC method is effpctive in characterizing earthquake clusters or isolated events at both global and local scales.The results indicate that link lengths for the whole investigated area in and around Japan follow negative exponential functi-onal or Gamma distribution functional increase.Besides,some difference is revealed among different areas,e.g.,link lengths in the area of the Japan Trench and the Kuril Trench are shorter than those in other areas;the densest spatial distribution of the SLC framework is also in the area of the Japan Trench and the Kuril Trench.The close investigations indicate that the a and 1/θvalues estimated respectively from the exponential function and Gamma distribution may relate to spatial clustering,which is supported by the results of the distribution of link lengths and the spatial distribution of the SLC framework in the investigated areas.

Studies of Atomic Structure and Physical Properties of Metal Clusters in MgO by HREM and Nano-probe Methods

Nanometer-sized metal clusters were prepared inside single crystalline MgO films by vacuum co-deposition of metals and MgO. The atomic structure was studied by high-resolution electron microscopy (HREM) and nm-area electron diffraction. The size of the clusters is ranging from 1 nm to 3 nm without those larger than 5 nm, and most of them have definite epitaxial orientations with the MgO matrix films. The character of the composite films is very much useful for the studies of various kinds of physical properties with anisotroPy. The physical properties such as electric transport, magnetic, optical absorption, sintering and catalytic ones were thus measured on the same samples analyzed by HREM by using high sensitivity apparatus with interest of clarifying the retationship between the atomic structure and physical properties

Genetic-Chicken Swarm Algorithm for Minimizing Energy in Wireless Sensor Network

Wireless Sensor Network(WSN)technology is the real-time applica-tion that is growing rapidly as the result of smart environments.Battery power is one of the most significant resources in WSN.For enhancing a power factor,the clustering techniques are used.During the forward of data in WSN,more power is consumed.In the existing system,it works with Load Balanced Cluster-ing Method(LBCM)and provides the lifespan of the network with scalability and reliability.In the existing system,it does not deal with end-to-end delay and deliv-ery of packets.For overcoming these issues in WSN,the proposed Genetic Algo-rithm based on Chicken Swarm Optimization(GA-CSO)with Load Balanced Clustering Method(LBCM)is used.Genetic Algorithm generates chromosomes in an arbitrary method then the chromosomes values are calculated using Fitness Function.Chicken Swarm Optimization(CSO)helps to solve the complex opti-mization problems.Also,it consists of chickens,hens,and rooster.It divides the chicken into clusters.Load Balanced Clustering Method(LBCM)maintains the energy during communication among the sensor nodes and also it balances the load in the gateways.The proposed GA-CSO with LBCM improves the life-span of the network.Moreover,it minimizes the energy consumption and also bal-ances the load over the network.The proposed method outperforms by using the following metrics such as energy efficiency,ratio of packet delivery,throughput of the network,lifetime of the sensor nodes.Therefore,the evaluation result shows the energy efficiency that has achieved 83.56%and the delivery ratio of the packet has reached 99.12%.Also,it has attained linear standard deviation and reduced the end-to-end delay as 97.32 ms.

Cluster-Based Stable BSM Dissemination System for Safe Autonomous Platooning

Recently,the importance of vehicle safety supporting system has been highlighted as autonomous driving and platooning has attracted the researchers.To ensure driving safety,each vehicle must broadcast a basic safety message(BSM)every 100 ms.However,stable BSM exchange is difficult because of the changing environment and limited bandwidth of vehicular wireless communication.The increasing number of vehicles on the road increases the competition to access wireless networks for BSM exchange;this increases the packet collision rate.An increased packet collision rate impairs the transmission and reception of BSM information,which can easily cause a traffic accident.We propose a solution,the vehicular safety support system(V3S),which exchanges BSMs reliably even when many vehicles are on the road.The V3S uses a clustering scheme to decrease network traffic by reducing the amount of data exchanged between a vehicle and the roadside unit(RSU).In addition,the V3S reduces the collision rate of wireless network packets by broadcasting the vehicle’s BSM in an allocated timeslot using the time division multiple access(TDMA)MAC protocol.The V3S also deals with insufficient bandwidth for dedicated short-range communications(DSRC)by changing DSRC channels according to traffic flow.In evaluating the packet error rate for stable BSM packet delivery,the V3S demonstrates an excellent packet error rate of less than 1%,compared to the 802.11p with its packet error rate of 82%.

Learning Hierarchical User Interest Models from Web Pages

We propose an algorithm for learning hierarchical user interest models according to the Web pages users have browsed. In this algorithm, the interests of a user are represented into a tree which is called a user interest tree, the content and the structure of which can change simultaneously to adapt to the changes in a user＇s interests. This expression represents a user＇s specific and general interests as a continuurn. In some sense, specific interests correspond to shortterm interests, while general interests correspond to longterm interests. So this representation more really reflects the users＇ interests. The algorithm can automatically model a us er＇s multiple interest domains, dynamically generate the in terest models and prune a user interest tree when the number of the nodes in it exceeds given value. Finally, we show the experiment results in a Chinese Web Site.

Application of a soft competition learning method in document clustering

Hard competition learning has the feature that each point modifies only one cluster centroid that wins. Correspondingly, soft competition learning has the feature that each point modifies not only the cluster centroid that wins, but also many other cluster centroids near this point. A soft competition learning method is proposed. Centroid all rank distance （CARD）, CARDx, and centroid all rank distance batch K-means （CARDBK） are three clustering algorithms that adopt the proposed soft competition learning method. Among them the extent to which one point affects a cluster centroid depends on the distances from this point to the other nearer cluster centroids, rather than just the rank number of the distance from this point to this cluster centroid among the distances from this point to all cluster centroids. In addition, the validation experiments are carried out in order to compare the three soft competition learning algorithms CARD, CARDx, and CARDBK with several hard competition learning algorithms as well as neural gas （NG） algorithm on five data sets from different sources. Judging from the values of five performance indexes in the clustering results, this kind of soft competition learning method has better clustering effect and efficiency, and has linear scalability.